Report of the Health Care Task Force


The Pennsylvania State University


April 18, 2014
















Table of Contents


Executive Summary…..……………………………….………………………………………2


 Report of the Academic Research Subcommittee…………...…….…………....16

 Report of the Benchmarking Subcommittee………………………………………48

 Report of the Health Spending, Utilization, Value

and Vendor Analysis Subcommittee……………………………………...…54

Report of the Communications Subcommittee…….……………………...…..…89

Report of the Health Care Trends Subcommittee………………………....……96


Appendix A……………………………………………………………………………………117

Appendix B……………………………………………………………………………………118

Appendix C……………………………………….……………………………………………119

Appendix D......................................................................................................................120

Executive Summary

The formation of the Health Care Task Force was, in part, motivated by the recent concerns raised by faculty and staff regarding the “Take Care of Your Health” initiative that the University introduced for the 2014 benefits year.  The membership of the Task Force, which is composed of six faculty, two staff, and five administrators, was selected to bring together knowledgeable individuals with backgrounds and expertise in health care policies and programs, insurance, finance and related aspects, with the goal of helping to inform decisions about the planning, programs, partnerships, financing and communications associated with the health care of Penn State employees and their dependents. 


While the charge of the Task Force is broad, specific items within the purview of the Task Force include the benchmarking of health care benefits programs, the assessment of alternative approaches to reduce the rate of increase in the University’s health care costs and to improve the health status of employees and their dependents, and the investigation of improvements to communication, consultation and transparency for current and future health care programs.  And, although the genesis of the Task Force was the most recent controversy over the wellness initiative, the Penn State community clearly needs to take a serious look at the value received for its health care expenditures.  Between 2005 and 2013, the costs to Penn State of health care rose by 72.2% while the centrally allocated raise pool increased only by 24.5%.


The Task Force Report begins with an analysis of the academic literature examining the efficacy of three programs of particular relevance to the Penn State community:  High Deductible Health Care Plans, Workplace Wellness Programs, and Tiered Benefit Designs.  While all of the studies have methodological shortcomings, several broad themes emerge.  First, the implementation of high-deductible health care plans appears to reduce substantially health care expenditures and there is some reassuring evidence that the use of preventive care need not be compromised when such care is excluded from the plan deductible.  Second, workplace wellness programs appear to have, at best, a small but statistically significant short run impact on some employee health behaviors, most notably smoking cessation and weight loss, but the longer run effects are still unclear.  And, to the extent that these programs generate cost savings, they seem to arise from intensive disease management rather than lifestyle management programs.  Third, the record suggests that tiered benefits effectively steer consumers away from non-preferred drugs when implemented in pharmacy benefit plans, but that the results are much more sparse in cases of hospital or physician care.


The Task Force also implemented a benchmarking survey of peer academic institutions.  The resulting survey includes data on Penn State and responses from eight other Big Ten universities, plus Temple University and the University of Pittsburgh.   The survey focused on the two types of plans offered by Penn State:  A Preferred Provider Organization (PPO) and a Qualified High Deductible Health Plan (QHDHP).  Of those universities in the survey, a majority offered more than two plans.  Penn State is among the five in the survey that offer a QHDHP and, of those five, three universities require salary-based premium contributions by employees while two have flat dollar charges.   In terms of the premium contributions for the PPO plans, five of the eleven use some form of salary-based indexing and six use flat dollar charges that are independent of salary level.  Penn State is the only institution that uses a flat percentage of salary in the indexing of consumer premiums.


Of the eleven universities in the survey, Penn State is alone in assessing surcharges for spouses/partners who have access to coverage through their own employer, and for those using tobacco.  Eight of the eleven universities report some type of wellness program, and six of those offering wellness programs report that incentives are used to encourage participation.  All but one university in the survey reports that they provide health care for both pre-Medicare and Medicare-eligible retirees.    In addition to Penn State, one other institution in the survey provides Medicare-eligible retirees with a Medicare Advantage Plan that includes prescription drug coverage.   Seven others offer a Medicare Supplement Plan, of which five include a prescription drug plan and two do not.


The Task Force also examined a number of policies that could correct market failures in health and medical care, with the specific goal of increasing the value that the Penn State community receives for its health care expenditures.  While the implementation of any of these policies would require further analysis and consultation with the affected parties, this Report provides a starting point.


Given that accurate and timely information is a prerequisite for making informed decisions, the University may wish to improve its data and analytic infrastructure through the establishment of a comprehensive data warehouse to house claims data.  Such a warehouse, which would adhere to the federal government’s privacy and confidentiality requirements for handling personal health information under the Health Insurance Portability and Accountability Act, would permit analyses, in a timely and systematic way, to support and guide decisions regarding contracted vendors, prices paid to providers, the quality of care received by beneficiaries, and other important questions. 


The Task Force also considered how Penn State could leverage its purchasing power by assessing the efficiency of contracted vendors and consideration of alternatives for the services that they provide.  First, Medicare-eligible retiree benefits are currently offered a group insurance plan from Highmark that is fully insured.   Many employers have moved instead to a Medicare Advantage Marketplace in which choices are made from a menu of plans with different prices and characteristics, and preliminary evidence suggests that this approach could permit our retirees to purchase better plans at a lower cost.  Second, the University could harness the Big Ten’s Committee on Institutional Cooperation to work together with peer institutions on mutually beneficial projects and to share information.   Third, an effort could be made to identify health care providers that serve significant portions of the Penn State population and engage in discussions about preferred arrangements.  This could result in more favorable pricing, jointly developed care management and quality programs, or centers of excellence for more complex treatments and procedures.   Fourth, the University could assess the value it is receiving from its current pharmacy benefits manager by “carving out” this service from the current contract with the third-party administrator (TPA) and putting it out for bid separately.  Fifth, Penn State could assess the market competitiveness of the current administrative services fee paid to the current TPA by developing an RFP to competitively bid for TPA services.


The Task Force also considered possible changes in the ways that Penn State and its TPA pay for health care services.  First, the University may wish to consider the use, where appropriate, of reference-based pricing that is a strategy designed to generate convergence in the market to a fair price for clinical services such as diagnostic tests, durable medical equipment, or laboratory tests.  Second, the University could work with payers to implement bundled pricing in some settings, which is a package price for a bundle of services or an episode of care.  Such pricing eliminates the incentives inherent in a fee-for-service environment to bill for unnecessary treatment.  Third, the University could follow the lead of Medicare and implement a policy of non-payment for medical errors.


Also examined were options that would increase access to urgent and primary care services, such as onsite health clinics and expanded primary care at campus locations.  A related concern is the value of different care management models for beneficiaries and, in this vein, the University may wish to evaluate the current Highmark disease management program against alternatives such as complex case management or gaps in care programs, second opinion services, or traditional disease management programs.


As health care benefit design has increasingly placed the employees in the position of making decisions about treatments and also bearing an increasing share of the associated costs, the Task Force explored several options designed to provide better information to consumers and tools for beneficiaries to help them navigate increasingly complex health systems.  One would be Choosing Wisely, an initiative of the American Board of Internal Medicine Foundation that is designed to use professionalism and accountability on the part of physicians and their specialty societies to address the problem of costly utilization of inappropriate health care services.  Another would be the provision of a comprehensive consumer transparency tool to provide more and better information to beneficiaries on provider quality and price.  These tools, which are currently provided by several vendors, may be web-based and can provide information on where the patient stands with respect to deductibles and savings accounts as well as provide concrete information on pricing by specific providers to help the patient make informed decisions about their medical care.


There are also several possibilities for continued changes in benefit design that may provide greater cost efficiencies for employees.  One obvious strategy would be to consider increasing enrollment in the high deductible health plan option that, according to the review of the academic literature, has been shown to be an effective tool for reducing health care expenditures.  Another approach would be to include incentives or penalties for engaging in wellness programs or engaging in risky behavior. While this is permitted by provisions in the Patient Protection and Affordable Care Act, any benefits of this approach must be balanced against the costs to employees, both in terms of their pocketbooks and their privacy.  The mixed success of such programs that is noted in the academic review section, in conjunction with the negative response to the recent “Take Care of Your Health” initiative at Penn State, suggests that this option would need careful thought and consultation if implementation were being considered. A third strategy might be to invest in intensive disease management which could offer potential savings while helping University employees and their families improve their health status.


The Task Force also considered several policies involving University operational decisions.  One of increasing importance is ensuring access to health care providers for beneficiaries in western Pennsylvania given the fierce competition between UPMC and Highmark Blue Shield with its affiliated health care delivery system (West Penn Allegheny Health System).   Another would be to eliminate payments for stop-loss insurance, and a third would be to evaluate the need for the continuation of internal university subsidies to the Hershey Medical Center.


The Report contains a summary of the Task Force deliberations on the need for effective communication and consultation with the Penn State community when changes in employee benefits are being considered.   An essential feature of effective communication policy is improving the transparency of the administration’s decision-making process and providing faculty and staff with opportunities to shape and test programs before they are implemented as policy.  It is viewed as important to build a culture of open, honest and frequent communication, and to recognize individuals’ sensitivity while respecting personal privacy.  Communications from the administration should be crafted to build credibility and to demonstrate good will.


The Task Force also examined the health care trends in the private and public sectors.  While the rate of increase in aggregate health care expenditures has flattened a bit recently, there are many factors that will continue to push costs upward.  These include an increase in chronic conditions, continued waste and abuse, the increased prevalence of health insurance with the associated incentives to over-consume low-value care, and the increased use of expensive pharmaceuticals.   The health care system is likely to face many challenges, most notable those associated with chronic conditions, a likely under-supply of primary care physicians, the slow adoption of evidence-based medicine, the incentive problems associated with fee-for-service payments, the shift from inpatient to outpatient care, and the growing focus by payers and patients alike on value.


The health care landscape is also in flux as the Patient Protection and Affordable Care Act (PPACA) is implemented in phases.  The Act provides a strong focus on prevention, including a substantial list of clinical preventive services that must be covered, such as screenings, immunizations, and counseling to prevent health problems from developing or to detect diseases at early stages when they are more amenable to treatment.   The PPACA also focuses on innovative delivery reform, the goal being to increase quality and reduce costs by combining new approaches to organizing and delivering care with payments designed to support these changes.  Three particular programs appear to have efficacy in achieving these goals:  management of chronic disease, improved care coordination, and the bundling of payments.


In terms of challenges on the horizon, the Task Force has identified several trends that bear continued scrutiny.  One is the new insurance exchanges, and how they develop over time (or do not) will have a strong influence not only on the individual insurance market, but on the group market as well.  Another is the increasing use of electronic medical records, the development of which has already been very expensive and disruptive, yet the benefits have yet to be fully realized.  The evolution of the understanding of the human genome promises to revolutionize health care in ways that are likely unimaginable at this time.  New technologies for interacting with patients at lower cost than the standard model continue to be developed, and of course the reality of an aging population casts a deep shadow over any discussion of future health care expenditures.


Finally, the Task Force discussed, at some length, the merits of including in the Report a list of "action items" for the benefit of those who will be tasked with making the decisions regarding future health care benefits.  Given that all of the options going forward involve a high level of uncertainty, and some more than others, in terms of their applicability to the situation faced by Penn State, the Task Force decided that such an approach would be counterproductive.  Given the limited time provided to craft this Report, and considering the complex and often subtle tradeoffs that will be faced in any formal evaluation of the options, the Task Force can do no more than to set the stage by providing a framework for the detailed analyses and discussion which must inevitably follow.


I.  Origins of the Task Force


"After roughly two hours of discussion and debate during a special meeting, the Faculty Senate voted 93-15 to approve an amended resolution supporting the University's suspension of the penalties for noncompliance with the wellness initiative in the Take Care of Your Health Plan.


The resolution further requested the suspension of the additional surcharge for coverage for spouses and same-sex domestic partners, citing the financial hardship on faculty and staff with lower salaries; the use of policy experts, both within the University Senate and throughout the Penn State community, in an effort to fairly evaluate the alternatives for controlling health care expenses; and engagement of the faculty and staff in a dialog, soliciting their views on how benefits changes would impact their health, finances and productivity. The Senate resolution also spelled out the body's expectations for how the announced health care task force should be populated.


The resolution calls for an update on reforms to the Take Care of Your Health initiative to be brought before the full University Senate by its last regular meeting, which is scheduled for Tuesday April 29, 2014."[1]

As a result of the September 24, 2013 Faculty Senate resolution, the Health Care Task Force was created, and formally charged on November 18, 2013. The Task Force, the membership of which is provided in Attachment B, is composed of 13 people, six of whom are faculty, two are staff, and five are administrators. Administrative support was provided by the Office of the President in the person of Monica Nachman, whose exemplary assistance has been of immeasurable value to the Task Force in the performance of its duties.


The charge letter to the Task Force is contained in Appendix A, the relevant part of which is reproduced below.

"Matters within the purview of the Health Care Task Force include, but are not restricted to, the following:


·         The benchmarking of health care benefits programs, including those with wellness components, at peer public and private universities and major corporations, and the identification of key trends in health care benefits programs and planning.

·         Data-driven and science-based assessments of alternative approaches to reduce the rate of increase in Penn State’s health care costs. These may include the benchmarking of peer university cost structures to assess the relative magnitudes of employer and employee contributions, the efficacy of earnings-based contributions, the advantages of promoting internal or favored external facilities and services, and possible ways to increase the choices in health care coverage and prices among covered employees and their dependents.


·         The exploration of the advantages and disadvantages of various kinds of programs, such employee wellness programs, which could result in an improved health status for Penn State employees and their dependents.


·         Investigate improvements to communication, understanding, consultation, transparency, and employee participation for current and future health care programming and health care benefits."


 II.  Procedure


After the November 18 meeting when the charge was presented, the Task Force next met on December 9, at which time several subcommittees were formed in order to process the anticipated workload.  A list of these subcommittees, and their membership, follows.


·         Health Spending, Utilization, Value and Vendor Analysis:  Keith Crocker, Joe Doncsecz, Robin Haas, Dennis Scanlon (Chair), Rachel Smith, Robin Wittenstein.

·         Benchmarking:  Martha Aynardi, Jo Anne Carrick, Robin Haas, Rachel Smith (Chair), Jeremy Warner.

·         General Trends:  Carol Weisman, Robin Wittenstein (Chair).

·         Communications:  John Harwood, Susan Johnson, S. Shyam Sundar (Chair).

·         Academic Research:  Keith Crocker (Chair), Kyoungrae Jung, John Moran, Dennis Scanlon, Dennis Shea.[2]

Beginning on January 6, 2014 the Task Force met biweekly.  The subcommittees set their own meeting schedules, and reported back to the full Task Force on a regular basis.  There were also regularly scheduled presentations at the Task Force meetings to inform the members during their deliberations.

·         January 6:  Highmark Presentation (William Straw, Kelly Lieblein, Kurt Keul).

·         January 20:  Towers-Watson Presentation[3] (Allan Khoury, Geri Recht).

·         February 3:  Take Care of your Health Initiative Presentation (Linda Tobin).

·         March 3:  Hershey Medical Center Presentation (Dr. William Bird).

·         March 17:  Castlight Presentation[4] (Jason Frushon, Steve Jacobson, Brian Jones).

The deadline faced by the Task Force was to produce a report that could be included in the agenda of the April 29 Faculty Senate meeting.

III.  Background

From the outset, the goal of the Task Force has been to examine the health care alternatives facing the Penn State community and, in the process, to consider the costs to employees of various policies, as well as the associated benefits to the University.  The purpose of the Task Force is not to make specific policy recommendations but, rather, to provide a rigorous evaluation of the alternatives that will be of use to decision makers.  This includes both (i) an assessment of the underlying tradeoffs inherent in each of the policies, and (ii) the types of data that would need to be gathered beforehand in order to make an informed decision.  In this fashion, our goal is to provide a dispassionate analysis of the potential paths forward.


This report is crafted in the shadow of two imposing facts of the Penn State health care landscape.  The first is retrospective--the historical pattern of spending by the University on health care--and is of particular concern since Penn State is self-insured and health care expenses come out of the same budgetary "pot" as everything else, including wages, travel and research budgets.  The second is prospective--the seismic changes in the provision of health care caused by the implementation of the Patient Protection and Affordable Care Act--and, while the full scope of the changes are still in flux, there are some clear and proximate impacts.

A. Penn State Health Care Expenditures.


The following table contains data provided by the Penn State Controller's Office on the University's health care expenditures for the past decade, the rate of increase of which is sobering.











Paid $

(% change)













(% of Total Claims)





























Paid $

(% change)













(% of Total Claims)



















The "Total Claims Paid" consists of the health care and prescription drug claims paid by Penn State for all active and retired employees (both Medicare-eligible and non-eligible), as well as administrative fees.  This represents the total cost to the Penn State community of health care, not including (after 2011) the employee deductibles and coinsurance.[5]  The "Premium Contributions" are the total of all premiums paid by Penn State employees and are deducted from employees' paychecks or paid directly by retirees.  The "Net PSU Cost" is the total cost to Penn State of paying health care claims. 


The increase in Total Claims Paid between 2005 and 2013 was $87,764,059.  To put this number in context, one percent of the General Funds salary budget base in 2013-2014 (excluding the Dickinson School of Law, the College of Medicine, and Penn College) is $9.7 million, and a one percent increase in tuition would generate about $11.5 million.  And, while the Total Claims Paid increased by 72.2% between 2005 and 2013, during the same time period the centrally allocated raise pool at Penn State increased by a compounded rate of only 24.5%.


In addition to the cost of paying claims, there are several other substantial health care expenses incurred by the University.  First is the payment to Hershey Medical Group to ramp up their new clinics in State College.  This payment, which totals $15,635,242 to date, has risen from $646,528 at its inception in 2010 to $6,076,356 in 2013.  Second is the payment by the University to the TIAA Retirement Health Savings Plan for those beginning employment at Penn State after December 31, 2009, which replaces the earlier retiree medical plan.  These contributions total $10,675,397 since 2010, although the total cost is reduced by the $1,008,195 that Penn State has received as a result of those leaving University employment and thereby forfeiting their balances.  The final substantial cost to the University is the ASO reimbursement to the Hershey Medical Center (HMC), which has amounted to $10,985,587 since 2008.


While Penn State is self insured, and so it ultimately pays all of the claims (net of co-premiums, deductibles and coinsurance), the University does employ a third-party administrator (TPA) to handle and process the claims.  Prior to 2008, HealthAmerica was the TPA for the major portion of University covered lives, and after 2008 the job was passed to Highmark, at which time they became the exclusive provider of health care insurance for faculty and staff.  2008 was a transition year in which both companies handled part of the case load.  The claims payments numbers above include payments to the TPAs.[6]  


As indicated in the table, prior to 2009 the rate of increase in health care costs averaged around 7% per year.  This rate increased to 13% in both 2009 and 2010, but then fell to an average of around 3% from 2011-2013, although projections (which are admittedly speculative) suggest that the rate may increase to 9% in 2014.[7]


The reasons behind the reduction in the rate of increase in Penn State's health care expenditure from double-digits prior to 2011 to a rate in the low single digits after 2011 are unclear.  One possibility is that the recession, which depressed health care costs for all employers, may be the culprit.[8]  If so, one might expect that the previous higher rates of increase may re-emerge as the economy improves.  Another possibility is the changes in the Penn State plan design, implemented in 2011, that introduced in-network deductibles and coinsurance may have resulted in a reduction in health care claims as consumers (who now incur a larger share of the cost of health care) reduced their utilization of health care services.  If this were to be the case, the important question is whether this is a one-shot reduction in health care costs (after which the previous rate of increase reasserts itself) or an actual longer-term reduction in the rate of cost inflation (sometimes referred to as "bending the curve").   A third possibility is the increased availability of the HMC clinics, by offering both greater access and increased competition for the community based physician groups, may have lead to positive changes in utilization and cost patterns. The bottom line is that we do not know the reasons for the recent reduction in rate of increase of Penn State's health care expenditures.


As a result, it is important that the Penn State community examine carefully the health care policy options going forward.  If our current experience is simply a hiatus in a longer term, and unsustainable, trend of increasing health care costs, then the need to carefully consider the options is obvious.  But even if we were fortunate enough to have successfully "bent the curve" in health care costs, it is important to know which policies are the most effective in delivering health care value for the dollar.  Policies resulting in inefficient expenditures on health care can be replaced by those offering the same value for less money or, alternatively, with ones that provide higher value for the same health care expenditure.[9]

B.  The Impact of the Affordable Care Act.


The 2010 Patient Protection and Affordable Care Act (PPACA) will impact the Penn State health care landscape in a myriad of ways.  The cost and structure of the health care market will change in response to the Act’s sections that encourage innovation and change in how health care services are organized, delivered and paid for.  These include the development of medical homes for patients with multiple chronic conditions, bundled payments, accountable care organizations, increased primary care funding, enhanced prevention activities, and the development of the Center for Medicare and Medicaid Innovation.  As noted below in the Report of the Health Trends Subcommittee, the increased emphasis on health care prevention and the opportunities for innovation in care delivery and payment will be particularly important in the growth of health care expenditures at Penn State.  There are also several other challenges that likely will arise from the implementation of the PPACA.


One is the rise of the insurance exchanges in the individual market.  While the dominant model for the delivery of health care to employees has been employer-sponsored health plans, which are group plans composed of the employer’s work force that either are underwritten by a third-party insurer (in the case of small employers) or are self insured (which is the course often chosen by large employers such as Penn State), this may be changing.  If the exchanges in the individual market turn out to be an effective way of providing insurance coverage, some employers may opt to change from providing group coverage to simply providing a fixed payment to employees who then purchase their own coverage on the insurance exchanges.  The result is that it becomes the employee that assumes the risks of escalating health care costs and inappropriate utilization patterns.  And, if this becomes a common mechanism to reduce employers’ exposure to escalating health care costs, there will be budgetary pressures to conform.[10]


A second challenge involves the requirement that employers with more than fifty full-time employees offer affordable coverage to all full-time employees.  Failure to comply with the mandate can result in substantial penalties levied on the employer.[11]   The definition of a full-time employee under the PPACA is defined as one working an average of 30 hours per week, or at least 130 hours in a month.  Given the large number of employees at Penn State that currently work an average of less than 40 hours per week, some in several different areas in the University, ensuring compliance with this requirement promises to be a difficult task.


A third challenge involves the newly mandated fees under the PPACA.  Given Penn State’s status as a large, self-insured, employer, it is required to pay the Transitional Reinsurance Fee that is $63 per member per year in 2014, and is scheduled to fall to $44 per member per year in 2015.  The University is also subject to the mandated Patient Centered Outcomes Research Institute (PCORI) fee that for 2013 was $2 per member per year.[12]  Of more substantial concern, however, is the 40% excise tax levied on certain high-value health insurance plans in 2018, which is sometimes referred to as the “Cadillac Tax”.[13]  This tax applies to both fully insured and self-funded employer plans and would be paid by the University, so there is clearly a strong incentive when considering plan design to avoid this liability.


 IV. Outline of the Report


The report is organized by subcommittee, presenting the findings of each in a separate chapter with supporting documentation contained in appendices.  The first section contains the report of the Academic Research Subcommittee that contains reviews of the academic literature on high deductible heath care plans, workplace wellness programs, and tiered pricing in benefit plans.  While there exists a voluminous academic literature on a host of other potential policies, these three were selected because they are directly relevant to the Penn State experience.


The second section contains the report of the Benchmarking Subcommittee which, with input from the full Task Force, from benefits experts in the Penn State Office of Human Resources, and from experts at Towers Watson (a consulting firm under contract to Penn State), used the Qualtrics survey software to gather information from CIC institutions, plus the University of Pittsburgh and Temple University, on the structure of their health care programs.  The data obtained from this survey are contained in Appendix C of this report.


The third chapter contains the report of the Health Spending, Utilization, Value and Vendor Analysis Subcommittee that contains assessments of alternative options or enhancements for maximizing the value of the University's investments in health care benefits.   Rather than making specific recommendations, this chapter identifies options and contains a discussion of their potential benefits and risks based upon available evidence.  In cases where more evidence would be required, suggestions are made regarding the type of data that would be useful as well as the nature of the analyses that should be performed in order to make an informed decision.


The fourth chapter contains the report of the Communications Subcommittee that examines potential improvements to communication, understanding, consultation, transparency, and employee participation for current health care programming and health care benefits.


The fifth chapter contains the report of the Health Trends Subcommittee, which examines general trends in health care expenditure, the factors impacting the increases in health care costs, the challenges facing health care systems, and key components of and potential impacts of the Patient Protection and Affordable Care Act (PPACA) of 2010.


A final section contains concluding remarks, including a discussion of the recent controversy over health care reforms at Penn State.

Report of the Academic Research Subcommittee


 I. Introduction


The recent opposition to the “Take Care of Your Health” initiative by Penn State faculty and staff was focused on privacy and efficacy concerns.   With regard to the latter, the University administration argued that three studies, two of which were authored by employees of Highmark, the program vendor, provided evidence that the wellness initiative would be effective in reducing health care costs.  As it turns out, there is a broad academic literature examining the efficacy of many health care programs, and these studies have the distinct advantage of being written by someone without a product to sell, and they have been subject to peer review and scrutiny.  The purpose of this chapter is to review the academic literature on three health care programs of particular relevance to the Penn State community:  high-deductible health care plans, workplace wellness programs, and tiered benefit designs.  This review should be helpful as the University considers future health care initiatives.


While all of the academic studies have methodological shortcomings, and some more than others, there are several broad themes that emerge.  At the risk of oversimplification—the caveats are carefully noted in the reviews below—this is what we know.  First, the implementation of high-deductible health care plans appears to reduce substantially health care expenditures and although the verdict isn’t in yet regarding their impact on health outcomes, there is some reassuring evidence that the use of preventive care need not be compromised when such care is excluded from the plan deductible.


Second, workplace wellness programs appear to have, at best, a small but statistically significant short-run impact on some employee health behaviors, most notably smoking cessation and weight loss, but the longer run effects are still unclear.  And, to the extent that these programs generate cost savings, they seem to arise primarily from intensive disease management programs.  Lifestyle management programs appear to be, at best, a breakeven proposition, and in many cases may cost more than they save in health care costs.


Third, the efficacy of tiered benefits designed to channel consumers to preferred providers is mixed.  The record suggests that, in the case of pharmacy benefits, tiered benefits effectively steer consumers away from non-preferred drugs, but the evidence is much more sparse in the case of hospital or physician care.  Moreover, some studies have indicated that a large copayment difference would be required to effectively influence consumers’ choices of providers.


While there is a substantial academic literature examining other important health care policies, the three that we have chosen to examine carefully in this section are particularly relevant to the Penn State community.  The high-deductible health care plan is already offered as an option (the “PPO Savings” plan), and of course the wellness initiative was implemented and then largely rescinded in the summer of 2013.  The tiered benefits design, while not currently a part of the Penn State plan design, has been discussed as a possibility in the future, and so merits our attention as well.


II.  High-Deductible Health Plans in an Employer Group Setting.[14]


A.  Background


Although there is no official definition of a “high-deductible health plan” (HDHP), such plans generally combine two elements that distinguish them from more traditional health insurance policies: the first, as the name suggests, is a deductible that is high relative to recent norms, and the second is a Health Savings Account (HSA), into which both employers and employees can contribute funds on a pre-tax basis that can be withdrawn, tax free, to pay for qualified medical expenses (Bundorf, 2012). These contributions are helpful in providing liquid funds that can be used to meet, or partially meet, the plan’s deductible, and the employer’s contributions also serve to reduce the financial burden workers bear prior to satisfying the deductible. Unspent account balances are transferable across employers, grow tax free, and function much like a traditional IRA in that balances remaining at age 65 can be used for non-medical purposes. This provides consumers with an incentive to use the funds judiciously since any monies spent on health care will not be available for other purposes after retirement. Internal Revenue Service regulations require that high-deductible plans meet certain requirements if an HSA is to be offered. For example, in 2014, a plan must impose a minimum deductible of $1250 for single coverage and $2500 for family coverage and must cap total out-of-pocket expenditures at no more than $6350 for an individual plan and no more than $12,700 for a family plan (


A key concern with HDHPs is that the imposition of a higher deductible will lead enrollees to indiscriminately reduce their utilization of necessary medical care, especially preventive care, which requires that individuals incur costs immediately in exchange for benefits that may only be realized, if at all, far in the future. In response to these concerns, most employers offering HDHPs opted to exclude preventive care from the deductible and are now required to do so by the Affordable Care Act (Bundorf, 2012).


HDHPs have experienced substantial growth in recent years, with enrollment tripling between 2006 and 2012 (Kaiser Family Foundation, 2012). As of 2012, 34 percent of workers with employment-based coverage were enrolled in a plan with an annual deductible exceeding $1000 and 14 percent were enrolled in a plan with a deductible of $2000 or more (Kaiser Family Foundation, 2012). Continued growth is expected, both in response to increasing health care costs and as a means of averting the tax on “Cadillac” health plans scheduled to take effect in 2017 under the Affordable Care Act (Buntin et al., 2011).


B.  Health Insurance: Some Basic Concepts


Before proceeding, it will be useful to set out some terminology. In this document, the term deductible refers to the annual amount an insured individual or family (henceforth, the “insured”) must pay out of pocket before the policy begins to pay claims. The term coinsurance refers to the fraction of each dollar of expenditures above the deductible that the insured party is responsible for. For example, a coinsurance rate of 20 percent would mean that after meeting the annual deductible, the insured would be responsible for 20 percent of each dollar of expenditure incurred, up to the out-of-pocket maximum, which is the total amount of out-of-pocket expenditures, not including the deductible, that the insured would have to pay in a given year before the policy began picking up the full cost of additional medical care.


To make these terms more concrete, consider a hypothetical health insurance policy with a $1000 deductible, 20 percent coinsurance, and an out-of-pocket maximum of $4000.[15]  Under the terms of this policy, an insured with $1000 or less of annual medical expenditures would receive no reimbursement from the policy and, consequently, no protection from financial costs of this magnitude. If the same insured were to have $3000 in annual expenses, they would be responsible for paying $1400 out of pocket ($1000 + 0.20×($3000 – $1000)), with the insurance company paying the remaining $1600, thereby transforming a $3000 cost into a $1400 cost, and providing some protection against medical expenses in this range. Under the hypothesized policy, the insured would have to accrue $21,000 in total expenditures, corresponding to $5000 in out-of-pocket expenses ($1000 + 0.20×$20,000), before the insurance company would pay 100 percent of any additional costs arising during the year. Thus, health insurance policies with relatively high deductibles offer little protection against the risk of incurring small to moderate expenses, but a great deal of protection against catastrophic events. For example, the insured’s financial exposure arising from a kidney transplant costing $300,000 would be capped at $5000 in this example.


C.  Rationale for High-Deductible Plans


High-deductible health plans represent one of several approaches for curtailing the use of medical care that is not worth its cost. Typically presented as a strategy for cost containment generally, the fundamental economic motivation behind deductibles and other forms of cost sharing is to strike a balance between the efficiency gains from providing protection against the financial risks associated with illness and the efficiency losses associated with the overconsumption of (potentially ineffective) medical care. From an economic standpoint, the key features of a health insurance policy are: (1) the amount of financial protection afforded by the policy; and (2) the implicit prices for additional units of medical care created by the cost-sharing provisions. Notice that in the example discussed earlier, the insured would pay full price for all health care below the deductible, 20 cents on the dollar for health care above the deductible (but below the out-of-pocket maximum), and an effective price of zero once they had accrued out-of-pocket expenditures above the maximum. Thus, prior to meeting the deductible, insureds have the usual incentives to equate costs and benefits when deciding how much care to purchase. Beyond that, the 80 percent subsidy created by the policy’s coinsurance provision will induce insureds to over utilize health care in the sense that they will purchase care whose incremental value is below the incremental costs required to produce it. This distortion is heightened once the out-of-pocket maximum is reached and insureds receive care for free.


Overconsumption of health care is of more than theoretical interest. Like all care, care that is not worth its cost must be paid for in the form of higher premiums, an expense that is borne by all members of the plan. Thus, regardless of the way in which health insurance is delivered - be it through a commercial insurance company, a self-insured employer, or an informal arrangement among individuals - there will always be a rationale for providing incentives to economize on the use of care that would not pass a cost-benefit test.


D.  Effects of Cost Sharing


The emergent and non-discretionary nature of some categories of health care notwithstanding, there is credible evidence that by lowering the net price of medical services, health insurance can result in large increases in the amount of care consumed, often with little discernable benefit to the recipient in terms of improved health outcomes. A well-known experiment (often referred to in the literature as the “RAND Health Insurance Experiment”) conducted in the 1970s, in which families were randomly assigned to health plans with levels of coinsurance ranging from zero to 95 percent, found a roughly 30 percent difference in expenditures between those paying (nearly) full price and those in the free care plan, with virtually no difference observed in health outcomes across the two groups[16] (Manning et al., 1987). It is important to recognize that these findings do not imply that medicine is of little value in general, but rather that the additional care people consume in response to a reduction in price - the price-sensitive component of medical care - may produce only modest benefits.


More recently, in 2008, a randomized experiment was conducted in Oregon in which low-income adults were randomly selected from a waiting list for the state’s Medicaid program and offered enrollment. A control group comprised of individuals on the list who were not offered enrollment was tracked simultaneously for comparison purposes. Oregon’s Medicaid program provides comprehensive coverage with no cost sharing. Two years after random assignment, researchers found that enrollment in the Medicaid program increased medical spending by 35 percent relative to the control group, however no significant differences were found in three biomarkers thought to be modifiable over the relevant time span: blood pressure, cholesterol, and glycated hemoglobin levels (Baicker et al., 2013). These findings have been the subject of some controversy due to the length of the follow-up period and the relatively small number of participants with hypertension, high cholesterol, or diabetes at baseline.


Owing to its experimental design and applicability to an employed population, the RAND Health Insurance Experiment probably provides the cleanest evidence overall regarding the impact of cost sharing on medical expenditures and clinical outcomes among employed working-age adults and their dependents. This is especially true of comparisons made relative to the 95-percent coinsurance plan which, to a large degree, captures the implicit price of medical care below the deductible in today’s high-deductible health plans.  Of course, the estimates from the RAND experiment apply to a period when medicine was less efficacious than it is today and, as a result, some caution is warranted in extrapolating the results of the RAND experiment to current day circumstances, especially with respect to clinical outcomes.


E.  Methodological Limitations of HDHP Studies


In the next section, we review more recent research that pertains to high-deductible health plans directly, but note that there has not been a randomized experiment conducted for HDHPs analogous to the RAND Health Insurance Experiment or the Oregon Medicaid Experiment. All studies of which we are aware are based on non-experimental data, often from convenience samples drawn from specific employers or insurers, and are subject to a variety of caveats, which we now describe.



1.  Selection Bias


Selection bias arises when the decision to participate in the treatment arm of a study is made by participants on the basis of an unobserved factor that influences the outcome of interest. When estimating the effects of HDHPs in an employer group setting, the treatment group is comprised of workers enrolled in a HDHP and the control group is comprised of workers enrolled in a traditional plan, such as a preferred provider organization (PPO) or a health maintenance organization (HMO). Random assignment, as was used in the RAND and Oregon experiments, eliminates this problem by assigning subjects to the treatment and control groups in a manner that is independent of any unobserved factors that might influence the outcome of interest, and which would otherwise be disproportionately represented in either the treatment or control group.


There are two primary sources of selection bias that must be considered when studying the effects of HDHPs. The first arises when HDHPs are offered alongside other types of plans, creating the possibility that HDHPs disproportionately attract healthier, wealthier or more sophisticated enrollees. In her review of the literature, Bundorf (2012) finds some evidence that HDHPs benefit from favorable selection along each of these margins. If true, this is likely to result in an overestimate of the impact of HDHPs on expenditures. Many researchers sidestep this issue by focusing on employers who adopt a HDHP as a full-replacement product.


The second source of selection bias occurs at the employer level and cannot be addressed by restricting attention to employers offering a HDHP as the sole insurance option. This form of bias exists when the decision to offer a HDHP is prompted by the employer’s health care cost growth exceeding that of a reference group, e.g. employers in the same region or industry. In such cases, the change in health care costs among employers who stick with a traditional plan (the control group) does not provide a good approximation to what cost growth would have been among HDHP adopters in the absence of a HDHP plan, resulting in an underestimate of the impact of HDHPs on expenditures.


2.  Announcement Effects


A different type of bias arises when the decision to offer a HDHP is announced in advance of its adoption. This would occur, for example, if the addition of a HDHP were announced in the summer or fall prior to its implementation in January. In such cases, a subset of insureds might accelerate elective procedures or stock up on medications during the current year, which would lead researchers to overestimate the effect of the HDHP on spending if its impact were assessed using one pre- and one post-implementation year, or if the years bracketing implementation were factored into a longer evaluation. To the best of our knowledge, the only researchers who have attempted to deal with this problem are Buntin et al. (2011), who demonstrated that restricting attention to expenditures incurred during the second and third calendar quarters of the pre- and post-implementation years did not have a large effect on their estimates. This suggests that anticipatory behavior may not be an important confounder in studies of HDHPs, but it would be useful to confirm this finding using a larger data set and/or an alternate sensitivity check.


3.  Generalizability


Until recently, the majority of research studies on HDHPs were based on an analysis of outcomes for insured workers from a single employer who switched from a conventional health plan to a HDHP, with covered workers from a comparable employer who did not switch plans serving as the control group. The limited generalizability of these single-employer studies led to more recent research based on data from multiple employers, with the largest such study following approximately 700 employers over a three-year period (LoSasso, Shah and Frogner, 2010).


F.  Effects of High-Deductible Health Plans


In 2012, M. Kate Bundorf, a health economist at Stanford University, conducted a comprehensive review of the academic literature on HDHPs for a report commissioned by the Robert Wood Johnson Foundation (Bundorf, 2012). The evidence summarized below draws heavily from her review.


Before proceeding, we should note that, although a methodological critique of each of the referenced studies is beyond the scope of this document, all suffer from one or more of the limitations described in the previous section.


1.  Expenditures


Early HDHP studies, which were based on the experiences of individual employers, produced a wide range of estimates from which little could be inferred. However, more recent studies, which have been able to draw on larger samples of employers, have generally found that HDHPs reduce expenditures relative to traditional plans, with estimates ranging from 5 percent  (LoSasso, Shah and Frogner, 2010) to 14 percent (Buntin et al., 2011). The bulk of the available evidence suggests that these expenditure reductions occur predominantly among enrollees who are in average or better health (Bundorf, 2012). It is worth noting that the estimates cited above apply to total expenditures (the sum of those borne by the employer and the employees) and therefore reflect reductions in the overall amount of health care consumed rather than costs shifted to employees.


A number of studies suggest that design features of HDHPs, such as the level of the deductible and the size of employer contributions to HSA-style accounts, influence the extent to which HDHPs reduce spending relative to other plans. Generally speaking, the higher the deductible, the greater the spending reduction (Buntin et al., 2011), with the opposite being true for employer contributions to tax-deferred savings accounts (Borah, Burns and Shah, 2011; Buntin et al., 2011; Charlton et al., 2011; LoSasso, Helmchen and Kaestner, 2010). A caveat on the latter finding is that the majority of studies examined the influence of an earlier incarnation of the HSA, the Health Reimbursement Account (HRA), which did not allow unspent balances to be used for non-medical purposes at retirement and, in some cases, did not permit funds to be rolled over from year to year, thus weakening the incentives to economize on health care financed through these accounts.


2.  Utilization of Services


Bundorf (2012) concluded that the spending reductions observed among HDHP enrollees were mainly attributable to decreases in the use of prescription drugs (Buntin et al., 2011; Charlton et al., 2011; LoSasso, Helmchen and Kaestner, 2010; Parente, Feldman and Chen, 2008) and outpatient services (Borah, Burns and Shaw, 2011; Buntin et al., 2011; Charlton et al., 2011). Bolstering this conclusion is a related finding by LoSasso, Helmchen and Kaestner (2010) that the increase in spending associated with higher employer contributions to employee savings accounts was primarily due to increased spending on prescriptions and outpatient care.


In the case of inpatient and emergency care, the existing evidence is largely contradictory, with some estimates suggesting HDHPs decrease utilization of both types of care, while others indicate null or positive effects (Borah, Burns and Shaw, 2011; Buntin et al., 2011; Charlton et al., 2011; Wharam et al., 2007; Wharam et al. 2011).


3.  Use of Preventive Care


An oft-cited concern about HDHPs is that the introduction of greater cost sharing will lead insureds to reduce their utilization of preventive care, such as screening tests, which may yield no immediate benefit but may have longer-term benefit. The existing empirical work on the subject suggests that the impact of HDHPs on preventive care is mainly determined by whether the services in question are excluded from the plan’s deductible. For example, in the studies surveyed by Bundorf, there was either no effect or a small effect of HDHP enrollment on immunization rates (Buntin et al., 2011), cancer screening rates (Buntin et al. 2011; Wharam et al., 2008), or diabetes A1C testing (Buntin et al., 2011) when these services were carved out of the deductible. When they weren’t, there were significant reductions in screenings for some types of cancer, particularly breast and cervical cancer (Charlton et al., 2011). This is likely to be less of an issue going forward because the Affordable Care Act mandates that most health plans provide a wide array of preventive services at no cost (

4.  Use of High-Value Care


A related worry is that, in becoming more cost-conscious, insureds may fail to distinguish between high- and low-value care. Here, as documented by Bundorf (2012), the evidence is equivocal, with researchers finding that HDHP enrollees have maintained utilization of necessary services in the areas of maternity care (Kozhimannil et al., 2011) and emergency room visits (Wharam et al., 2007; Wharam et al., 2011), but have failed to selectively cut back on low-value care in the context of physician visits (Hibbard, Greene and Tusler, 2008) and medication adherence (Chen, Levin and Gartner, 2010; Greene et al., 2008).


5.  Price Shopping


A finding that a significant fraction of HDHP-induced expenditure reductions were the result of individuals procuring lower-cost care, rather than opting out of certain types of care or care entirely, would be of interest because it would highlight a mechanism through which cost reductions could be achieved without compromising patient outcomes. At the time of Bundorf’s review, no published study had examined whether HDHPs encourage consumers to use less expensive physicians or hospitals, but there was some evidence of substitution toward generic drugs and an increase in the use of mail order pharmacies among HDHP enrollees (Greene et al., 2008; Parente,  Feldman and Christianson, 2008). More recently, Sood et al. (2013) studied this topic using claims data on nine common outpatient services. They found that HDHP enrollees did not pay significantly different prices for eight of the nine services considered. The exception was physician office visits, but the effect was small; HDHP enrollees paid 2.3 percent less for office visits.


G.  Concluding Remarks


There is growing evidence that HDHPs reduce overall expenditures, at least in the first year after adoption. Research on whether these reductions can be sustained over longer time horizons has been hindered by the relatively short follow-up periods available in most studies. In her review, Bundorf (2012) found a maximum follow-up period of three years. More extensive observation periods will be required to determine how HDHPs affect longer-term trends, such as whether HDHPs have been successful at “bending the cost curve.”


In the same vein, researchers have been unable to ascertain whether HDHPs affect the health of enrollees. Evidence on clinical outcomes has been virtually nonexistent, presumably due to an inability to track insureds over a long enough period of time for clinical differences to become apparent. It does seem, however, that exempting preventive services from cost-sharing, as is now required by the Affordable Care Act, will lessen concerns that HDHPs might impede efforts to prevent chronic disease or detect it at its earliest stages.


Another area that, to our knowledge, has not been addressed is how HDHPs affect the value of the financial protection provided by insurance.  As discussed earlier, the fundamental tradeoff in designing an optimal insurance arrangement involves identifying the level of cost sharing that balances the efficiency gains from risk avoidance against the efficiency losses that result when insureds are induced to purchase care that isn’t worth its cost. Little is known about how to strike this balance, primarily because quantifying the value of financial protection against health risks is difficult. A recent working paper by Kowalski (2013) finds that at typical levels of cost sharing, the losses attributable to overconsumption of medical care are greater than the gains from risk protection, implying that overall efficiency would be enhanced by a move toward greater cost sharing. It should be noted that her estimates are based on the health plans offered by a single employer and have not yet been published in a peer-reviewed journal.


Finally, it is worth reiterating that the studies reviewed here were subject to many methodological difficulties. As a result, it is important to interpret the findings from any particular study, as well as those from the HDHP literature as a whole, with a degree of caution.


H.  References


1.      Baicker, Katherine, Sarah Taubman, Heidi Allen, Mira Bernstein, Jonathan Gruber, Joseph Newhouse, Eric Schneider, Bill Wright, Alan Zaslavsky and Amy Finkelstein. 2013. “The Oregon Experiment – Effects of Medicaid on Clinical Outcomes.” New England Journal of Medicine 368(18): 1713-1722.


2.      Borah, B.J., M.E. Burns, N.D. Shah. 2011. “Assessing the Impact of High Deductible Health Plans on Health-Care Utilization and Cost: A Changes-in-Changes Approach.” Health Economics 20(9): 1025-1042.


3.      Bundorf, M. Kate. 2012. “Consumer-Directed Health Plans: Do They Deliver?” Research Synthesis Report no. 24. The Synthesis Project. The Robert Wood Johnson Foundation.


4.      Buntin, Melinda, Amelia Haviland, Roland McDevitt and Neeraj Sood. 2011. “Healthcare Spending and Preventive Care in High-Deductible and Consumer-Directed Health Plans.” American Journal of Managed Care 17(3): 222-230.


5.      Charlton, Mary, Barcey Levy, Robin High, John Schneider and John Brooks. 2011. “Effects of Health Savings Account-Eligible Plans on Utilization and Expenditures.” American Journal of Managed Care 17(1): 79-86.


6.      Chen, S., R. Levin and J. Gartner. 2010. ‘‘Medication Adherence and Enrollment in a Consumer-Driven Health Plan.’’ American Journal of Managed Care 16: 43–50.


7.      Greene, J., J. Hibbard, J.F. Murray, S.M. Teutsch and M.L. Berger. 2008. “The Impact of Consumer-Directed Health Plans on Prescription Drug Use.” Health Affairs 27(4): 1111-1119.


8.      Hibbard, J.H., J. Greene and M. Tusler. 2008. “Does Enrollment in a CDHP Stimulate Cost-Effective Utilization?” Medical Care Research and Review 65(4): 437-449.


9.      Kaiser Family Foundation and Health Research & Educational Trust. 2012. Employer Health Benefits, 2012 Annual Survey. Menlo Park, CA.


10.  Kowalski, Amanda. “Estimating the Tradeoff Between Risk Protection and Moral Hazard with a Nonlinear Budget Set Model of Health Insurance.” Unpublished working paper, Yale University.


11.  Kozhimannil, K.B., H.A. Huskamp, A.J. Graves, S.B. Soumerai, D. Ross-Degnan and J.F. Wharam. 2011. American Journal of Managed Care 17(1): e17-e25.


12.  LoSasso, Anthony, Lorens Helmchen, and Robert Kaestner. 2010. “The Effects of Consumer-Directed Health Plans on Health Care Spending.” Journal of Risk and Insurance 77(1): 85-103.


13.  LoSasso, Anthony, Mona Shah and Bianca Frogner. 2010. “Health Savings Accounts and Health Care Spending.” Health Services Research 45(4): 1041-1060.


14.  Manning, Willard, Joseph Newhouse, Naihua Duan, Emmett Keeler, Arleen Leibowitz and Susan Marquis. 1987. “Health Insurance and the Demand for Medical Care: Evidence from a Randomized Experiment.” American Economic Review 77(3):251-277.


15.  Newhouse, J.P. 2004. “Consumer-Directed Health Plans and the RAND Health Insurance Experiment.” Health Affairs 23(6): 107-113.


16.  Parente, Stephen, Roger Feldman and Song Chen. 2008. “Effects of a Consumer-Driven Health Plan on Pharmaceutical Spending and Utilization.” Health Services Research 43(5):  1542-1556.


17.  Parente, Stephen, Roger Feldman and Jon Christianson. 2008. “The Impact of Health Status and Price on Plan Selection in a Multiple-Choice Health Benefit Program Including HRA and HSA Options.” Unpublished manuscript. University of Minnesota.


18.  Sood, Neeraj, Zachary Wagner, Peter Huckfeldt and Amelia Haviland. 2013. “Price Shopping in Consumer-Directed Health Plans.” Forum for Health Economics and Policy 16(1): 35-53.


19.  Wharam, J.F., B.E. Landon, A.A. Galbraith, K.P. Kleinman, S.B. Soumerai and D. Ross-Degnan. 2007. “Emergency Department Use and Subsequent Hospitalizations among Members of a High-Deductible Health Plan.” Journal of the American Medical Association 297(10): 1093–1102.


20.  Wharam, J.F., A.A. Galbraith, K.P. Kleinman, S.B. Soumerai, D. Ross-Degnan and B.E. Landon. 2008. “Cancer Screening Before and After Switching to a High-Deductible Health Plan.” Annals of  Internal Medicine 148(9): 647–55.


21.  Wharam, J.F., B.E. Landon, F. Zhang, S.B. Soumerai and D. Ross-Degnan. 2011. “High-Deductible Insurance: Two-Year Emergency Department and Hospital Use.” American Journal of Managed Care 17(10): e410-e418.


22.  Wharam, J.F., A.J. Graves, B.E. Landon, F. Zhang, S.B. Soumerai and D. Ross-Degnan. 2011. “Two-Year Trends in Colorectal Cancer Screening After Switch to a High-Deductible Health Plan.” Medical Care 49(9): 865–871.


III.  Assessing the Evidence on Workplace Wellness Programs[17]


This section focuses on summarizing the current literature on the impact of workplace wellness programs (WWP) on two primary outcomes—health care costs and employee health.  Reduced health benefits costs and improved employee and beneficiary health have been identified by the University as important goals of the health benefit changes (Shockey and Jensen, 2013).


The review of research concerning the impacts of WWP supports the following conclusions to be detailed below:


·         Workplace wellness programs appear to have, at best, a small, but statistically significant impact on a few employee health behaviors or health risks. These appear to be most commonly found in programs addressing smoking cessation or weight loss. The clinical relevance of these changes may be very modest. The long-run maintenance of the improved behaviors/risk reductions and the impact on actual health outcomes, such as mortality and morbidity, are not well established by research. The short-run improvements appear to occur primarily in programs that apply fairly significant and ongoing incentives and/or involve more than basic lifestyle management programs.


·         While some evaluations and reviews suggest significant costs savings and a return on investment (ROI) ranging from $3 to $6 for each $1 of wellness program costs, recent carefully designed studies and other reviews of programs, including studies from Highmark cited by the University, suggest a range of cost-benefit ratios closer to 1.5 to 2.0. Furthermore, studies appear to indicate that the most significant portion of these cost savings come from intensive disease management programs of those suffering an active chronic illness or at high risk for such illness; estimates of the cost-benefit of stand-alone life-style management programs or of the lifestyle management program of a more comprehensive workplace wellness program appear to be, at best, breakeven, and may even cost more than they save in health care costs. Broad and comprehensive studies of full cost savings, including not just health care costs, but absenteeism and other labor costs are extremely rare and thus no reliable conclusions about whether these might add to the return on investment of either lifestyle management or disease management programs can be drawn. The health care cost savings Penn State estimates from the first year of the workplace wellness effort in Take Care of Your Health, which is a lifestyle management program, suggests an assumption of an ROI of approximately 3 to 1. The studies noted below suggest that this cost savings from the first year of this effort may be overestimated by a factor of 3 to 6 times.


A. Methodological Limitations of Wellness Studies


Before summarizing the research literature, it is important to emphasize several caveats that were noted in Scanlon and Shea (2013), and are often mentioned in the studies referenced below. Much of the research conducted is not conducted as a randomized clinical trial (RCT), and the challenges of research design in workplace settings are often at the root of the problems.  As was described in that paper (and greater detail on these challenges is outlined there), there are several challenges to deriving good evidence from the current research literature on WWP, including.


1.  Definition and Measurement of the Wellness Program Intervention


There is no a standardized or well accepted definition of what constitutes a wellness program (Mattke et al., 2013). Thus, each evaluation is essentially assessing a different intervention.


2.  Measurement of Wellness Program Participation and Participation Intensity


Measuring the effects of an intervention requires a standard measure of who participated or received the intervention and, ideally, how much they participated or received. Often, this is difficult in a workplace setting, and different measures are used in different studies.


3.  Single Case Studies


Much of the wellness literature is based on individual case studies at a small number of employers. While useful on the one hand, such studies are limited in terms of their potential generalizability to other employers for a number of reasons.


4.  Self-Selected Participation in Wellness Programs


The fact that many of the research designs in the wellness literature do not rely on randomization raises major concerns about the external validity of study findings.  Employees who voluntarily participate in wellness programs are different than most employees, and likely in ways that are favorably correlated with program outcomes. As a result, program effects will be biased towards more favorable conclusions about the impact of wellness programs. One of the consistent findings noted below is that studies that employ elements of randomized design of participation typically find much smaller impacts of wellness programs.


5.  Simultaneous Changing of Other Components of Employer Sponsored Health Insurance   Benefits


Many employers make changes in other aspects of their health benefits at the same time they implement workplace wellness programs. As a result, an analysis of the wellness program that does not account for these other changes might attribute outcomes to the wellness program.


6.  Measurement of Costs or Health


Assessment of wellness programs requires careful assessments of both costs and health. Both areas are challenging. Costs may be limited to include only employer/plan payments for health care utilization (excluding employee payments), and program effects may also impact disability costs, absenteeism, and a host of other labor costs. Similarly, measurement of health outcomes, health risks, and health behaviors raises enormous challenges and has varied significantly across studies.


7.  Time Horizon Examined


Most studies of wellness programs are limited in duration often little more than 6 months to 2 years. Savings and health benefits might be concentrated in the early years (as people may slack off from their improved health behaviors) or may not occur until later (as those who do persist may be benefitted far in the future). Since many studies in the wellness literature have very short follow-up or post-intervention periods, studies may not have incorporated a long enough time horizon to get an accurate assessment.


8.  Positive Publication Bias


This effect is a common problem throughout the research literature, as studies that demonstrate an effect are much more likely to be published than those not finding an effect. Thus, the published literature may be more favorably biased towards the success of wellness programs. Indeed, in a 2010 article, Baicker et al. (p. 6) state that, “it is difficult to gauge the extent of publication bias, with programs seeing high return on investment most likely to be written about and studies with significant findings of positive returns most likely to be published.”


It is also important to clarify a few terms. This paper uses terminology found in a recent study by research from RAND which categorizes wellness programs as including one or more of three types of activities including; a) screening activities to identify health risks; b) lifestyle management including primary prevention (e.g., weight reduction programs) and secondary prevention activities (including management of existing diseases or disease management); and c) health promotion activities (smoking bans, healthy options in cafeterias, occupational health interventions) (Mattke et al., 2013). As noted above, one of the key emerging elements in research is the differential impact of lifestyle management versus disease management programs.


B.  The Impact of Workplace Wellness Programs on Health


Scanlon and Shea (2013) referenced a handful of studies on the impact of WWP, including several that were cited by Penn State in explaining its reasons for implementing Take Care of Your Health. Only one of the studies cited by Penn State referenced a detailed study of the health outcomes associated with WWP, a report from the Centers for Disease Control’s Task Force on Community Preventive Services (2010). The report, Recommendations for Worksite-Based Interventions to Improve Worker’s Health, assessed the value of workplace Health Risk Assessment Programs (e.g., HRAs) and other commonly used workplace interventions such as smoking cessation policies and immunization initiatives.


The study reviewed the evidence on assessment of health risk with feedback (AHRF), defined to include the following three elements, according to the report (p. S233): (1) the collection of information about at least two personal health behaviors or indicators; (2) translation of the information collected into one or more individual risk scores or categorical descriptions of risk status; and (3) provision to the participants of feedback regarding their risk status, either overall or with respect to specifıc risk behaviors, as well as AHRF Plus, defined to be AHRF programs that then include  other intervention components in a worksite setting.


The CDC task force recommendations are mixed regarding the effectiveness of these various workplace interventions. In its review, the CDC task force indicated that there was insufficient evidence about the effectiveness of AHRF’s when implemented alone as a primary intervention. For example, the CDC task force explained:


“The [CDC] Task Force finding of insufficient evidence to determine effectiveness is based on concerns with recurring combinations of flaws in individual studies across the body of evidence. The most important concern was the paucity of comparative studies in which the intervention was offered to one defined population and outcomes compared to another defined population that received a lesser (or no) intervention. Many of the studies identified in this review provided the intervention of interest (AHRF alone) to the “control” arm of a trial that was primarily intended to evaluate the effectiveness of a more comprehensive intervention that included AHRF as a single component. The absence of measurements from a relevant concurrent comparison group in these studies raised the potential for bias in the estimated intervention effects, particularly for self-reported changes in behavior. Most studies analyzed only a small subset of participants for whom there were complete follow-up data, which may have favored the inclusion of results from individuals who had changed their health behaviors in the interval.” (Task Force on Community Preventive Services, (2010), p. S232).


While the CDC task force found that AHRF by itself was not supported in the published evidence, the task force did recommend AHRF when combined with meaningful worksite intervention programs, AHRF Plus. When combined with well-developed health education programs, the CDC task force found the potential for positive outcomes related to a number of behaviors, including tobacco and risky alcohol use, increased physical activity, and other behaviors.


A recent review article on workplace wellness programs, Horwitz, Kelly and Dinardo (2013) suggests that the expectation that workplace wellness programs will be effective is based on three key assumptions. First, that wellness programs can accurately identify employees with specific health risks and effectively target incentives to employees for participation in wellness interventions to address these risks more effectively than usual care (i.e., care without workplace wellness programs). Second, financial incentives to participate in wellness programs will lead to behavior change on the part of employees that will improve health. Third, improvements in health will result in cost savings for employers. The first two are critical for being able to have an impact on employee health.  The authors find little evidence of improved health. In examining weight loss programs, they find four comprehensive literature reviews, none of which find evidence of long-term sustained weight loss. Their review of smoking cessation programs similarly found that programs had initial effects, but the long-term impact was often nil. They found no comprehensive, high quality reviews of programs to manage high blood pressure or cholesterol.


The major effort by a group of researchers at RAND (Mattke, et al., 2013), as well as a systematic review of evidence by a related group of investigators in Osilla et al. (2012), both tried to summarize the impacts of these programs on the health of employees.  The RAND report concludes that “lifestyle management interventions can reduce risk factors, such as smoking, and increase healthy behaviors, such as exercise. We find that these effects are sustainable over time and clinically meaningful.” The review by Osilla, et al. is less positive, noting mixed evidence for positive impact on health related behaviors, substance use, and risk factors. Reviewing a total of 33 studies published from 2000-2011, they examined outcomes in physical activity, diet, body mass index/weight, mental health, tobacco use, and alcohol use. In most cases, except for tobacco use and alcohol use where effects were most favorable, they found that approximately half of the studies showed improvements and half showed no improvement or decline. Studies with nonrandomized designs found positive health effects approximately 75 percent of the time, while randomized control trials were less likely to find effects, raising the question of whether selection bias explains some or all of the health effects.


The findings by Osilla, et al. (2012) on the impact of tobacco cessation programs seem to be supported by other studies, although not universally so. Cahill and Perera (2011) examine the impact of financial incentives in workplace wellness programs, reviewing 19 studies from 2000-2010. They find only one study demonstrated significantly higher quit rates for a group receiving incentives compared to a control group. They conclude that while programs and financial incentives may meet with early success, this tends to dissipate when rewards are removed, leading to no enhancement in long-term cessation rates.


Several recent studies have also looked at the impact of WWP on weight control, obesity or weight loss. Paloyo at al. (2013) reviews studies between 1972 and 2010 on the impact of financial incentives on weight control. They find that the results of studies vary widely due to programmatic and contextual differences, making it difficult to draw conclusions.  Cawley (2013) examined a workplace wellness program with incentives for weight loss implemented for over 2,000 employees at 24 worksites.  As with many similar worksite wellness programs, the intervention had high attrition, with 43 percent dropping out within 3 months, and less than a third continuing for a full year. The study finds that some of the programs resulted in a 5-8 pound greater weight loss among participants, although caution that selection bias may explain the result. RAND’s assessment of their data (Mattke, et al., 2013) found that weight control programs appear to have an impact on reducing employee weight by approximately one pound, and that incentives can add to this effect, but only in a small amount which is unlikely to be clinically meaningful.


Other studies have looked at an array of impacts on cardiovascular disease (CVD) risk factors, including, but not limited to weight control. Aneni, et al. (2014), for example, reviews 29 papers on the impact of WWP on CVD risk, including 18 randomized trials.  They note the problems with duration, nonstandardized interventions and measures, and selection.  With regard to programs addressing weight control, they find that greater than 50 percent of the studies show a favorable outcome, echoing the results above, as well as those in Benedict and Arterburn’s (2008) review.  On the other hand, they find no improvement in almost all WWP addressing physical activity, insufficient evidence on dietary outcomes, and too little evidence on blood pressure, lipid control, and glucose management. They do point to a few high quality studies that focus on high-risk populations and reduce hypertension. Cahalina (2014) also examines the impact of worksite health and wellness on CVD risk factors.  They conclude that there is consistent evidence that programs can achieve modest weight reductions, but weak evidence that other CVD risk factors are reduced.  As with Aneni, they highlight that health benefits are related to more comprehensive interventions and/or targeted interventions that go beyond basic lifestyle management efforts.


Two additional recent studies find evidence of favorable health effects from WWP, though both suffer from many of the weaknesses of other studies. Maeng, et al. (2013) examine the Geisinger MyHealthRewards program.  They find that the program reduced stroke and myocardial infarction by 0.5 to 1 percent among participants during a 4 year period. Rolando, et al. (2013) examined the impact of a risk factor management program on the development of diabetes, and found that those that managed their weight to a body mass index under 30 were less likely to develop diabetes. The study examined individuals who had completed 8 consecutive years of health risk assessments (excluding all those who had missed any assessments), and examined a number of interventions. Only BMI management was related to diabetes, with all programs to manage other risk factors--moving from a high-fat to low-fat diet, better stress management, changes in amount of sleep, type of grain in diet, amount of salt in diet, fruit and vegetable intake, reduced alcohol use, blood use--having no significant effect on future development of diabetes in this population.


In summary, workplace wellness programs appear to have, at best, a small, but statistically significant impact on a few employee health behaviors or health risks. These appear to be most commonly found in programs addressing smoking cessation or weight loss. The clinical relevance of these changes may be very modest. The long-run maintenance of the improved behaviors/risk reductions and the impact on actual health outcomes, such as mortality and morbidity, are not well established by research. The short-run improvements appear to occur primarily in programs that apply fairly significant and ongoing incentives and/or involve more than basic lifestyle management programs


C.  Impact of Workplace Wellness Programs on Costs


The second impact of WWP to be considered in this document is the impact on employer costs. When questions were submitted to the Penn State Benefits Office by the Faculty Senate in advance of its meeting on Tuesday September 10, 2013, one question specifically addressed the evidence base for the design of the “Take Care” program. In its response, the university referenced three studies that focused primarily on the cost savings associated with wellness programs (see



Before considering other evidence, we briefly summarize the conclusions of those studies.


One study cited in the University’s response was authored by Katherine Baicker and colleagues in 2010 and published in the journal Health Affairs. In that article, Baicker and colleagues report a positive return on investment of $3.27 in reduced health care sending per $1 spent on wellness programs. In addition, the authors report a positive return related to reductions in employee absenteeism, estimated to be valued at $2.73 per dollar spent on wellness programs. The authors also report that there are important caveats listed in the limitations section suggesting caution when projecting to other employers. For example, the authors state; “Our analysis cannot address the important question of which attributes of wellness programs are most important, and how such programs should be optimally designed. Well-designed field experiments that compare the effectiveness of program components such as patient education and professional counseling across different industries and populations are needed to answer it” (Baicker et al., 2010, p. 7).


The university’s response to the Faculty Senate question on evidence also referenced two published Highmark studies as evidence for the proposed “Take Care” program. One study published in 2011 (Naydeck et al., 2008) examines the impact of a wellness program initiated by Highmark for its own employees in 2002, estimating the impact of the program on costs for four years after wellness program implementation, and calculating an overall ROI. Reported results suggest an overall ROI of $1.76 for every $1 spent on the wellness program. Specifically, the authors estimated that wellness program participants had annual health care expenditures that were $176.47 lower than those that did not participate in the wellness program, with the majority of the savings resulting from lower hospitalization costs ($181.78) for program participants versus non-participants. While this study is interesting, there are a number of limitations regarding participation, cost allocations, or other co-occurring changes that are emphasized by the author or not clear in the methods.


The second Highmark study, published in 2011 (Williams and Day, 2011), attempted to examine the impact and value of “web based wellness program components” added to existing wellness programs during the period of 2004-2007. The study utilized employees of employers with Highmark coverage that adopted the web-based wellness features (the treatment group) and compared outcomes relative to employees of other employers with Highmark coverage that did not adopt any wellness program components (the control group). The results showed lower costs for program participants relative to non-participants, and also suggested that web based content can have value. Like the previous study however, there are a number of limitations, including measurement of program participation, whether employee out-of-pocket costs were accounted for in the overall costs of the wellness program, and a lack of details regarding important differences in benefit designs across the employers of the employees included in the study sample. All of these are important issues in the literature generally, as identified above, and as such pose significant caveats for Penn State.


One of the most important new research articles is the RAND research team’s recent paper examining the PepsiCo wellness programs; the paper studies its cost savings over a 7-year period, which is one of the longest studies of employer based programs. (Caloyeras, et al., 2014; see also an earlier PepsiCo evaluation by the same team in Liu, 2013 and the overall RAND report in Mattke, 2013).  Their results indicate that those individuals who had continuous participation in the programs over a 7-year period achieved a small, but significant savings of $30 per person per month.  This represented an overall ROI of 1.46.


The researchers also looked into the issue of savings from lifestyle management versus disease management programs. They found lifestyle management programs had no significant long-term savings associated with participation in them, while disease management programs—those focused on the currently chronically ill and/or identified high risk populations—achieved savings of approximately $136 per person per month.  The researchers also consider why their results lead to smaller estimates than the well-known study by Baicker and colleagues and write: “…we closely reviewed the seven papers that Baicker and colleagues analyzed. First, five papers looked at programs that operated more than twenty years ago, a time in which smoking was permitted in offices and statins were just emerging. These factors make it likely that the gains from lifestyle management interventions were higher twenty years ago than they are today. Second, the studies have a variety of methodological weaknesses, such as a lack of statistical controls for health status; a lack of adjustment for concomitant participation in disease management; and data limitations, such as imputation of costs from self-reported use. Lastly, the included populations are not easily generalizable—one study was of retirees and another was of 1,000 small-town city employees” (Caloyeras, et al., 2014, p.129).


The RAND results are similar to another recent evaluation conducted by John Nyman and colleagues of a worksite program at the University of Minnesota over a 3 year program (Nyman, et al., 2012). This study found an ROI from the overall program of 1.76, with the results driven entirely by the disease management features of the program, with no cost savings associated with lifestyle management programs. A related study on a fitness rewards program at the university showed a small cost reduction effect for those employees who were persistent participants in the fitness program, however even that effect was sensitive to the inclusion or exclusion of a small number of high cost persons in the employee population (Abraham, et al., 2012).


Gowrisankaran et al. (2013) examine a hospital based wellness program in St. Louis. The program was largely a lifestyle management program. During the two-year period examined, there was an overall decrease in hospitalizations of 12 percent (41% for key targeted conditions) leading to a decrease in inpatient costs of more than $22 per person per month. Those inpatient costs, however, were offset by a rise in outpatient costs (primarily medications and doctor visits) of $19 per person per month.  Factoring in the direct costs of the wellness programs as well, the authors conclude that despite widespread participation in the program, there was no net change in health care costs.


Lerner, et al. (2013) conducted a comprehensive literature review on the economic impact of employee focused health promotion/wellness programs. After examining more than 2,000 studies, the authors identified 44 that met inclusion criteria and 10 as “high quality” studies.  They found that only 3 of these studies looked at direct and indirect costs and concluded that the evidence is too limited to draw any conclusions about the impact of programs on costs. A similar conclusion was drawn in the previously referenced Osilla et al. (2012) study, which cited mixed evidence and poor research design as complicating factors for drawing conclusions about program impacts on costs.


A literature review conducted by van Dongen et al. (2011) critically appraised studies of programs targeting nutrition or physical activity behavior changes. While they found a positive economic return to programs in the initial years for absenteeism and health care costs, they also conducted a subanalysis comparing results in randomized designs versus observational designs. They found the favorable cost outcomes were concentrated in the observational studies, with studies that incorporated randomization typically finding costs were increased or not affected.


Several recent studies and reviews provide a more positive assessment of the economic impact of workplace wellness programs. Grossmeier et al. (2013) provide an evaluation of a comprehensive population health management program conducted by BP that included both lifestyle management (LM) and disease management (DM) elements. Using a quasi-experimental design and propensity score matching to provide controls, they estimate savings over a short 1-2 year period from the program. They estimate savings of $60.65 per participant per month in the LM programs and $214.66 per DM participant per month, and a full ROI from both programs of $3 for every program dollar spent. In addition to the caveats noted at the beginning of this section on wellness programs—which apply to essentially all of the studies reviewed—Grossmeier’s research team truncates costs for high cost individuals, essentially removing all costs above a threshold prior to their analysis. The thresholds range from $8,278 to $14,241. In total, as a result, costs were reduced for 45 program participants and 27 nonparticipants.  Since more high-cost participants were excluded, and it is well known in health services research that the high-cost spenders represent a significant share of total health spending, this truncation of costs may affect the estimated economic impact.


Maeng et al. (2013), noted above, also estimates cost savings from the Geisinger MyHealthRewards program. They found costs were reduced by 10-13 percent during the second and third years of the program, creating a ROI of 1.6, but those costs savings were not significant in later years.


Soler et al. (2010), also referenced above, reviews the impact of AHRF and AHRF Plus programs on health services use, but not costs.  Examining 5 studies that relied on AHRF-type wellness programs, they find mixed evidence, with some indications that these programs improve (increase) appropriate use of preventive care, but results that are difficult to interpret on their impact on reducing inappropriate care.  In reviewing 6 studies that evaluate AHRF Plus programs, they find a consistent pattern of increases in use of preventive services in most and a decline in hospital use in two studies. They conclude that the evidence supports a favorable impact of these programs on health services use, though they do not estimate what impact this might have on costs. The Gowrisankaran study above had a similar pattern of health services use change (lowered hospitalization and increased outpatient care), with no impact on costs.


Chapman (2012) conducts a comprehensive meta-evaluation of 62 worksite health promotion evaluations from 1985 to 2012. The studies are scored according to seven criteria, including research design, sample, size, baseline delineation, measurement quality, appropriateness and replicability of the intervention, observational length and recency, and then ranked from highest scoring to lowest. He summarizes economic impacts of the studies in percentage change in absenteeism, percentage change in health care costs, percentage change in worker’s compensation or disability costs, and overall cost-benefit ratio. Averaging over all the studies that reported results, Chapman finds an average reduction in absenteeism of 25% (26 studies), an average reduction in health care costs of 25% (32 studies), an average reduction in worker’s compensation costs of 32% (7 studies) and an average cost-benefit ratio of 5.56 (25 studies). He also comments that “…newer prevention technologies are also associated with higher levels of economic impact….instead of the typical 1:3.0 cost-benefit ratio they report 1:6.1”. However, it is not clear which studies he includes in this assessment. The 12 most recent studies as indicated by his ranking system have an average cost-benefit ratio of 5.3 (median of 4.2). Of the 11 studies Chapman ranks as highest quality, the average ROI is 5.9, but this is driven primarily by one study with a reported 19.4 ROI. The other 4 studies range from 1.65 to 3.60. If we extend to Chapman’s 19 top ranked studies, the ROI is 5.3.


Significant questions about Chapman’s approach appear upon closer examination.  For example, examination of the one outlier reporting an ROI of 19.4 (Harvey et al, 1993) raises questions overall about the ranking methods used in the study. The study examined a WWP introduced in the late 1980s in the city of Birmingham. While the study reports costs savings of 50 percent over trend, the employer also introduced a health maintenance organization (HMO), a simultaneous change in insurance coverage that can threaten internal validity, as noted above. In a very short time period, nearly 90 percent of employees switched into the HMO; neither the authors nor Chapman can disentangle the effects of the two changes, and there is no way to conclude that the savings found are the result of the workplace wellness programs. Given that this study is ranked as the fourth best study of the 62 examined, a more detailed examination of the other studies should be considered before accepting the reported ROI estimates.


In summary, while some evaluations and reviews suggest significant costs savings and ROIs ranging from $3 to $6 for each $1 of wellness program costs, recent carefully designed studies and other reviews of programs, including studies from Highmark cited by the University, suggest a range of cost-benefit ratios closer to 1.5 to 2.0. Furthermore, studies appear to indicate that the most significant portion of these cost savings from come from intensive disease management programs of those suffering an active chronic illness or at high risk for such illness.  Estimates of the cost-benefit of stand-alone life-style management programs or of the lifestyle management program of a more comprehensive workplace wellness program appear to be, at best, breakeven, and may even cost more than they save in health care costs. Broad and comprehensive studies of full cost-savings, including not just health care costs, but absenteeism and other labor costs are extremely rare and thus no reliable conclusions about whether these might add to the return on investment of either lifestyle management or disease management programs can be drawn.


D.  Additional Considerations


Several authors, including proponents of workplace wellness programs and those who question their effectiveness, tend to share the view that for such programs to have their most positive effects on employee health and health care costs, several conditions must be in place. While many of the studies cited note these factors, the recent RAND review provides an excellent summary of these “Key Facilitators of Successful Wellness Programs” (Mattke et al. 2013, pp. xxiv).  They include:


1.  Effective communication strategies


Successful programs include broad outreach, clear leadership messaging, and face-to-face interaction, especially important for large geographically dispersed employers


2.  Opportunities for employee engagement


Engagement on both health and the impact of health costs on employers and employee is critical, long before even introducing programs. As noted in the initial white paper: “Importantly, as many analysts have argued, simply providing workers with information on their health risks will not necessarily change long established behaviors. Instead, as models of patient engagement and health behavior have illustrated, changing health habits is difficult, and is often a long incremental process that takes time, and most occur at a pace the employee is comfortable with” (Mittler et al., 2013; Prochaska et al., 1992, and Hibbard et al., 2004) Once programs are introduced, engagement requires activities to be convenient, accessible, targeted to employees and low cost.


3.  Leadership engaged at all levels


Senior leaders must consider employee health to be an organizational priority, and must invest time and effort in explaining the wellness and culture changes to all direct supervisors, and provide these supervisors with the materials, time, and resources to communicate with employees.


4.  Use of existing resources and relationships


Successful programs build from resources offered through health plans or other providers, but insist on tailoring programs, communications, and other materials to their specific workforce and worksite. Successful programs often require significant environmental changes and have included onsite exercise facilities, changes in vending, cafeteria, and food service programs, and building modifications to enhance physical activity.


5.  Continuous evaluation


While successful programs do not always conduct regular evaluations of wellness, regular feedback, needs assessments, and adjustments is often part of the successful programs.


Both supporters and those who question the value of WWP agree that the probability of success in improving employee health and reducing health costs is lowered if employers do not first establish a strong foundation for the program through these key preconditions. As noted in a Center for Studying Health System Change report (Tu and Mayrell, 2010), there is a key message for all employers seeking to address concerns about health care costs as they examine the opportunities and challenges from the Affordable Care Act:


“Employers looking to wellness as a quick fix for high health costs are those least likely to see positive returns, as they are also the least likely to have undertaken the measures to gain true employee engagement in health.” (p.12)


E.  References


1.  Abraham, J., Nyman, J., Feldman, R., and Barleen, N. (2012). “The effect of participation in a fitness rewards program on medical care expenditures in an employee population”. Journal of Occupational and Environmental Medicine, 54(3):280-285.


2.  Aneni E.C., Roberson L.L., Maziak W., Agatston A.S., Feldman T., Rouseff, M., Tran, T., Blumenthal, R., Blaha, M., Blankstein, R., Al-Mallah,M.,  Budoff, M. (2014) A Systematic Review of Internet-Based Worksite Wellness Approaches for Cardiovascular Disease Risk Management: Outcomes, Challenges & Opportunities. PLoS ONE 9(1): e83594.


3.  Baicker, K., Cutler, D., and Song, Z. (2010). “Workplace Wellness Programs Can Generate Savings”. Health Affairs: 29:2304-311.


4.  Benedict M., Arterburn D (2008) Worksite-based weight loss programs: a systematic review of recent literature. American Journal of Health Promotion, (22):408–416.


5.  Cahalin,L., Myers, J., Kaminsky, L, Briggs, P., Forman, D., Patel,M.,  Pinkstaff, S.,  Arena, R. (2014). “ Current Trends in Reducing Cardiovascular Risk Factors in the United States: Focus on Worksite Health and Wellness” . Progress in Cardiovascular Diseases 56(5): 476-483.


6.  Cahill K, Perera R. (2011). Competitions and incentives for smoking cessation. Cochrane Database Systematic Reviews. 2011 Apr 13;(4):CD004307.


7.  Caloyeras J. P., Liu H., Exum E., Broderick M., Mattke S. (2014). "Managing manifest diseases, but not health risks, saved PepsiCo money over seven years". Health Affairs, 33(1):124-131.


8.  Cawley, J. and Price, J. (2013). “A Case Study of a Workplace Wellness Program That Offers Financial Incentives for Weight Loss”. Journal of Health Economics, (32):794-803.


9.  Chapman, L. (2012). “Meta-Evaluation of Worksite Health Promotion Economic Return Studies:  2012 Update” in The Art of Health Promotion, American Journal of Health Promotion, Cleveland, OH.


10. Grossmeier, J., Seaverson, E., Mangen, D., Wright, S., Dalal, K., Phalen,C.  and Gold, D.  (2013). “Impact of a Comprehensive Population Health Management Program on Health Care Costs”. Journal of Occupational and Environmental Medicine, 55(6):634-643.


11. Gowrisankaran, G., Norberg, K., Kymes, S., Chernew, M., Stwalley, D., Kemper, L., Peck, W. (2013).”Wellness Programs: A Hospital System’s Wellness Program Linked To Health Plan Enrollment Cut Hospitalizations But Not Overall Costs”. Health Affairs, (32):3477-485.


12. Harvey M., Whitmer R.,  Hilyer J., Brown K. (1993). “The impact of a comprehensive medical benefits cost management program for the city of Birmingham: results at five years”. American Journal of Health Promotion, 7:296–303.


13. Hibbard, J.H., Stockard, J., Mahoney, E.R. and Tusler, M. (2004). “Development of the Patient Activation Measure (PAM): Conceptualizing and Measuring Activation in Patients and Consumers.” Health Services Research, 39(4 PT1): 1005-1026.


14. Lerner, D., Rodday, A, Cohen, J.,  and Rogers, W. (2013). “A Systematic Review of the Evidence Concerning the Economic Impact of Employee-Focused Health Promotion and Wellness Programs”. Journal of Occupational and Environmental Medicine, 55(2):209-222.


15. Liu, H., Mattke, S., Harris, K., Weinberger, S., Serxner, S., Caloyeras, J., Exum, E. (2013). “Do Workplace Wellness Programs Reduce Medical Costs? Evidence from a Fortune 500 Company”. INQUIRY: The Journal of Health Care Organization, Provision, and Financing, 50(2) 150–158.


16. Mattke S., Liu H., Caloyeras J.P., Huang C.Y., Van Busum K.R., Khodyakov D., and Shier V. (2013). “Workplace Wellness Programs Study: Final Report”, Santa Monica, Calif. RAND Corporation, RR-254-DOL (available at


17. Mittler, J.N., Martsolf, G.R., Telenko, S, and Scanlon, D. (2013). “Making Sense of Consumer Engagement Initiatives: A Conceptual Framework.” Milbank Quarterly, 91(1):37-77.


18. Mountz, A. (2013). “Health care changes to be discussed at Faculty Senate meeting”. Retrieved March 19, 2013 from


19. Naydeck, B.L., Pearson, J.A., Ozminkowski, R.J., Day, B.T. and R.Z. Goetzel. (2008). “The Impact of the Highmark Employee Wellness Programs on 4-Year Healthcare Costs.” American College of Occupational and Environmental Medicine, 50:146-156.


20. Nyman, J., Abraham, J., Jeffery, M., and Barleen, N. (2012). “The effectiveness of a health promotion program after 3 years: Evidence from the University of Minnesota”. Medical Care. 50(9):772-778.


21. Osilla, K. , Van Busum, K. Schnyer, C., Wozar Larkin, J. Eibner, C. and Mattke, S. (2012). “Systematic Review of the Impact of Worksite Wellness Programs”. American Journal of Managed Care: 18(2):e68-e81.


22. Paloyo, A.R., Reichert, A.R., Reinermann, H., Tauchmann, H. (2013). “The causal link between financial incentives and weight loss: an evidence-based survey of the literature.” Journal of Economic Surveys, Article first published online: 25 JAN 2013.


23. Prochaska, J.O., DiClemente, C.C., and Norcross, J. (1992). “In search of how people change: Applications to the addictive behaviors.” American Psychologist 47, 1102-1114.


24. Scanlon, D. and Shea, D. (2013). “Assessing the Evidence for Penn State University’s “Take Care of Your Health” Benefits Program”.  Retreived March 13, 2014 from


25. Shockey, Jill and Jensen, L. Reider. (2013). “Benefits changes focus on employee wellness, long-term cost savings”. Retrieved March 13, 2013 from


26. Task Force on Community Preventive Services. (2010). “Recommendations for Worksite-Based Interventions to Improve Worker’s Health”. American Journal of Preventive Medicine: 38(2S):S232–S236.


27. Maeng, D., Pitcavage, J., Tomcavage, J. and  Steinhubl, S. (2013). “Can Health Insurance Improve Employee Health Outcome and Reduce Cost? An Evaluation of Geisinger’s Employee Health and Wellness Program. Journal of Occupational and Environment Medicine, 55(11): 1271-1275.


28. Rolando, L., Byrne, D., McGown, P., Goetzel, R., Elasy, T., Yarbrough, M. (2013). “Health Risk Factor Modification Predicts Incidence of Diabetes in an Employee Population Results of an 8-Year Longitudinal Cohort Study”. Journal of Occupational and Environment Medicine, 55(4): 410-415.


29. Tu, H. and Mayrell, R. (2010). “Employer Wellness Initiatives Grow, But Effectiveness Varies Widely”. National Institute for Health Reform Research Brief 1, Center for Studying Health System Change, Washington DC.


30. van Dongen, J., Proper, K., van Wier, M., van der Beek, A., Bongers, P. van Mechelen, W. and van Tulder, M. (2011). “Systematic review on the financial return of worksite health promotion programmes aimed at improving nutrition and/or increasing physical activity”. Obesity Reviews, 12:1031-1049.


31. Williams, L.C. and B.T. Day. (2011). “Medical Cost Savings for Web-Based Wellness Program Participants From Employers Engaged in Health Promotion Activities.” American Journal of Health Promotion, 25(4):272-80.

IV.  Tiered Benefit Programs[18]


A.  Background


Tiered benefit designs aim to channel consumers to “preferred” providers (or treatments), which are determined based on their high performance on cost, quality or efficiency measures. By placing preferred providers in a tier with the lowest cost-sharing, tiered benefits incentivize consumers to select high-performing providers. They are thus expected to achieve cost saving and quality improvement without restricting consumers’ choice of providers.


Tiered cost-sharing has rapidly grown for pharmacy benefits since the late 1990s, and is currently a nearly universal feature. In 2013, 92% of enrollees in employer-sponsored plans have tiered schemes for their prescription drug coverage, where generic drugs are placed in a tier with the lowest copayments and brand-name drugs are assigned to one or more tiers with higher copayments (Kaiser Family Foundation, 2013). While much less prevalent than for drug coverage, tiered benefits have been increasingly used for hospital or physician services (Kaiser Family Foundation, 2013): among employers offering health insurance, 23% include a tiered provider network in their plan with the largest enrollment in 2013. This corresponds to a 53% increase from 2007 (15%).


B.  Tiered Incentive Designs for Pharmacy Benefits


An extensive literature has examined how tiered incentives for pharmacy benefits influence consumers’ drug choice, overall drug use, and total drug spending. The literature generally suggests that a higher or a greater increase in the copayment for drugs in a certain tier led to a decrease in use of those drugs and total drug spending. However, the results on overall drug use are mixed, implying that consumers facing differential copayments do not always substitute less costly drugs for those with similar clinical indications but higher copayments. Gibson et al., (2005) and Gibson et al. (2010) review this literature. Below is a summary of findings of several key articles.


A few studies examined consumers’ choice between preferred and non-preferred branded drugs under tiered benefits. Huskamp et al. (2005) analyzed changes in drug use among employees at a firm that switched its drug benefits from a 2-tier (generic and branded drug tiers) to a 3-tier (generic, preferred brand, and non-preferred brand tiers) scheme and increased the copayment only for non-preferred branded drugs from $12 to $24 (the copayments for generic and preferred branded drugs were $6 and $12, respectively). The authors used workers at a firm with no copayment change as a control group and found that the introduction of the 3-tier structure decreased non-preferred branded drug use but did not influence overall drug use. However, another study using a similar setting and design found that consumers decreased their overall drug use as well as non-preferred branded drug use after facing the change in their pharmacy benefit from a 2-tier to a 3-tier scheme (Motheral and Fairman, 2001).


Several studies investigated how consumers changed their drug use when facing a greater increase in the copayment for brand-name drugs than that for generic drugs. Motheral and Henderson (1999) analyzed data from firms that increased the copayment for branded drugs from $10 to $15 while changing the copayments for generic drugs from $4 to $5 or $5 to $7. They reported that total drug spending and branded drug use among employees at those firms decreased significantly compared with those by workers at a firm with no change in copayments ($10 for branded drugs; $5 for generic drugs). This study showed no changes in the total drug use and chronic medication use. Another analysis found that enrollees facing the copayment increase from $2 to $7 for branded drugs (no price change for generic drugs) had a decrease in total drug spending and branded drug use (Gibson et al., 2005b). However, this analysis indicated that the reduction in spending diminished over time. Further, unlike Motheral and Henderson (1999), it found that generic drug use and total drug use also decreased after the price change for branded drugs, suggesting that consumers discontinuing the use of branded drugs did not always switch to generic drugs.


Studies that used variation in cost-sharing across plans or analyzed different populations also reported similar findings. For example, a study of Medicare beneficiaries receiving retiree benefits showed that beneficiaries in plans with tiered schemes had 14.3% lower total drug spending, 14.6% fewer fills, but 57.6% higher out-of-pocket spending than those in plans with no tier (Gilman and Kautter, 2008).


In summary, the literature on pharmacy benefits generally concurs that tiered schemes effectively steer consumers away from non-preferred drugs. However, it also indicates that consumers may reduce overall drug use by discontinuing medications whose copayment increased, which suggests tiered benefits may limit access to certain drugs when drugs in preferred and non-preferred tiers are not perfect substitutes.


C.  Tiered Provider Network


Tiered benefits for hospital care or physician services are far less prevalent than for drug coverage, which is partially because provider performance is harder to measure and standardize than pharmaceuticals. Various methods can be used to place providers into specific tiers, but placement has often been based on performance on quality and/or cost-efficiency measures (Sinaiko, 2011). Literature evaluating tiered incentives for hospital or physician services is scant. Only a few studies have examined provider choice under tiered schemes, and no evidence from the academic literature exists on the impacts of tiered benefits on cost savings and quality outcomes.


Scanlon et al. (2008) examined hospital choice after a large firm initiated a tiered program for hospital care. This firm changed the standard coinsurance for union employees from zero to 5%, which was waived if the union beneficiary chose a hospital that adopted certain patient safety practices, such as computerized physician order entry. Using non-union employees whose cost-sharing remained zero as a control group, the study showed that the engineer union members admitted to a hospital for medical conditions were more likely to visit a “safe” hospital than non-union engineers. However, it found no significant effect for surgical admissions or within the machinist union members. The authors discussed that their mixed finding may be due to some features of the study setting, such as a small sample size (N=517 – 1,205) and a small choice set in the preferred tier (5 out of 27 hospitals complied with the safety standards).


While limited and mixed, the findings of this study offer some evidence supporting that tiered benefits can be used as a tool to channel consumers to certain providers. However, its interpretation requires some cautions. First, the finding from a single firm may not be generalizable to other plans or firms. Second, the difference in consumer cost-sharing between preferred and non-preferred hospitals in the study was on average $403, suggesting that a large price differential may be needed to have an influence on consumers’ hospital choice.


For physician services, no research has analyzed how tiered schemes affect consumers’ actual choice of physicians. Recently, Sinaiko et al. (2010) surveyed a random sample of Massachusetts state employees, who were in plans with tiered physician networks, to examine consumers’ experience with tiered benefits. The survey results indicated a relatively low awareness and trust about the tiered benefit among consumers: about half of the respondents were aware of their tiered provider networks, only 19% knew which tier their doctor was in, and further, only 40% considered their health plan as a trustful source of information for identifying “better” physicians.


In a follow up study, Sinaiko (2011) assessed how quality and/or cost information influences consumers’ choice of physicians using an experimental approach. The experiment used the same population as above, but constructed a “hypothetical” tiered benefit structure. It randomly assigned the sample into one of six groups, each of which was shown two hypothetical tiers and was told that doctors in Tier 1 have lower costs and higher quality while doctors in Tier 2 have higher costs and lower quality. The copayment differential between the two tiers varied across groups, ranging from $0 to $35. This experiment found that copayment differences of $10 – $35 would influence 3.5% – 11.7% of consumers to select physicians in a preferred tier. A simulation from the experiment indicated copayment differentials greater than $300 would be necessary to counteract recommendations of physicians from friends, family or referring physicians.


Tiered provider networks are theoretically appealing because they give an incentive for patients to select high-value providers, which can help reduce costs and/or improve quality. However, little empirical evidence on impacts of tiered provider networks exists, making it difficult to predict outcomes of adopting tiered incentives for hospital care and physician services.


D.  Conclusions


Prior evaluations of tiered schemes for pharmacy coverage generally indicate that consumers reduce use of drugs that are placed in tiers with higher copayments. However, evidence on the impacts of tiered benefits for hospital or physician care is sparse. The results from a few studies available suggest that a large copayment difference would be needed to effectively influence consumers’ choices of providers.


 As discussed above, ultimate goals of tiered benefit designs are to improve quality or value, and save costs. As such, it is important to consider alternative approaches to achieve those goals. For example, one similar concept that has recently appeared in the market is reference pricing (Robinson and MacPherson, 2012; Robinson and Brown, 2013). Reference pricing is a benefit design that purchasers or insurers make a fixed contribution toward the full price of a service, and if patients choose a provider charging a higher price than the defined contribution, the remainder is patients’ responsibility. Reference pricing is expected to work well for services with wide cost variation but small quality difference. While the evidence base for the impacts of alternative approaches on saving costs and improving value may also be limited, it would be worthwhile for those considering adopting tiered incentives to assess costs and benefits of possible alternatives.


E.  References


1.  Gibson, T. B., Ozminkowski, R. J., & Goetzel, R. Z. (2005a). The effects of prescription drug cost sharing: a review of the evidence. The American Journal of Managed Care, 11(11), 730-740. Retrieved from


2.  Gibson, T. B., McLaughlin, C. G., & Smith, D. G. (2005b). A copayment increase for prescription drugs: the long-term and short-term effects on use and expenditures. Inquiry, 42(3), 293-310. doi: 10.5034/inquiryjrnl_42.3.293


3.  Gibson, T. B., McLaughlin, C. G., & Smith, D. G. (2010). Generic utilization and cost-sharing for prescription drugs. Advances in health economics and health services research, 22,      195-219. doi: 10.1108/ S0731-2199(2010)0000022012


4.  Gilman, B. H., & Kautter, J. (2008). Impact of multitiered copayments on the use and cost of  prescription drugs among Medicare beneficiaries. Health Services Research, 43(2), 478-495. doi: 10.1111/j.1475-6773.2007.00774.x


5.  Huskamp, H. A., Frank, R. G., McGuigan, K. A., & Zhang, Y. (2005). The Impact of a ThreeTier Formulary on Demand Response for Prescription Drugs. Journal of Economics & Management Strategy, 14(3), 729-753. doi: 10.1111/j.1530-9134.2005.00080.x


6.  Kaiser Family Foundation and the Health Research & Educational Trust. (2013). Employer health benefits annual survey. Menlo Park, CA: The Henry J. Kaiser Family Foundation.


7.  Motheral, B., & Fairman, K. A. (2001). Effect of a three-tier prescription copay on pharmaceutical and other medical utilization. Medical care, 39(12), 1293-1304.


8.  Motheral, B. R., & Henderson, R. (1999). The effect of a copay increase on pharmaceutical utilization, expenditures, and treatment continuation. The American Journal of Managed Care, 5(11), 1383-1394.


9.  Robinson, J. C., & Brown, T. T. (2013). Increases in consumer cost sharing redirect patient volumes and reduce hospital prices for orthopedic surgery. Health Affairs, 32(8), 1392-          1397. doi: 10.1377/hlthaff.2013.0188


10.  Robinson, J. C., & MacPherson, K. (2012). Payers test reference pricing and centers of excellence to steer patients to low-price and high-quality providers. Health Affairs, 31(9), 2028-2036. doi: 10.1377/hlthaff.2011.1313


11.  Scanlon, D. P., Lindrooth, R. C., & Christianson, J. B. (2008). Steering patients to safer hospitals? The effect of a Tiered Hospital Network on hospital admissions. Health services research, 43(5 Pt 2), 1849-1868. doi: 10.1111/j.1475-6773.2008.00889.x


12.  Sinaiko, A. D., & Rosenthal, M. B. (2010). Consumer Experience With a Tiered Physician Network: Early Evidence-Page. The American Journal of Managed Care, 16(2), 123-130. Retrieved from


13.  Sinaiko, A. D. (2011). How do quality information and cost affect patient choice of provider in a tiered network setting? Results from a survey. Health services research, 46(2), 437-456. doi: 10.1111/j.1475-6773.2010.01217.x


14.  Sinaiko, A. D. (2012). Tiered Provider Networks As a Strategy to Improve Health Care Quality And Efficiency. NIHCM Foundation. Retrieved from



Report of the Benchmarking Subcommittee


I.  Introduction


Penn State places a high priority on offering competitive compensation to attract and retain highly talented faculty and staff.  Health care benefits are an extremely important component of an employee’s total compensation.  The Health Care Task Force appointed a subcommittee to collect and compile data from peer public and private universities regarding their current health care benefits programs and information about their future health care planning.


Major benchmarking studies of employer-based health care programs might normally include hundreds or even thousands of organizations.  The data from these types of studies are highly summarized, which makes them very useful to show trends and the popularity of different approaches.  But the highly summarized nature of the results from large benchmarking studies makes it more difficult to know the combinations of approaches used by specific institutions.  The survey conducted by the Benchmarking Subcommittee was targeted to the CIC institutions[19] and the other two large state-related research universities in the Commonwealth of Pennsylvania, the University of Pittsburgh and Temple University.


II. The Survey


With assistance from the full Task Force, review and suggestions by Towers Watson and benefits experts in Penn State’s Office of Human Resources, benchmarking questions were written such that the responses would allow comparison of current plan designs, changes to plan designs that are under consideration for the future, cost trends, approaches to retiree health care, etc.  An online survey was conducted using Qualtrics software.


An initial contact was made through the Chief Human Resources Officers.  Immediately following the initial contact with each university, the survey was sent to the Chief Benefits Officers/Managers with a personalized letter from the Chair of the Task Force.  As an incentive to participate, respondents were guaranteed access to the summarized results for all of the respondents except those who requested that their responses remain confidential. 


Survey responses were received from:


            Michigan State University                             University of Illinois

Northwestern University                             University of Maryland

            Ohio State University                                    University of Michigan

            Rutgers University                                        University of Minnesota

            Temple University                                         University of Pittsburgh


Those who received the survey but declined to participate are:


            Indiana University                                         University of Iowa

Purdue University                                         University of Nebraska

            University of Chicago                                    University of Wisconsin                              

The survey was not of trivial length, and not surprisingly, those who did respond chose to omit answers to some of the questions.  Searches of institutional websites and targeted follow-up turned up information that has been used to supplement our understanding and will be noted in the following section that attempts to summarize and interpret the survey responses. 


Appendix C presents the results of the survey in table form.  A survey form has also been included for Penn State, so there are 11 universities contained in survey.  Only Penn State’s response to the survey is identified; other universities are labeled with letters.


III. The Results


This section expands on comparisons of some of the elements of the survey that are most related to the more significant aspects of the plans as well as comparisons of selected components most closely related to recent changes in Penn State’s health care plans.


A.  Insurance Arrangements


An understanding of the various insurance arrangements (e.g., self- vs. fully-insured), as well as the differences in treatment of benefit costs by the states for public institutions, is important to the comparisons of relative situations.  The different approaches could influence the likelihood that certain plan design elements are included or considered for the future.


All of the respondents indicate that they, like Penn State, are self-insured.  Some of the public universities surveyed participate in state-level plans that are self-insured, but the institution itself is not self-insured.  The survey did not include a specific question about whether benefit costs were covered in the institution’s budget, but we do know that health care costs for Rutgers University and the University of Illinois are not directly budgeted at the institution.  Instead, those costs for educational and general activities are borne by those institutions' respective states. 


The State of New Jersey, on behalf of Rutgers University, pays health and all benefits costs for a set number of positions as authorized in their appropriations act. For any positions over that number (for grants/auxiliaries, etc.), Rutgers must reimburse the state at the state-mandated fringe rate.  Rutgers financial statements reflect the cost of benefits as both revenue and expense non-cash items. 


The University of Illinois has a similar, but different, arrangement.  Most of their health care costs are budgeted and funded at the state level.  While the cost increases at the state level may influence the funding available for direct operating support, the state’s coverage of the annual increases in health care costs are, in effect, an increase in state appropriations for at least that part of the operating budget.  Due to State budget rescissions over the past decade, the University of Illinois now pays nearly $25 million to the State of Illinois Central Management Services to cover a portion of the health care benefits for University employees.


Participation in the state’s health plan is handled in yet a different manner for the University of Maryland.  The university is charged directly by the state for health care benefits for employees who are participants in the state health plans.  The charges are paid from the university’s operating budget.  Unlike the University of Illinois and Rutgers University, the University of Maryland is subject to health care cost increases as a budget driver/consideration yet, like the University of Illinois and Rutgers University, they do not have control of the plan design.


B.  Qualified High Deductible Health Plans (QHDHP) and Health Savings Accounts (HSA)


Penn State is among five of the eleven universities in the survey that currently offer a QHDHP.  Penn State has the highest participation percentage (one of the institutions did not report participation percentage across their many plans nor did they report on HSA seed funding), which can probably be explained by a number of factors.  First, the other institutions offer more health plan options while Penn State offers just this plan and a PPO plan.  Second, the much higher participation rate at Penn State could be a result of the very steep salary indexing of employee contributions that serves as a substantial incentive for employees to move to the QHDHP. 


The survey also measured specific aspects of QHDHPs because they may be critical to employees' decision making.  The design parameters of the QHDHPs, in conjunction with the level of risk an employee is willing to assume, are important considerations to employees making choices amongst various plans.  The health status of employees and covered family members also play into the level of risk employees are willing to assume.  Parameters in isolation, for example just considering the maximum out-of-pocket costs or just the deductible, do not provide all of the information needed to make the risk-adjusted decision to opt for a high deductible plan.  The survey summary provides this information, but for ease of comparing Penn State’s plan design with others, some of the most important aspects are presented together in the following table.


Selected Components of Qualified High Deductible Health Plans



Individual Coverage

Family Coverage


HSA seed funding-employee


Max Out-of-Pocket


HSA seed funding-employee


Max Out-of-Pocket




































Penn State










C.  Salary Indexing of Employee Contributions


The majority of universities that responded to the survey offer more than two health care plan choices to their employees.  The survey questions around plan design focused on the two types of plans offered by Penn State, a Preferred Provider Organization (PPO) and a Qualified High Deductible Health Plan (QHDHP). 


In addition to Penn State, four of the other ten respondents use some version of salary-based indexing for their PPO to assess employee contributions.  Six respondents use flat dollar charges regardless of salary level.


As previously noted, five of the respondents, including Penn State, offer QHDHPs.  Penn State and two other universities use some form of salary-based employee contributions for their QHDHPs, while the other two use a flat dollar cost approach.


Penn State’s flat percentage of salary is unique among the three salary-based approaches.  Institution A uses salary bands and reports their structure as a decrease in the university’s subsidy as the salary bands increase from lower salaries to higher salaries.   Institution J’s response provided a link to a website.  The website contained a calculator that revealed that as salaries increase, the percent of salary that is charged went down although the dollar amount of the charge increased, i.e., unlike Penn State’s approach, the charge is not a flat percentage of salary.

D.  Surcharges


Of the eleven universities in the survey, Penn State is alone in assessing surcharges for spouses/partners who have access to coverage through their own employer.  Material on Institution H’s website, however, indicates that spouses who have access to a health plan through their own employer at an annual cost for single-person coverage of $850 or less must purchase their own coverage.


Penn State is the only institution that reports a surcharge for tobacco users, although institution E reports that employees who do not use tobacco are eligible for a $300/$400 reduction in their annual premiums.


E.  Wellness


Eight of the eleven universities in the survey report the provision for some type of wellness program.  Six of the eight, including Penn State, report that incentives are used to encourage participation.  Institution E reports that the incentive is in the form of a premium discount for those who do not use tobacco.  Institution D gives employees $100 for completing a health questionnaire.  Institution K reports that participation can lower deductibles and out-of-pocket maximums.


Just one respondent indicated that their institution could quantify savings to the institution from their wellness initiatives.  This respondent reported that they could approximate savings of 2.5 percent of total health insurance plan costs.


F.  Health Care Benefits for Retirees


All but one of the institutions in the survey reports that they provide health care benefits for both pre-Medicare and Medicare-eligible retirees.  Further review of the website of the institution that provided a negative response indicated that, while the institution itself does not provide retiree health care, their retirees are eligible to participate in the state plan that includes subsidies for retirees.


Those institutions in the survey providing retiree health care benefits report that their pre-Medicare retirees continue in the active employee plan until they become eligible for Medicare.


In addition to Penn State, one other institution in the survey provides Medicare-eligible retirees with a Medicare Advantage Plan with prescription drug coverage.  Seven others provide a Medicare Supplement Plan, while five include a prescription drug plan and two do not.


Further website research indicated that one of the institutions in the survey has made changes to the University's contribution (subsidy) provided for retiree health care as well as changes to eligibility.  One of the changes is a reduction in the number of years of employment required to qualify for a lower subsidy as part of an overall strategy to be more competitive for mid-career or late-career faculty and staff.


G.  Future Plans


The final survey question asked institutions to report their current or planned use of various cost saving strategies.   Current and planned offerings can be significantly influenced by the individual situations of the public institutions in particular if their employees participate in their state’s health plans or if they have autonomy to design their own plans.  


As previously mentioned, five of eleven universities in the survey, including Penn State, offer a QHDHP.  Of the other six respondents, three report that they are considering offering such a plan in either the next one to two years, or three to five years.


The responses regarding many of the other strategies or tactics may be of interest in the context of the work of the Health Spending, Utilization, Value and Vendor Analyses Subcommittee.


IV. Conclusions


Comparison of benchmarking data to account for all of the differences across plans is challenging. The data compiled from the survey does, however, provide quite useful information about some of our peer institutions that could be used to inform Penn State’s future positioning to remain competitive for the best faculty and staff.


Penn State’s basic plan designs and employer share of the costs are generally consistent with those of our peers.  As noted earlier, some of the relatively new features, such as spousal and tobacco surcharges and steep salary indexing, are in sharp contrast to those in the survey sample of peer institutions.  With that said, the health care landscape is changing so rapidly that the practices of our peers today may well not reflect what they are doing in the future.  As a result, it is important for the University to benchmark the features of its health care program with that of peers on a regular basis.

Report of the Health Spending, Utilization, Value and Vendor Analysis Subcommittee

I.         Introduction

As earlier chapters indicate, Penn State and its employees and eligible beneficiaries are projected to spend $228 million dollars on health and medical care services in fiscal year 2014 (July 1, 2013 to June 30, 2014).  This amount includes required insurance co-premiums on the part of employees and retirees (averaging 20.82% of the total between 2005 and 2013), amounts paid by the University as part of its self-insured benefits structure, and administrative fees to Highmark, the University’s third party administrator (averaging 3.16% of the total between 2009 and 2013).  The amount does not include dollars paid by employees and beneficiaries in the form of deductibles, co-payments and co-insurance, as these numbers were not available to the task force, but these are likely sizeable as well and if these actual numbers are added, certainly moves the overall projected annual expenditures for 2014 to near one quarter of a billion dollars.


Importantly, while the rate at which these numbers have grown over the past ten years varies – from 13% inflation between 2008-2009 and 2009-2010 to 2% inflation between years 2011-2012, the cumulative effect of the overall expenditure inflation is estimated to be over $100 million dollars between 2005 and 2014.  There are many potential factors that drive these increased expenditures, including growth in the employed and retiree populations and covered beneficiaries, increased health care utilization due to aging and chronic disease prevalence of the employed and retiree populations, and general trends in health care inflation caused by increased utilization of health care services due to the availability of advanced medical technologies, availability of new pharmaceuticals, as well as other market based and regulatory factors such as increased provider market concentration and the impact of the Affordable Care Act (ACA).  It is not the objective of this chapter to precisely decompose the sources of historical inflation, and in fact doing so is difficult given current data availability.  However, an investment in an analysis that decomposes the sources and magnitudes of expenditure increases might be useful in helping to better understand the best levers for managing health benefits expenditure growth to the satisfaction of the university and its covered beneficiaries.


Instead, the purpose of this chapter is to consider various options for Penn State and its beneficiaries to get better value for the amount it collectively spends on health care annually.  Value is, quite simply, the benefits received for the amount spent.  In this case benefits can be measured by beneficiary satisfaction, health status, quality of care provided, and efficacy and effectiveness of health and medical care received.  Value can be improved in three specific ways:

·         Same Spend – More Benefits: Value can improve when the benefits received (i.e., improvements in satisfaction, quality of care, and health status) improve for the same amount of expenditures (e.g., the projected $250,000,000 in fiscal year 2014).

·         Less Spend – Same Benefits: Value can improve when the benefits received (i.e., improvements in satisfaction, quality of care, and health status) stay constant for less expenditure (e.g., less than the projected $250,000,000 in fiscal year 2014).

·         Less Spend – More Benefits:  Value can improve when expenditures decrease, (e.g., less than the projected $250,000,000 in fiscal year 2014), and benefits also increase (i.e., improvements in satisfaction, quality of care, and health status)

Given that better value can be achieved in various ways as illustrated above, with the two key pieces to the equation being dollars spent and benefits received for those dollars, alternatives available to the Penn State community for improving value should focus on the ability of various alternatives to impact health care spending, health care benefits, or both.  Lest this discussion seem esoteric, very similar scenarios apply in most individual and household purchasing and consumption decisions on a day to day basis.  For example, a “buy one get one free” sale provides more value by providing the consumer with more product for the same amount of expenditure.  A “20% off sale” provides more value by providing the same product at a lower overall expenditure.  Finally, the addition of competitors in a particular market (e.g., pizza delivery stores in a town) may result in an overall reduction in price and an improvement in the overall quality of the product (e.g., better pizza that includes fresher ingredients or quicker home delivery).  Thus, many believe that extracting additional value from health care markets can and should function just as many other product and service markets do in our economy.  To date, there is broad agreement that the health care market has operated inefficiently in aggregate, and thus there is significant room for improvement.  But health care markets also have features that make them distinct from other markets for products and services, and thus in order to facilitate value for the purchaser and patient, alternatives may be judged on the degree to which they address, or partially address the following areas believed to create inefficiencies in health care.  Economists often refer to these inefficiencies as “market failures”.


II.  Notable Market Failures in Health Care


A.  Lack of Quality and Price Transparency


Unlike many other commodity markets, the price and quality of specific health care services are not easily known by consumers (i.e., patients).  This lack of transparency is believed to undermine competition and to perpetuate inefficiency by making it difficult for patients to be informed about the specific benefits received for health and medical care services purchased.


B.  Potential Conflicts of Interest Due to Principal-Agent Relationships


This concept is related to transparency as described above, but generally reflects the fact that many health care decisions, such as referrals to specialists or the prescription of a drug, are made by health care organizations or providers that also benefit from these transactions under historical methods of payment in health care.  As a result, it can be difficult for the consumer (e.g., the patient) to question the recommendations of health care professionals or to know if what is being suggested is in the best interest of the patient.  Examples of similar agency concerns in other markets include financial planners and auto mechanics, where consumers (the “principals”) typically rely on these professionals (the “agents”) to recommend financial products or needed repairs, but in some cases, there is evidence that what is recommended may benefit the agent more than the principal.  Many analysts believe that conflicts of interest that currently exist in health care markets (e.g., provider ownership of imaging equipment from which providers benefit based on their own referrals) generate substantial inefficiencies and should be addressed.  


C.  Inefficiencies Due to Third Party Insurance


In health care most services are paid by third party insurers on behalf of patients, and, in traditional insurance contracts, patients pay less than the full price of services once any required deductibles are met.  As a result, for services patients consume, patients are typically responsible for just a small percentage of the overall cost of the service out of their own pockets (i.e., often only have to pay a $20 copay for a $100 office visit).  Research has shown that overconsumption of low value services can result, driving up overall health care expenditures.  Thus, many analysts believe that innovations in insurance design can help to promote efficiency by reducing spending on low-value services.


D.  Antitrust Concerns Due to Market Consolidation


Health care provider markets (which include the local availability of hospital and doctor/specialist care, for example) have become increasingly consolidated over the last two decades as hospitals have formed ‘health systems’ and health systems have acquired physician practices.  This movement has occurred, in part, to help give health care providers leverage in their dealings with health insurance plans that have also consolidated over the same time period.  One of the effects of consolidation in the provider and insurance markets, also referred to as bilateral monopoly, has been an increase in market power that can lead to higher prices (defined as prices significantly in excess of the marginal costs of production) when insufficient competition exists.  While the degree to which market consolidation is a problem varies in markets across the country, and even in markets within a state or region, many analysts believe that this can be a significant problem.  In Western Pennsylvania for example, Highmark and the University of Pittsburgh Medical Center (UPMC) have been embroiled in a controversy that would make UPMC doctors and hospitals not available to Highmark insurance beneficiaries.  Legislation has been offered in the state house and Senate that would require providers associated with integrated healthcare delivery systems to contract with any willing insurer[20].


E.  Production Inefficiency


Just as any other product or service is created, health care services are produced by combining different ‘inputs’ to produce ‘outputs’.  For example, in the production of restaurant meals, labor, food ingredients, overhead for the building, etc., are combined to create meals that are produced and sold.  In health care, labor, equipment, diagnostic tests, liability insurance, office and hospital space, etc. are combined to produce health care services. In competitive markets, product and service prices, and operating margin, are significantly related to how efficiently the products and service can be produced.  In health care, there is evidence that care is produced less efficiently than it could be, with a recent Institute of Medicine (IOM) panel estimating that 30% of total health expenditures are due to waste and inefficiency; $765 billion in total including $210 billion in unnecessary services; $130 billion in inefficiently delivered services; $190 billion in excessive administrative costs; $105 billion in excessively high prices; $55 billion in missed opportunities for disease prevention; and $75 billion in fraud.[21]  Thus, many efforts to create value in health care, by providers and payers alike, are focused on increased production efficiency, meaning the development of more innovative ways to combine inputs for the purpose of producing outputs, which can be defined as improvements in the management and the health of populations.


F.  Sub-Optimal Patient Behaviors Due to Moral Hazard


While much of the focus is placed on health care providers, analysts also believe that patients and consumers play an important role in achieving value in health care.  For example, patient compliance with the recommendations of health providers (e.g., taking prescribed medications or seeking recommended preventive care, such as flu shots) is believed to be important for prevention of avoidable complications.  In addition, engaging in healthy behaviors, such as exercise, healthy eating, and seatbelt use are believed to be important in promoting long term health and avoiding illness and injury and the costs of treating those events.  Health economists and others have identified the concept of “moral hazard” to explain what may appear to some to be sub-optimal behavior on the part of health consumers.  In economics moral hazard is a situation where an individual is willing to take risks because he or she will not fully bear the costs associated with risky behavior.  In health care this manifests through the presence of third party insurance where, for example, the costs associated with treating health problems (e.g., obesity) that may be due to risky individual behavior (e.g., poor diet and lack of exercise) are borne mostly by the other individuals belonging to an insurance pool. 


G.  Public Failures and Externalities


It should also be noted that some of the problems in health care and in fact some of the market failures discussed above may be due to ‘public failures’, which may be difficult for employers to resolve on their own.  For example, licensure of health care professionals is a form of government regulation that is an attempt to ensure the quality of those delivering health care services.  While such regulation certainly has positive attributes, it can also serve to limit the amount of provider competition that can lead to the market failures (e.g., market concentration) described above.  In addition, these same regulations can impose significant additional requirements on health care providers that increase costs of care, with limited evidence of improvements in care.  Similarly, lobbying by professional societies on behalf of their membership can result in regulations that prevent other health care professionals from practicing in clinical areas that others deem them qualified to practice in, or that licensure in some states allow them to practice in.  For example, nurse practitioners and physician assistants have been challenged in some states by licensure restrictions that forbid them to independently deliver primary care services or that allow them to do so only under supervision of a licensed physician.  While such restrictions can be controversial and there are reasonable arguments on both sides of this issue, the resulting government regulations can sometimes create market inefficiencies and serve to exacerbate certain problems, such as timely access to appropriate primary care services.  Likewise, government policies can lead to externalities, which are defined by economists as costs or benefits incurred by one party as the result of the behavior of another party.  A recent example of an externality due to the passage of the Affordable Care Act is the provision of the law that allows children to remain on their parents’ insurance policies to age twenty-six years of age.  This provision impacts, for example Penn State’s student graduate insurance policy because some students might now opt out of selecting Penn State coverage to be covered on a parental policy.  The result is a changing risk pool for the university’s graduate insurance policy that can have cost implications for both the university and graduate students.


III.  Specific Goals Related to Improving Value for the University and its Beneficiaries


As described above, various alternatives for improving the value of health benefits for the Penn State community can be assessed based on the degree to which they address the market failures described previously.  But in addition to addressing these market failures, which can seem abstract at times, it is also important to assess the alternatives against their ability to achieve specific goals important to the Penn State population that are consistent with improving value.  In this section we discuss several practical goals that will promote value for the Penn State community so that alternatives can be evaluated based on their ability to achieve these specific goals.


A.  Appropriate Utilization of Health Care Services


The consumption of health care services are valuable in so far as these services prevent illness and disease or help to restore, maintain, or improve one’s existing health status.  Given that a key component of the value equation is total expenditures, value can be improved by ensuring that money is being spent on utilization of health care services that maximize the potential to improve, restore, or maintain health or to prevent future illness.  Thus, health benefits expenditures should support scientifically efficacious and appropriate health care services, or appropriate services which include all needed services based on the scientific evidence currently available.  Obviously, consumption of appropriate health and medical care services will increase overall spending.[22]  Consumption of ineffective or low value services, which has been documented to exist in the United States, is not valuable because these services spend money without adding value to health and, in many cases, this consumption exposes patients to unnecessary risks (for example for side effects, such as infections after an unnecessary surgery).[23]  Several of the market failures identified above can exacerbate the consumption of inappropriate utilization.  For example, insufficient evidence about the price and quality of services, third party insurance and moral hazard, and overconsumption as a result of conflicts of interest due to principal-agency relationships can serve to promote overconsumption or unnecessary consumption of health and medical care services. 


B.  Receipt of High Quality Health Care


Health is no different than other product or service markets where one can purchase or receive the desired commodity or service, but the quality of that product or service can be poor or inferior.  For example, paying a contractor for a repair that is poorly done leads to the need for additional work and can often create additional problems and require additional expenditures to correct quality deficiencies.  There are many definitions and dimensions of quality of care in health, but many have gravitated towards the one used by the Institute of Medicine (IOM) in its Crossing the Quality Chasm report released in 2001  The definition of quality provided by the IOM is:


“The Institute of Medicine defines healthcare quality as the extent to which health services provided to individuals and patient populations improve desired health outcomes. The care should be based on the strongest clinical evidence and provided in a technically and culturally competent manner with good communication and shared decision making.”


To make this definition more concrete, the IOM further elucidated six ‘aims’ or dimensions of high quality health care.  These are as follows:[24]


·         SafeAvoiding preventable injuries, reducing medical errors

·         Effective Providing services based on scientific knowledge (clinical guidelines)

·         Family CenteredCare that is respectful and responsive to individuals and their families

·         Timely Reducing wait times, improving the practice flow

·         EfficientAvoiding wasting time and other resources

·         EqualConsistent care regardless of patient characteristics and demographics

Several of the market failures identified above can serve to perpetuate the existence of low or inferior quality health care. These include, for example, the lack of transparency of health care price and quality and conflicts of interest stemming from principal-agent relationships.


C.  Fair Pricing


In its simplest form, the level of overall health care expenditures for the University and beneficiaries is determined by quantities of health care utilized (described in A above) and the prices charged for this consumption.  Thus, another important dimension for maximizing value is to obtain the best pricing possible for health and medical care services consumed.  Several of the market failures described above, including price and quality transparency, third party insurance, and market consolidation, have served to result in prices that many analysts believe are not as competitive as they should be.


D.  Timely Access to Health Care Providers


High quality health care requires access to appropriate services in a timely fashion.  Access to care is determined by many factors including the availability of a sufficient number of providers in a region to care for the population and the financial (e.g., insurance and income) and non-financial (e.g., transportation) means for patients to seek care from those providers.  Access is also related to making sure that patients receive care in the appropriate setting and avoid care in less than optimal settings.  For example, a significant focus has been placed on reducing the provision of primary care in emergency departments (ED) which is not only expensive, is generally not well coordinated, and can contribute to the problem of ED overcrowding.  Several market failures described above can impact access.  For example, market consolidation or competition that becomes anti-competitive can serve to limit access; this is the concern in Western Pennsylvania given the relationship between the the UPMC health system and Highmark.  Regarding the availability of a sufficient number of providers, this is also a concern in certain regions, for example rural areas, where the availability of primary care providers and specialists is of concern.  The development of the HMC clinics in State College, at the request of the University, was necessary to provide sufficient primary and specialty care.


E.  Optimal Benefit Design and Out-of-Pocket Expenditures for Beneficiaries


Benefit design refers to the specifics of health insurance contracts that dictate components such as deductibles, co-payments, co-insurance, provider network availability, benefit limitations (e.g., number of post-surgical physical therapy visits allowed) and other important dimensions.  As described above in the section on market failure, benefit design may serve to promote over-consumption of low benefit health care services since third party insurance has historically reduced the out-of-pocket prices paid by consumers for services consumed.  In addition, third party insurance and benefit design decisions can impact market consolidation and prices paid in health care markets in a variety of ways.  Making optimal decisions regarding benefit design is both important and difficult and often involves decisions that weigh protecting insured’s from the significant financial risk associated with illness, injury and disease on the one hand, and promoting utilization of valuable care (e.g., flu shots and other preventive care) on the other hand.  Benefit design can also be important for ensuring that care is received in the appropriate setting; for example ensuring that non-emergent primary care is received in office based clinics rather than in more costly emergency departments.  Since these decisions ultimately impact utilization and thus overall expenditures, choices related to benefit design are critically important to achieving high value.


IV.  A Framework for Evaluating Various Options to Improve Value


The discussion of market failure and specific goals for achieving greater value in health care provides a framework for evaluating various options available to Penn State and its beneficiaries.  Each option can be compared along the dimensions defined above, and as outlined in Table 1, to consider the degree to which it is likely that employing the option would yield greater value for the Penn State population.  While this framework can be valuable as a tool for comparing options, it is important to note that there is uncertainty around most options with some options having a wide degree of uncertainty.  Uncertainty exists because the scientific evidence for the options varies, and in many situations, the feasibility of implementing an option, and realizing a positive return on investment, depends on the particular context of the employer and the geographic regions where its employees reside (i.e., the local health care market characteristics in specific areas, such as availability of primary care services, specialists, etc.).  Thus, we suggest the framework be used as a starting point, to consider if and how each option might address the fundamental causes of market failure and whether employment of the option might help to increase value by addressing one of the specific goals important to Penn State and its beneficiaries.


Table 1: Important Market Failures and Specific Goals for Improving Value



Specific Market Failures in Health and Medical Care

Specific Goals for Improving Value

Ensuring Appropriate Utilization of Health and Medical Care Services

Receipt of High Quality Health and Medical Care

Fair Pricing for Health and Medical Care Services Consumed

Timely Access to Health Care Providers and Services

Optimal Benefit Design and Financial Risk

Lack of Quality & Price Transparency






Conflicts of Interest Due to Principal-Agent Relationships






Inefficiencies of 3rd Party Insurance






Antitrust Concerns Due to Market Consolidation






Production Inefficiency






Suboptimal Consumer Behavior Due to Moral Hazard







In the next section, we discuss specific options that Penn State might consider and that have been identified by the task force based on research, in consultation with internal Penn State experts, and in consultation with Towers-Watson, a benefits consulting firm that has been retained by the University to assist with long range planning.  The options are organized in categories, though it is possible that options can fit multiple categories.  It is also important to be clear that in presenting these options, neither the sub-committee nor the task force is making a recommendation that the option be implemented.  The task force views its role as identifying options and discussing their potential benefits and risks, to the extent we are able given existing evidence.  Ultimately decisions regarding which options to employ, if any, are a decision for university administrators in consultation with the faculty and staff.  Thus, in the section that follows, we discuss each option and how it might address specific market failures and accomplish specific goals related to improved value.  In the discussion of each option we identify the expected benefits and risks if the option was employed and we discuss the required financial investments for implementing the option as well as the timeframe required for implementation.  We also identify additional data analyses that might be helpful for reducing the uncertainty associated with the option.  Table 2 illustrates this framework.


Table 2: Framework to Evaluate Options for Improving Value







Option for Improving Value

Attributes of Options for Improving Value for Penn State’s Health Expenditures

Description of Option

Market Failures Addressed by the Option

Specific Goals for Achieving Value Addressed by the Option

Required Investment for Implemen-ting the Option

Potential Risks of Implemen-ting the Option

Timeframe for Implemen-tation of the Option

Additional Analysis Useful for Understanding the Option



















































V.  Discussion of Different Options for Improving Value for Penn State’s Health Care Spending


This section discusses possible alternatives for improving the value received for the health care dollars spent by Penn State and its employees, retirees and beneficiaries.  These options were identified from several sources including the following: a) an assessment of what other employers and payers are doing; b) options suggested from members of the health care task force or those consulted by the task force for advice; c) options suggested by Towers Watson, a health benefits consultant engaged by Penn State and available to the task force for advice; d) options identified from the health policy and health benefits literature.  For clarity of presentation and for the benefit of the reader, particularly those readers less familiar with health care vernacular, the options are sorted into seven categories.  In the sections that follow, each option is described followed by a discussion of market failures and goals using the framework described above.  It should be noted that the options are not necessarily mutually exclusive and some options can complement existing benefit designs while other options might replace existing offerings.   The discussion below is meant to provide enough detail so that decision makers can engage in productive dialogue about the relative merits of employing various options.  Because there is uncertainty associated with each option and the degree of uncertainty varies across options,  the discussion varies in depth based on the complexity of the option itself as well as the degree to which there is an established evidence base for the option.


A.  Improve the Data and Analytic Infrastructure to Effectively Manage Penn State’s $225,000,000 Annual Health Spend


1.  Establishment of a Comprehensive Data Warehouse


Through the course of its work it has become clear to the task force and the value sub-committee that Penn State lacks a comprehensive data and analysis infrastructure to adequately manage its more than $200 million dollar annual health care spend, relying instead on its contracted vendors for analysis and advice.  Other large self-insured employers have invested in building a comprehensive data warehouse and the analytic capacity to query the warehouse for purposes of informing key questions related to understanding the value received for dollars expended.  For example, the specific value of many of the options described below can be informed by an analysis of the University’s own data.  As an example the option of ‘reference pricing’ below seeks to address the problem of significant price variation for relatively standardized or homogenous health care services, such as x-ray, MRI or other imaging services.  In the case where there is significant price variation, with some providers of these imaging services charging higher prices, reference pricing is a tool for helping prices to converge to more competitive levels.  But the feasibility and return to implementing reference pricing depends on the degree to which there is actionable price variation among providers in specific markets and sufficient options of providers in these same markets.  Querying an existing data warehouse that contains longitudinal claims data can help to determine whether reference pricing is feasible and if so, which clinical areas provide the best options. 


Another example is using claims data to monitor appropriate utilization of services in the insured population over time or in comparison to external benchmarks.  Health plans often measure receipt of recommended clinical preventive services (e.g., flu shots, cancer screening tests) by the eligible population as a means of assessing quality of primary care.  Measures of receipt of clinical preventive services can be compared with benchmarks, such as those provided by the National Committee for Quality Assurance (NCQA) HEDIS measures.  A comprehensive data warehouse assembles longitudinal claims data for medical care services and prescription drugs consumed for beneficiaries in the relevant employer and retiree populations. 


The data comes from the contracted TPA on the medical side and the pharmacy benefits manager (PBM) on the pharmaceutical side.  The warehouse is assembled by a contracted vendor who meets the federal government’s privacy and confidentiality requirements to receive personal health information (PHI) as a ‘covered entity’ under the Health Insurance Portability and Accountability Act (HIPAA).  The contracted vendor develops business associates agreements with the TPA and PBM and receives the information on paid claims.  The vendor de-identifies these claims to remove information that can identify specific individuals, creating a file that can be useful for analysis while still protecting individuals’ personal health information (PHI). 


The data in the warehouse is only useful if the self-insured employer has identified a series of analyses and questions that can be answered from the data, and if there is analytic capacity to write programmatic code to query the data warehouse to answer the questions that have been identified.  Typically this requires analyst capacity trained in working with large claims data sets and familiarity with health care reimbursement and medical and pharmaceutical claims coding.  This analytic expertise can be housed within the self-insured employer or by the vendor or both.  Regardless, the warehouse is only valuable if the data can be analyzed in a systematic way to form decisions regarding contracted vendors, prices paid to providers, quality of care received by beneficiaries and other important questions.  Having provided this general description of a data warehouse, we now describe details of important attributes identified in our framework.


·            Primary Market Failures Addressed – Lack of quality and price transparency; Production inefficiency;

·            Primary Goals Addressed – Appropriate utilization; Receipt of high quality care; Fair pricing; Optimal benefit design

·            Required Investment for Implementing the Option - Uncertain, although discussions with Towers Watson indicate that a low-end estimate would be about $100,000.

·            Timeframe for Implementation – About one year to develop a fully functional warehouse.  This includes time required to seek proposals from vendors, finalize contract with selected vendor, sign data use and business associates agreements with TPA and PBM and have claims data transferred to the contracted vendor.

·            Potential Risks – Concerns on the part of employees and beneficiaries regarding the privacy of information about their medical and pharmacy utilization.  These risks can be mitigated by following existing examples and contracting with credible third party vendors that are in clear compliance with the HIPAA law.

·            Financial Return – Estimates would need to be received from other employers that have instituted a data warehouse.  Such savings are difficult to compute as the primary value of the warehouse is in conducting analysis that can then be used to evaluate all facets of health benefits expenditures and to inform better contracting and benefit design in the future.  The value of the warehouse for research purposes could be monetized as well.

·            Additional Analyses Required – More exploration of how other self-insured employers have successfully built a data warehouse and details regarding the monetized value of this investment.  


B.  Better Leverage Penn State’s Purchasing Power by Assessing the Efficiency of Contracted Vendors and Comparing the Services They Provide to Available Alternatives


1.  Assess Value for Current Provision of Retiree Benefit Options


Penn State currently provides retirement benefits to 1326 beneficiaries that are pre-65 and not Medicare eligible and to 4698 Medicare eligible retirees.  Benefits are provided to former PSU employees and their eligible beneficiaries.  Pre-65 and non-Medicare eligible remain in the same plan as active employees.  For the Medicare eligible retirees, Penn State offers a group health insurance plan that is fully insured by Highmark Blue Shield.  Because this group plan covers only Penn State’s beneficiaries, the risk pool is smaller than other available risk pools available to employers in the open market.  For example, one option is the retiree exchange market that has been something many employers have moved to recently.  A retiree exchange is a private option where the employer makes contributions to eligible retirees to be used to select a plan from a menu of available options on the open Medicare Advantage marketplace.  The Medicare Advantage marketplace is a well established and robust market place with many options provided by reputable health insurance plans that meet the federal government’s requirements for participating in Medicare.  


This exchange marketplace, not to be confused with the health insurance exchanges that have been created by the Affordable Care Act (ACA) and that have been in the news recently, provides beneficiaries with a variety of plan options that vary along dimensions such as the breadth of the provider network, differences in deductibles and co-payments, and other important dimensions.  Plan premiums are based on market competition and prices are inclusive of the amount contributed by the federal Medicare program towards coverage that the beneficiary is entitled to because of achieving Medicare eligible status.  Retirees that select a plan with a premium that is less than the employer contribution can put the difference into a health retirement account available to pay for future health related expenses.  Estimates have shown that by providing a fixed dollar benefit to retirees and having them shop on the exchanges results in benefits for both the employer and the retiree.  According to estimates provided by Towers Watson, retirees could typically save about $2,000 in out-of-pocket expenses annually since the exchange based plans are able to offer better terms than individual self-insured employer sponsored plans.  In addition, employers are typically able to reduce their annual expenditures and aggregate post-retirement benefit liability due to the fact that they are no longer providing group coverage to insure the risks of retirees.


In addition to the retiree exchange option, there are other options as well that employers have pursued to better control and predict their expenditures on retiree health expenses.  For example, as part of its task force work, the benchmarking sub-committee identified changes that are being implemented for retiree benefits at the University of Michigan as part of its Committee on Retiree Health Benefits (CORHB).  Details can be found at, but the changes include reductions in university subsidies for retirees (to 80% of total costs) and their dependents (50% of total costs) and changes in subsidies for newly hired employees.  Unlike the retiree exchange option described above, this option maintains the University’s group sponsored retirement plan, including post retirement benefit liability, but changes the percentage and therefore overall amount that the university contributes for retirement coverage. 


·            Primary Market Failures Addressed – Production inefficiency;

·            Primary Goals Addressed – Fair pricing; Optimal benefit design and out-of-pocket expenditures for beneficiaries;

·         Required Investment for Implementing the Option – Primarily benefits office staff time to design and implement changes.  Nominal administrative fees paid to exchange administrator to manage retiree spending accounts;

·         Timeframe for Implementation – Exchange option can take about one year to implement.  Other options, such as the change in subsidy approach implemented by the University of Michigan, may require a phase in over multiple years.

·         Potential Risks – Primary concern is how proposed changes are perceived by retirees.  While some argue that certain approaches, such as the move to retiree exchanges can actually make retirees and their beneficiaries better off, communication is important as there may be a perceived ‘take away’ associated with any changes in the way the university’s retiree health benefit is administered. 

·         Financial Return – Tower’s Watson estimates that Penn State can save about $2 million in cash annually while over 90% of retirees would have equal or better benefits and still save about $2,000 per year in out-of-pocket benefits.  In addition, the aggregate post-retirement benefit liability could be significantly reduced.

·         Additional Analyses Required – A more in depth understanding of the various options, including an assessment of how other universities have designed and implemented similar changes.  Since retiree benefits are administered as part of the existing TPA contract, verify the ability to eliminate the group plan in place of moving to an exchange option. 


2.  Leverage CIC Collective Partnership


The Big Ten Conference’s Committee on Institutional Cooperation is a consortium of Big Ten universities and the University of Chicago (and now including the newest members – the University of Maryland and Rutgers) that agree to work together on mutually beneficial projects and share information.  For example, the CIC has a purchasing consortium that is described as follows:

In 1998, CIC formed what has become a remarkably effective Purchasing Consortium for commodities and services. By joining forces, purchasing directors from member universities are able to negotiate better terms and gain significant price advantages on products with relatively high margins.  To date, CIC has saved more than $19 million. [Source:]. 

Given that all of the CIC institutions are likely experiencing similar challenges with the provision of health care benefits for employees and retirees, there may be value gained by exploring a partnership among CIC institutions or a subset of institutions that might be interested in such a partnership.  While the exact nature of partnership would need to be discussed, this could include anything from basic information sharing and benchmarking to the possibility of group negotiation of TPA services.  For example, Pharmacy Benefits Management (PBM) services is one significant area of health care expenditures where it seems possible to reap the advantages of the CIC’s collective purchasing power.


·         Primary Market Failures Addressed – Lack of quality and price transparency; Antitrust concerns due to market consolidation;

·         Primary Goals Addressed – Fair pricing;

·         Required Investment for Implementing the Option – The CIC infrastructure exists.  An investment in leadership required to gauge interest of other consortium members in pursuing a partnership focused on health care benefits.

·         Timeframe for Implementation – Depends on the specific activities the consortium might wish to pursue.

·         Potential Risks – Potential concerns among CIC members when it comes to sharing information on benefits offered to employees.  Risk is likely small however given other topics where CIC members have agreed to collaborate.

·         Financial Return – Uncertain.

·         Additional Analyses Required – Contacting CIC leadership to gauge the interest of consortium members.

3.  Identify Health Care Providers That Serve Significant Portions of the PSU Population and Engage in Discussions about Preferred Arrangements


While the task force was not able to examine specific data it is believed that, in many of the markets where Penn State beneficiaries receive care, there are specific providers and provider organizations that provide care to a large percentage of eligible Penn State beneficiaries, or that would be capable of providing care to most Penn State beneficiaries.  One option then, recognizing the size of Penn State’s population in its major geographies, is to develop close relationships with specific providers on both pricing as well as quality of care.  Similar to the concept of preferred provider relationships, the University can explore providers’ willingness to provide favorable contract terms for Penn State’s eligible beneficiaries.  This might take many forms, including more favorable pricing, jointly developed care management and quality programs, or centers of excellence for more complex treatments and procedures.  For example, primary care providers’ practices may employ reminder systems to notify patients by mail when they are due to receive recommended clinical preventive services appropriate for their age/sex group.  Penn State beneficiaries might have lower amounts of required contributions to use these providers relative to others.  Providers that are not part of the partnership can still be used but will require greater levels of cost sharing by beneficiaries.  Partnerships can be broad (e.g., discounts from current prices for all care provided by the partner for Penn State beneficiaries) or narrow (e.g., center of excellence for heart care or cancer care).


·         Primary Market Failures Addressed – Lack of price and quality transparency; Production inefficiency;

·         Primary Goals Addressed – Appropriate utilization; Receipt of high quality health care; Fair pricing; Timely access to Health care providers;

·         Required Investment for Implementing the Option – Time on the part of benefits staff and their agents to identify partners willing to enter into strategic partnerships and to develop those partnerships and translate them into benefit design changes that can be communicated to Penn State beneficiaries.  Significant expertise in contract negotiation will be required to implement this option.

·         Timeframe for Implementation – Varies by specific partnership.

·         Potential Risks – Concern on the part of beneficiaries that choice of provider might be limited.  Communication is important and would need to convey to beneficiaries the benefits if care was received by a partnering provider.

·         Financial Return – Uncertain and depends on the specific details of potential partnerships.


4.  Assess Value Received from Current Pharmacy Benefits Manager (PBM)


Penn State’s current Third Party Administrator (TPA) contract with Highmark Blue Shield includes a PBM contract with Express Scripts.  A PBM provides many functions including negotiating rebates with pharmaceutical manufactures and negotiating dispensing discounts with retail pharmacies.  In addition, PBMs provide claims adjudication services, benefit design advice and implementation, and pharmaceutical prescription management services, such as monitoring for polypharmacy or medication errors related to improper dosing, contraindications, and other problems.[25]  Like the provider and health insurance marketplaces, there has been significant consolidation in the PBM marketplace with just a few PBMs accounting for a significant portion of the market share in the United States.  Since PBMs provide a relatively homogenous service, savings may be possible if the PBM contract was “carved out” and bid separately from the current TPA contract.  For example, Towers Watson estimates that savings could amount to 0.5% of annual expenditures, or a total dollar savings of about $1 million dollars, though this number could be more or less pending a formal review of the current PBM contract and the opportunity to solicit competitive bids.


·         Primary Market Failures Addressed – Lack of price and quality transparency; Antitrust concerns due to market consolidation;

·         Primary Goals Addressed – Fair pricing; Optimal benefit design and out-of-pocket expenditures for beneficiaries

·         Required Investment for Implementing the Option – An assessment of the value received from the existing PBM contract within the existing TPA contract.  Implementation of a competitive bid process to gauge opportunities for additional value.

·         Timeframe for Implementation – About 9-12 months would be necessary if a change were to be made for implementation in the following year.

·         Potential Risks – Existing PBM contract is included as part of the current TPA contract.  The possibility of separating (i.e., ‘carving out’) the PBM contract would need to be reviewed.

·         Financial Return – Towers Watson estimates the possibility of about $1 million in annual savings or 0.05% of annual health care expenditures.


5.  Assess Market Competitiveness of Current Administrative Services Fee Paid to Current Third Party Plan Administrator


Effective January 1, 2008, Penn State contracted with Highmark Blue Shield to become the exclusive third party administrator (TPA) of Penn State’s self-insured health benefits plans for employees of the University, the Hershey Medical Center, and all retirees.  The contract is currently scheduled to run through December 31, 2017, with approximately 3.5 years remaining.  As part of the contract a TPA administrative services fee was negotiated and was to be paid by Penn State on a per employee contract basis.  This TPA fee covers services provided by the TPA including plan enrollment, claims adjudication, provision of a network of providers with contracted discounts, etc.  Since such a long term contract is unprecedented in employer sponsored health benefits programs, the question is whether existing contract terms deviate from current fair market rates for administrative services only fees?  The only way to know this is to develop an RFP and to competitively bid for TPA services, but consultants to Penn State have estimated that current rates may be higher than market rates by anywhere from $1.50-$3.00 per employee per month, or a total of $306,000-$712,000 annually.


·         Primary Market Failures Addressed – Lack of price and quality transparency;

·         Primary Goals Addressed – Fair pricing;

·         Required Investment for Implementing the Option – An assessment of the value received from the existing TPA contract.  Implementation of a competitive bid process to gauge opportunities for additional value.

·         Timeframe for Implementation – About 3 months would be necessary to secure competitive bids.  If a change in TPA was sought, several additional months would be required for implementation in the following year.

·         Potential Risks – Existing TPA contract is in effect until December 31, 2017 and it is not clear whether it is possible for Penn State to renegotiate at this time.

·         Financial Return – Towers Watson estimates the possibility of about $306,000-$712,000 in annual savings, or 0.15%-0.3% of annual expenditures.


C.  Changes in the Way Penn State and Its TPA Pays for Health Care Services


1.  Implement Reference Based Pricing Strategy for Relevant Services


As described above in the discussion of the data warehouse option, reference pricing is an option that seeks to gain convergence to a fair price for specific clinical services used, such as diagnostic tests, durable medical equipment, or laboratory tests.  Some existing self-insured employers have implemented reference pricing by working with their TPA or by contracting with vendors that have experience in this area.  The process of implementing reference pricing begins with an analysis of recent claims data, which can be easily queried if the employer has an existing data warehouse in place.  If a comprehensive data warehouse does not exist, then the employer can work with vendors to conduct necessary analysis in conjunction with the existing TPA.  Regardless, the analysis will examine claims to understand if there is significant variation in prices paid for the same services within a relevant geographic region where patients seek care.  For example, Penn State might examine prices paid for outpatient MRI’s, examining such variation in markets with concentrations of Penn State beneficiaries such as in State College, Philadelphia, Harrisburg, Erie, and Pittsburgh.  If significant variation does exist, and if there is enough existing provider capacity in the market, then this analysis can inform several different strategies that are designed to facilitate prices converging to a lower average.  Strategies that have been used by employers include setting an upper limit to the amount paid for the service in the benefit design, to directly contracting with providers and developing preferred provider networks where providers agree to the established reference price. 


·         Market Failures Addressed – Lack of quality and price transparency.

·         Primary Goals Addressed – Fair pricing for health and medical care services; Optimal benefit design and financial risk.

·         Required Investment for Implementing the Option – Relatively low if existing data warehouse or if done by a contracted vendor.  Primarily involves staff time to conduct analysis of claims data.  If analysis suggests this is a useful strategy, will need to work with TPA to modify the claims system and enrollment communications to implement the reference pricing benefit.

·         Timeframe for Implementing the Option – About one year to implement for the following benefit year.

·         Potential Risks – Employees and beneficiaries, and their health care providers may view this approach as limiting choice.  However, since reference pricing is implemented for relatively homogenous services, this should be less of a concern if the benefit change is well communicated.  Experience from other employers suggests that provider prices converge quickly, reducing this concern since all providers are still available to the beneficiary.

·         Financial Return – Towers Watson estimates the savings to Penn State to be an annual net reduction of between $1-$4 per employee per month, or $12-$48 per employee per year.  This results in an estimated savings of $204,000-$816,000, though these estimates include uncertainty because the specific details of the reference pricing program to be implemented matter.

·         Additional Analyses Required – More information about how this option has worked for other self-insured employers can be obtained before moving forward.  There are potential vendors that have experience implementing this type of program and they should be able to provide more details on this option.


2.  Work with Payers to Implement Bundled Pricing


Health care services in commercial self-insured populations have traditionally been paid as fee-for-service transactions where the TPA negotiates discounts on fees charged by health care providers and pays for services based on the fee schedule and the pre-arranged discount.  For some services, primarily hospital based services or procedural care, payment may be made for a ‘bundle’ of services associated with, for example, the inpatient fees related to a hospitalization (e.g., room and board, lab and diagnostic services, and other items), but typically these bundled payments do not include fees for physician consultations and they often don’t include care provided by other outpatient providers before and after the hospitalization.  Importantly, fees paid typically don’t cover any fees associated with “re-work” that might be required should the patient need to be re-hospitalized or if care is needed to correct a deficiency in the original care provided.[26]  New models of payment that have received attention in recent years include “bundled payment” models or “episode payment models” that determine a package price for a bundle of services or an episode of care that spans a time horizon that falls within the typical range of care for the clinical service or procedure in question, and which includes all payments including those for inpatient and outpatient care as well as payments for any complications or re-work that might be necessary. 


The theory behind bundled or episode based payment is that prices can be set at a rate that should cover all care required for a specific measurable clinical situation, thus incentivizing providers to be efficient in the delivery of care since the amount paid must cover all care needed.  In theory, this type of payment eliminates the inefficiencies associated with fee-for-service reimbursement where the incentive may be to provide more care that can be billed.  The efficacy of bundled payment approaches hinges on the ability to identify clinical conditions or episodes of care amenable to such payment and where there is enough certainty about the likely treatment required such that fair payments can be established a priori. 


There has been work done on this topic with an increased number of examples where bundled payment has been successfully implemented.  For example, in California, the Integrated Healthcare Association (IHA) has successfully implemented bundled payments for certain clinical conditions such as orthopedic and maternity care (see for more details).  The Center for Medicare and Medicaid has launched a comprehensive Bundled Payment Program that includes 48 distinctly defined episodes of care; based on the strong evidence of savings from bundled payment pilot programs.  Hershey Medical Center is currently participating in this program, which covers the period of inpatient hospitalization plus all required care over the next 90 days, including hospital, physician, skilled nursing, rehabilitation, and outpatient services.  In Pennsylvania, Geisinger Health Plan launched Proven Care in 2006, which is a program where for certain procedural care (e.g., treatment for open heart surgery), Geisinger will charge a bundled price which includes the costs of all care related to the procedure as well as a warranty that commits that any additional care required due to mistakes or errors, will not be billed to the payer or patient.  Geisinger states that it  is able to provide efficient bundled prices and the warranty because they invest heavily in studying the care delivered for the relevant Proven Care procedures, implement the best known scientific evidence, and work with patients and families to engage the patient in optimal care pre and post procedure. At this point Geisinger only offers the Proven Care payment option to members of its own Health Plan, the Geisinger HMO, though the Proven Care clinical pathways are followed for all patients regardless of their insurer.


    Primary Market Failures Addressed – Potential conflicts of interest due to principal agent relationships; Production inefficiency; 

    Primary Goals Addressed – Fair pricing; Appropriate utilization of health care services; Receipt of high quality health care;

    Required Investment for Implementing the Option – Identification of providers capable of offering a bundled price or willing to accept the risk associated with episode based payment.  Agree on contract terms and work with TPA to incorporate into benefit design.  Significant clinical expertise will be required to appropriately define the “bundle”.

    Timeframe for Implementation – Uncertain.  Depends on identification of willing providers.

    Potential Risks – Risk to providers for taking on some risk related to the care they provide.  Potential risks to Penn State if quality of care is not monitored.  Since not all providers are ready to engage in such payment, the overall availability of capable providers to the Penn State population may be limited.

    Financial Return – Uncertain.  A RAND study has estimated that the United States would save as much as 5% if health care expenditures were subject to bundled payment.


3.  Evaluate  Non-Payment Policy for Medical Errors


Health care providers have historically been paid for care even when the care provided results in a medical error or harm to the patient.  This is counterintuitive and atypical from what is observed in other product and service markets.  Recently the federal government, as part of its Medicare program, instituted a policy where Medicare will not pay for ‘never events’, and the Centers for Medicare and Medicaid Services (CMS) also encouraged state Medicaid programs to adopt similar policies and encouraged states to do the same in the other health benefits programs they administer.  The impetus for this policy change was a study by the Centers for Disease Control (CDC) estimating that “common medical errors total more than $4.5 billion in additional health spending per year.”  These so called ‘never events’ can include hospital acquired infections, preventable falls and trauma, stage III and IV pressure ulcers, and other situations, such as when a foreign object (e.g., surgical instrument) is left inside the patient’s body cavity after the surgery.  Starting in October of 2008, CMS stopped paying for services when these errors existed, as well as the associated costs of treatment to remedy these errors.[27]  Penn State’s TPA implements similar policies but this should be reviewed and updated.


·         Primary Market Failures Addressed –Lack of quality and price transparency; Production inefficiency; 

·         Primary Goals Addressed – Fair pricing; Receipt of high quality health care;

    Required Investment for Implementing the Option – Working with TPA to implement payment policy in contracts with network providers.

    Timeframe for Implementation – Within one year pending TPA’s existing contract with providers.

    Potential Risks – Potential for some network providers to not agree to these payment terms.

    Financial Return – Uncertain, though some degree of savings is likely.  A historical claims analysis and literature review should yield better estimates.



D.  Options That Increase Access to Urgent and Primary Care Services and Valuable Care Management Models for Beneficiaries


1.  Onsite Health Clinics and Expanded Primary Care Access at Campus Locations


A strategy that is becoming increasingly common in employed populations of as many as 500 employees at a specific location is access to an onsite health care clinic, typically run by a contracted health care provider.  Since Penn State runs its own health system and has a College of Nursing, onsite clinics could be run in affiliation with these internal partners.  Onsite clinics can offer a range of services, depending on the size and scope, including same day urgent care appointments, preventative examinations, wellness visits and counseling, and other services.  In some employed populations these clinics have been received favorably by employers and employees because of their convenience to the workplace.  In addition, since the Affordable Care Act requires health plans to provide specific clinical preventive services without charge to patients, onsite clinics can also serve to meet the likely increased demand for preventive services by increasing access, enhancing convenience and reducing the patient load of community providers. 


One caution is that linked medical records would be needed to track patients’ receipt of the preventive services to avoid “overuse” that could occur if patients received these services from both the clinics and their regular providers.  Onsite clinics are not meant to be a substitute for all health care needs and thus the care provided at these clinics should integrate and coordinate with care provided by the community providers typically seen by employees.  Towers Watson, a consultant working with Penn State estimates that onsite clinics can result in a savings of 0.5-1.5% per year ($1-$3 per member per year) or as much as $4 million annually for Penn State.  These savings are realized from utilization of employed providers rather than contracted network providers, lifestyle risk reversal and disease control and prevention. 


Required investments include the acquisition and construction of a facility, and the costs of training and employing staff, typically through a vendor contract.  The utility of this option for Penn State should consider several facts.  First, Penn State is a university with multiple campus locations, each of different size.  Thus, this option might make sense in some locations but not in others based on number of employees and the local health care provider market capacity.  Second, Penn State has already been subsidizing the establishment of Hershey Medicine primary care clinics in State College since 2010.  These investments were meant to address a local primary care shortage as well as a shortage in other medical specialties and the total subsidy for starting and operating these clinics has been more than $15 million since 2010.  While these clinics are not located on the Penn State campus and do not exclusively serve Penn State employees, their creation was an attempt to address similar needs to those typically addressed by onsite health clinics, though with some differences.  Thus, it is not clear that an investment in a separate on campus clinic would add value, or as much value, given the availability of these newly established clinics.


·         Primary Market Failures Addressed –Production inefficiency. 

·         Primary Goals Addressed –Appropriate utilization of health care services; Receipt of high quality health care; Timely access to health care providers.

    Required Investment for Implementing the Option – Depends on the scale and scope of the clinics to be implemented.  Penn State has already invested $15 million since 2010 in the Hershey Medical Group for the establishment of physician practices in State College near the University Park campus.

    Timeframe for Implementation – Depends on scope but could likely be established quickly.

    Potential Risks – Uncertainty about how much demand there would be for these clinics among the employed population.  Location and access given the size of the University Park campus could influence demand.

·         Financial Return – Towers Watson estimates 0.5-1.5% savings (about $4 million) annually though more precise estimates would be warranted given the specific circumstances at Penn State.


2.  Evaluate Value of Current Highmark Disease Management Program Against Alternatives Including a Complex Case Management Program, a Gaps in Care Program, and a Second Opinion Service


As part of the fee Penn State currently pays to Highmark to administer the University’s self-insured health benefits program, Penn State employees, retirees and their eligible beneficiaries receive access to Highmark’s sponsored disease, care and case management programs (DM/CM) through their Blues on Call program.  As far as the task force is able to discern, the specific costs to the University of these programs relative to other services provided under the TPA’s administrative services contract are uncertain, though some estimates provided by other employers and their consultants suggest that these services can cost as much as $3-$9 per employee per month or $36 to $72 annually.  Given the size of Penn State’s population, this could total as much as $544,000 to $1,088,000 annually.  Lacking specific pricing details it is difficult to know how much could be saved by the University if it were to eliminate these services from the contract and whether better value could be received by contracting independently for care management services with another provider(s).  To answer this question specific pricing detail would need to be acquired from Highmark and compared to competitors available in the open market while simultaneously comparing the efficacy of the care management programs offered.  In addition, DM/CM programs have various degrees of aggressiveness ranging from ‘gaps in care’ programs that are the least aggressive, involving mailings to members and their physicians about needed care interventions (e.g., cholesterol tests) to more aggressive programs such as those that offer a dedicated staff to help provide counseling and support to plan beneficiaries (e.g., assistance after discharge from a hospital).  Penn State’s DM/CM programs seem to be in the mid-range of aggressiveness though the effectiveness of these programs is important and is not well understood.  Towers Watson reports that many employers are moving to more aggressive programs that seem to yield higher net savings despite costing more.  Common care management programs that some employers offer for their insured populations include the following:


(a)  Complex Case and Care ManagementThis involves telephonic or in-person management of complex cases with a high risk of incurring substantial medical costs.  Specially trained complex case managers can provide help to patients with health system navigation and coordinating and communicating across providers and finding access to pertinent clinical information.  Managers can also link patients to available resources covered under existing employee benefits programs, such as benefits for behavioral health services.  The trigger to reach out to patients depends on the aggressiveness of the program.  Some of the aggressive programs try to engage about 20% of plan members but succeed in actually engaging 6-8%.  It is worth noting that many insurance companies, including Highmark, are moving their care coordinators into the physicians’ practice, given the limited impact that telephonic care management has had.



(b)  Gaps in Care Program A gaps in care program is often employed as a lower cost alternative to disease management programs.  The program typically involves mailings to members and their physicians about needed care interventions (e.g., cholesterol tests).  These programs may cost as little as $4 per employee per month.  The literature suggests they do result in net savings, albeit a small number.


(c)  Second Opinion Service – As its name implies, a second opinion program can serve as a resource to patients who wish to determine the necessity of a recommended health care procedure or treatment.  Some employers require second opinions for procedures that have been shown to be over utilized and that result in unnecessary care that places patients at risk.  Examples of these procedures include hysterectomy, laminectomy, hip, knee and surgical treatment for carpal tunnel syndrome.  Towers Watson estimates that this could generate annual savings of  $800,000.


(d) Traditional Disease Management Program – A traditional disease management program typically involves a modeling based approach to segmenting the risk of an insured population using past health care claims information.  The focus is on patients with chronic disease, particularly those at high predicted risk of incurring significant health care costs.  Based on predictions of the model, individuals are contacted and provided with information and coaching and other services.  For example, one typical risk strata would include patients that have already been diagnosed with an existing chronic condition such as diabetes, hypertension or heart disease.  Specific outreach would be made by trained coaches and counselors, typically by phone, to encourage compliance with recommended treatments and behavioral modifications.  Another risk strata would include those identified by the model as ‘at elevated risk for developing chronic diseases’ such as those with pre-diabetes.  Similarly, specific outreach would be provided for this group by trained providers.


·         Primary Market Failures Addressed – Production inefficiency; Sub-Optimal patient behaviors due to moral hazard. 

·         Primary Goals Addressed –Appropriate utilization of health care services; Receipt of high quality health care; Timely access to health care providers.

    Required Investment for Implementing the Option – Depends on the specific programs to be implemented and estimates that would need to be secured from vendors providing these services.

    Timeframe for Implementation – Depends on scope but could likely be established for the following plan benefit year.

    Potential Risks – Receptivity of employees and beneficiaries to these programs.  Similar programs have been offered for years by Penn State’s TPA under the Highmark name.

·         Financial Return – PSU has three choices.  It can renegotiate its current program, with the goal of reducing costs by 0.1% or $200,000 per year.  PSU can choose to enhance its program, targeting high cost cases and individuals who may become high cost based on predictive modeling.  This would increase the program expense, but could decrease net overall medical cost by 0.6% or $1.2 milliion annually.  Alternatively, PSU could eliminate its current program and substitute a "gaps in care" approach.  There would still be net savings of $3 per employee per month, or 0.3% as compared to no program, but that savings would be more assured.  However, the current trend is to increase aggressiveness and intensity of programs rather than to decrease intensity.  Thus, Towers Watson estimates savings of $200,000 to $1,200,000 annually, depending on the approach selected.  Exact estimates would require more analysis and discussions with various vendors, including Highmark.

·         Additional Information Required – Details regarding current take up rate and effectiveness of existing care management programs would be helpful for assessing the value of moving in this direction.


E.  Explore Options to Provide Better Information and Tools for Beneficiaries to Navigate the Complex Health System


1.  Choosing Wisely


“Choosing Wisely” is an initiative of the American Board of Internal Medicine (ABIM) Foundation, designed to use professionalism and accountability on the part of physicians and their specialty societies, to address the problem of costly utilization of inappropriate health care services (see   To spark these conversations, leading specialty societies have created lists of “Things Physicians and Patients Should Question” — evidence-based recommendations that should be discussed to help make wise decisions about the most appropriate care based on a patients’ individual situation.”  Currently there are more than 50 specialty societies that have developed such lists, identifying medical treatments and procedures that are commonly prescribed but often are unnecessary or don’t add value for the patient.  In addition, Consumer Reports is working to disseminate this information to patients to increase awareness of the lists and the importance of discussing the contents of the lists with their physicians.  Specialty societies are also working with their membership to increase awareness and to identify ways in which providers can incorporate this information into their treatment decisions.”  In addition to addressing overuse, another opportunity is to encourage appropriate utilization by informing enrollees of their newly covered clinical preventive services benefits under the Affordable Care Act.


·         Primary Market Failures Addressed – Potential conflicts of interest due to principal agent relationships; Sub-Optimal Patient Behaviors due to Moral Hazard; 

·         Primary Goals Addressed –Appropriate utilization of health care services; Receipt of high quality health care;

    Required Investment for Implementing the Option – Little investment required to make beneficiaries aware of the Choosing Wisely campaign and its associated lists.  Could incorporate this information into a consumer transparency tool if one is selected.  Could work with TPA to engage network provides in the Choosing Wisely evidence.

    Timeframe for Implementation – Could be implemented immediately.

    Potential Risks – Communication is important to ensure trust and credibility by being clear that the Choosing Wisely recommendations come from credible and well respected medical specialty societies (e.g., the American Academy of Pediatrics) and not the employer or the insurance company.

    Financial Return – Uncertain.  No existing evidence to date though investment costs seem low.


2.  Provision of a Comprehensive Consumer Transparency Tool


As more and better information on provider quality and price becomes available to consumers, and as more beneficiaries face first dollar coverage in high deductible health plan arrangements, many plan sponsors, including employers, are making available a comprehensive consumer transparency tool.  These tools are offered by several vendors.  For example, Castlight Health (see is the tool that has been selected by another Big Ten institution, the University of Indiana.  These transparency tools can incorporate a number of features depending on the goals of the plan sponsor. For example, the web based tool can provide up to date information on where a patient stands with respect to deductibles and savings account balances in a high deductible health plan. The tool can also provide concrete information on pricing for medical and health care services offered by specific providers to help patients understand what they will need to pay out-of-pocket for specific services to be utilized.  In addition, the tool can provide access to available quality and safety information so that beneficiaries can compare the quality of services offered by different providers.  The tool can also be used to lead patients to important information about specific health care services and to incentivize patients, through the use of a ‘rewards program’ where beneficiaries can accumulate awards points that can be redeemed for benefits.  For example, patients that obtain an annual flu shot might earn rewards points similar to rewards earned in a frequent flier or frequent shopper program.  Changes in benefit designs can also be linked to the tool as well, such as the implementation of a reference pricing program as discussed above.  For an example of how Indiana University has used Castlight’s transparency tool, see (


·         Primary Market Failures Addressed – Lack of quality and price transparency; Antitrust concerns due to market consolidation;  

·         Primary Goals Addressed –Appropriate utilization of health care services; Receipt of high quality health care; Fair pricing; Optimal benefit design and out-of-pocket expenditures for beneficiaries.

    Required Investment for Implementing the Option – Would need to contract with a vendor that provides transparency tool services.  Specific pricing unknown at the time though vendors may be willing to price based on a realized shared savings agreement.

    Timeframe for Implementation – Likely 6-12 months for development prior to implementation.

    Potential Risks – Tool is only valuable if used by beneficiaries.  Would need communication strategy to drive beneficiaries to the tool.  Also, this approach would require the transfer of data from the TPA to the vendor and so security and privacy would be of utmost importance to faculty and staff.

    Financial Return – Uncertain.  No existing evidence to date.

·         Additional Analyses Required – Would be good to learn from Indiana University and other employers that have implemented a transparency tool.


F.  Explore Possibilities for Optimal Benefit Design of Insurance Offerings


1.  Increase Enrollment in Account Based Health Plan (High Deductible Plan Option)


During the 2014 open enrollment benefits period, Penn State offered a qualified high deductible plan with a health savings account (known as the PPO savings plan) for the first time as an option to the traditional PPO plan (known as PPO Blue).  The deductible amounts for the PPO savings plan are $1,250 for employee only coverage or $2,500 for family coverage with out-of-pocket maximum amounts set at $3,000 for employee only coverage and $6,000 for family coverage for in network care ($6,000 and $12,000 for out-of-network care).  Monthly employee co-premiums for the PPO savings plans are set at lower levels than the PPO Blue plan, and for those selecting the PPO savings plan option, the university seeds a health savings account with $400 for individual coverage and $800 for family coverage.  Employees can also contribute pre-tax dollars to the health savings account via payroll deduction.  Funds in the savings account accrue and can be rolled over and used to pay out-of-pocket fees for future health care expenses and stay with the employee even when he or she retires or leaves Penn State. 


As discussed in the academic literature section, the theory behind HDHP’s is that because consumers are faced with the full price of non-preventive care consumed, up to the amount of the applicable deductible, consumers will become more conscientious about care consumed, being careful to not utilize unnecessary or undervalued care.  Indeed, the evidence reported in that chapter suggests that HDHPs achieve savings in terms of health expenditures when compared to traditional PPO insurance arrangements.  As it planned for its 2014 health care budget, the Penn State benefits office assumed that 25% of the eligible population would enroll in the PPO savings option.  Instead, the actual number was about 16% of the eligible population.  Thus, the savings expected from offering this new option is likely to be less than predicted unless enrollment in the HDHP increases.  One option to be considered is driving additional enrollment in the PPO savings plan by using several available mechanisms, such as better and more focused communication about the advantages of HDHPs and incentives to drive enrollment in the PPO savings plan.  Incentives may include lower payroll deduction premium pricing for the PPO savings plan, additional seed money in the health savings account, or other possibilities.


·         Primary Market Failures Addressed – Inefficiencies due to third party insurance; Sub-optimal patient behaviors due to moral hazard;  

·         Primary Goals Addressed –Appropriate utilization of health care services; Optimal benefit design and out-of-pocket expenditures for beneficiaries.

    Required Investment for Implementing the Option – Minimal since PPO Savings Option has already been designed and offered.  Enhanced communication efforts and possibly increased financial incentives to drive enrollment in the HDHP would likely help.

    Timeframe for Implementation – Changes could be implemented for calendar year 2015 open enrollment.

    Potential Risks – Risks include lack of enrollment in the HDHP option because of lack of understanding of the mechanics of these plans.  In addition, since beneficiaries are exposed to significant first dollar coverage, there can be frustration on the part of beneficiaries if they do not have good information about prices.  A web based transparency tool and other options to promote price transparency, discussed above, can be helpful in addressing this potential risk.

    Financial Return – Experience from Towers Watson clients and evidence from a study by the American Academy of Actuaries suggests that net costs could decrease by 2-6%.  This would amount to annual savings of $4 million to $12 million dollars, though more analysis would be necessary to project potential savings for Penn State.  Employer funding of the Health Savings Accounts would offset some of these savings.

·         Additional Analyses Required – Given that Penn State implemented the PPO Savings plan for 2014 with about 16% enrollment, the claims and utilization experience of enrollees in this plan can be examined to better project the likely savings if enrollment was expanded.


2.  Introducing Narrow Provider Network Health Plan Options and Centers of Excellence


A growing number of employers and insurance products are instituting health plans that have ‘narrow networks’ or ‘Centers of Excellence’.  Narrow network plans are those that contract with a more limited set of health care providers (i.e., hospitals and physicians), that are typically included based on both quality and pricing data.  The theory is that these providers have been screened by the plan administrator as having met the criteria to provide the best value for plan beneficiaries.  Narrow network plans can be strict as in HMOs where the plan benefit does not provide coverage for any services received from out of network providers, or less strict such as in tiered network or preferred provider plans which provide less generous coverage for services received from non-preferred providers. 


The Center of Excellence concept is similar but pertains to preferred providers or a narrow network of providers in specific clinical areas such as treatment for heart surgery.  As an illustration, Lowes, the home improvement company, has contracted with the Cleveland Clinic as the Center of Excellence for cardiac surgery.  As part of its benefit design, Lowe will fly any employee or covered beneficiary to Cleveland for cardiac surgery if needed and will pay all travel costs for the patient and one other person.  Pepsi has instituted a similar program with Johns Hopkins in Baltimore for cardiac and complex joint surgeries and in addition to paying for travel costs will also waive required deductible and coinsurance for patients requiring these procedures (see  


·         Primary Market Failures Addressed – Lack of quality and price transparency. 

·         Primary Goals Addressed – Receipt of high quality health care; Optimal benefit design and out-of-pocket expenditures for beneficiaries.

    Required Investment for Implementing the Option – Working with TPA and other consultants to initiate conversations with a select set of providers.  First step is to examine data to identify candidate providers based on the quality of care delivered and prices charged.

    Timeframe for Implementation – Would depend on the nature of the program to be implemented and the time to perform due diligence.

    Potential Risks – Skepticism and lack of trust among employees about narrow network options.  This risk may be limited if providers with a high degree of brand recognition are selected. 

    Financial Return – Depends on the nature of the specific program to be implemented. 

·         Additional Analyses Required – A thorough assessment of similar types of programs offered by other employers and the rationale for those programs, as documented in scientific evidence, as well as important lessons from the rollout of these programs.


3.  Adding Value Based Insurance Design (VBID) Components to Health Plans


Value Based Insurance Design (VBID) programs challenge the traditional insurance benefit design that prices beneficiaries out of pocket costs (e.g., deductibles, copayments and co-insurance) the same regardless of the specific clinical service consumed.  For example, traditional insurance designs often charge the same amount for office visits regardless of the nature of the office visit.  Likewise, pharmaceutical copayments are often the same regardless of the drug therapy being utilized. VBID principles suggest that some treatments are so valuable that there should be no financial barriers to the beneficiary for consuming these services or treatments.  Thus, VBID designs seek to align patient cost sharing with the documented clinical value of the services and treatments consumed (  For example, VBID theory would suggest that anti-hypertensive drugs or cholesterol lowering drugs ought to be provided free of charge in the benefit design because they have great clinical benefit that they can improve health and avoid the costs of complications down the road for non-compliant patients.  Some have even argued that patients should be paid incentives to document they are adherent to drug regimens in these clinical areas.  The VBID literature also documents examples of situations in non-pharmacy areas that are also effective.  The ACA’s requirement that certain preventive services recommended by the United States Clinical Preventive Services Task Force (USCPSTF) be covered free of charge is another example of a VBID design that was written into federal legislation.  In its changes for 2014 benefits enrollment, Penn State implemented a VBID design for patients with specific chronic disease enrolled in the PPO Blue Plan.  Specifically this plan provides for 100% coverage for primary care and specialist visits as well as medical tests and supplies that are associated with the management of diabetes, high cholesterol, and hypertension.


·         Primary Market Failures Addressed – Sub-optimal behaviors due to moral hazard. 

·         Primary Goals Addressed – Receipt of high quality health care; Optimal benefit design and out-of-pocket expenditures for beneficiaries; Appropriate utilization of health care services.

    Required Investment for Implementing the Option – Minimal as PSU has already begun to offer a VBID design.

    Timeframe for Implementation – Expansion of the VBID benefit can happen for the next plan year.

    Potential Risks – Main risk is whether the return on investment will cover the increases in plan expenditures to Penn State given the increase in benefit generosity. 

    Financial Return – Depends on the specifics of the benefit changes implemented.  In the benchmarking survey one university estimated a 2.5% reduction in health care expenses due to its wellness program. 

·         Additional Analyses Required – A thorough assessment of similar types of programs offered by other employers and the rationale for those programs, as documented in scientific evidence, as well as important lessons from the rollout of these programs.  The University of Michigan’s VBID Center is the pre-eminent national resource on this topic and has assisted many employers and purchasers in VBID benefit design.


4.  Incentives or Penalties for Beneficiary Participation in Wellness Programs or for Engaging in Risky Behaviors


The Affordable Care Act created provisions to allow employers to use financial penalties and rewards for participation in so called ‘wellness programs’.  Penn State attempted such changes as part of its “Take Care of Your Health” initiative which was rolled out in the summer of 2013.  Specifically, Penn State originally imposed a financial penalty for employees, and in some cases eligible spouses, who did not participate in completion of a health risk appraisal (also known as a ‘wellness profile’), a biometric testing battery, and a commitment to have an annual health care exam with a physician.  In addition, a penalty or “surcharge” was added for beneficiaries who use tobacco as well as a surcharge for insurance coverage for spouses or same sex domestic partners (SSDPs) when these individuals were offered coverage from another employer.  These changes met with great resistance prompting the University to suspend the penalties associated with the wellness program, although the smoking and spouse/SSDP surcharges were maintained.  The tobacco surchage did not seem to have been very successful since it was based on self-reported information provided.  For example, a much smaller percentage of Penn State employees and their spouses/SSDP reported using tobacco, compared to the general population, suggesting that the policy was not effective at identifying smokers and imposing the penalty.  Importantly, there was a lot of negative reaction by employees to the Take Care of Your Health initiative.  This negative reaction is discussed in other chapters of this report.  The academic literature chapter also discusses the scientific evidence for wellness programs based on published literature, suggesting that the evidence is mixed both in terms of the expected savings and expected health benefits of wellness initiatives.   


·         Primary Market Failures Addressed – Sub-optimal patient behaviors due to moral hazard. 

·         Primary Goals Addressed – Appropriate utilization of health care services; Optimal benefit design and out-of-pocket expenditures for beneficiaries.

    Required Investment for Implementing the Option – Given the experience of the Take Care of Your Health initiative implemented in 2013, it is critical that any program be well designed with documentation from the scientific literature about expected benefits.  In addition, enhanced communication would be critical to explaining the purpose of the program, its intended impacts based on available evidence, and other important factors, such as how the confidentiality of private health information will be protected and shared with contracted third parties.

    Timeframe for Implementation – Would depend on the nature of the program to be implemented and the time to perform due diligence around scientific evidence and communications.

    Potential Risks – Skepticism and lack of trust among employees and beneficiaries based on the rollout of the Take Care of Your Health initiative in 2013.  The burden of proof would be on the University to explain the potential value of the program to be adopted. 

    Financial Return – Depends on the nature of the specific program to be implemented. 

·         Additional Analyses Required – A thorough assessment of similar types of programs offered by other employers and the rationale for those programs, as documented in scientific evidence, as well as important lessons from the rollout of these programs.


G.  Additional Options Related to University Operational Decisions


1.  Ensuring Access to Health Care Providers for Beneficiaries in Western Pennsylvania


Fierce competition between the University of Pittsburgh Medical Center (UPMC) and Highmark Blue Shield, and their respective health care delivery systems, has resulted in public concern about the availability of health care providers for patients and beneficiaries residing and working in Pittsburgh and surrounding areas (e.g., New Kensington, Beaver) and extending North to the campuses in Shenango and Erie and West to the Altoona campus.  The issue at hand is whether certain providers will be available, starting in 2015, in the insurance networks offered by the UPMC health plan and the Highmark Blue Shield plan.  For example, UPMC has said that its providers will not contract with Highmark’s insurance products and there is concern that UPMC’s products will not cover West Penn Allegheny Health System providers.  These ongoing developments have played out publicly in the local media and even legislators have questioned the degree to which the exclusion of specific providers from insurance networks is a violation of anti-trust law or is inconsistent with regulatory statutes of the Pennsylvania Department of insurance (see for example,  .  For Penn State beneficiaries residing in these areas the concern is that their UPMC providers may not be covered by the network offered by Highmark.  As a major purchaser of health care in these markets, Penn State could use its purchasing power and its ‘voice’ to weigh in on the situation due to the potential impact on its beneficiaries.  This might be done by working directly with Highmark to address the issue and/or by working with legislators and others who are trying to address this impasse.  Additionally, the contractual obligation with Highmark through our ten-year agreement will restrict the University's opportunity to provide another insurance option to faculty and staff who may be affected by this situation.


·         Primary Market Failures Addressed – Antitrust concerns due to market consolidation.  

·          Primary Goals Addressed – Timely access to health care providers.

    Required Investment for Implementing the Option – Depends on the specific approach taken.  For example, lobbying or working with legislators is a different approach than trying to work directly with Highmark.

    Timeframe for Implementation – The impending changes are expected at the beginning of 2015, so time is of the essence to avoid significant gaps in access to providers.

    Potential Risks – The risk of doing nothing seems significant.  If enough Penn State beneficiaries were to be affected, there would be serious concerns about timely access to health care providers.

    Financial Return – Beneficiaries could face significant out-of-pocket expenditures to maintain existing provider relationships.

·         Additional Analyses Required – Use claims data to assess the number of Penn State employees and beneficiaries at risk if their existing providers are not covered by Highmark’s network.  For example, look at historical billing to see how many UPMC vs. West Penn Allegheny Health System providers were seen by PSU beneficiaries.  This would provide an estimate of the likely problem should Highmark’s network exclude UPMC providers.


2.  Eliminate Payments for Stop Loss Insurance


Penn State self-insures for the costs of medical claims for employees and retirees that are not covered by beneficiary premium contributions or cost sharing.  Some self-insured employers choose to purchase ‘stop loss’ insurance coverage to protect the employer in the case where claims experience reaches exceedingly high levels.  This insurance is purchased through a reinsurance vendor that charges a premium for taking on this risk.  Currently Penn State is paying over $1 million dollars annually for stop loss insurance.  Towers Watson has advised that Penn State could save the amount paid for stop loss insurance by managing the risk on its own.


·         Primary Market Failures Addressed – Not applicable.  

·         Primary Goals Addressed – Reductions in health benefits expenditures and trends in expenditures.

    Required Investment for Implementing the Option – A consideration of whether additional cash must be kept in reserve in order to cover potential risks due to claims fluctuations.

    Timeframe for Implementation – A change could be made at the next stop loss policy renewal period.

    Potential Risks – Facing a period of claims experience that was unusually high.

    Financial Return – Savings of monies currently paid to others to help protect against unusually high claims experience.

·         Additional Analyses Required – In consultation with university risk managers, determine the claim level that would put too much financial burden on the university.


3.  Evaluate the Need for Continuation of Internal University Subsidies to Hershey Medical Center


As part of its employee and retiree health benefits budgets, Penn State has been subsidizing Hershey Medical Center and its' providers in two ways.  First, there is an approximately $2.1 million dollar transfer (2013 value) made annually to the medical center that was initiated in 2008 at the time the University switched its TPA contract from Coventry/Health America to Highmark.  While there may have been practical reasons for this transfer when it was initially made, it is not clear why this expenditure should be continued or included as part of the employee and retiree health benefits budgets.  Second, the University has paid annual amounts, ranging from $646,000 in 2010 to approximately $6.1 million  in 2013 for a cumulative total of $15 million to date, to support the opening of additional Penn State Hershey Medical Group clinics in State College.  The rationale for supporting these clinics out of the employee and retiree health benefits budget is to increase access to health care providers, particularly primary care providers, and to create competition in the State College markets.  At the time these clinics were developed, and the subsidies started, the University was experiencing significant increases in costs and was faced with very large potential increases in its health care costs due to a lack of available alternatives to existing providers.  Since both of these subsidies are included in the health benefits budgets, and thus contribute to the calculation of annual trends, it is worth thinking about whether these transfers are necessary, and if so how they are justified, given that they contribute to overall health benefits spending on an annual basis and if eliminated from these budgets, could result in sizeable savings.  If these subsidies are strategically important, such that the presence of the subsidy results in other savings to the health benefits budget, then these benefits should be clearly articulated and monetized. 


·         Primary Market Failures Addressed – Although the establishment of the HMC clinics was designed to address an increasingly less competitive market in State College, going forward there could be antitrust concerns due to market consolidation.  

·         Primary Goals Addressed – Fair pricing; Timely access to health care providers.

    Required Investment for Implementing the Option – Internal discussions about eliminating these subsidies or clearly articulating why they should be included in the health benefits budget.

    Timeframe for Implementation – A change could be made quickly to remove these subsidies from the health benefits budgets.  If subsidies are continued they could be moved to other university budget categories that make more logical sense.

    Potential Risks – Reduction in revenue for Hershey Medical Center.  A reduction in subsidies has an impact on the fiscal health of the new primary care clinics that have been recently established in State College.  The elimination of the subsidies could result in the closure of the clinics, since these medical practices have an annual operating loss, and the elimination of a viable alternative could lead to an increase in health care costs overall, especially given the positive performance of these practices on a quality and cost basis.

    Financial Return – Immediate savings realized in the health benefits budget, although these would have to be balanced against anticipated future costs if competition in the market is reduced.

·         Additional Analyses Required – A better understanding about whether there is direct value received for these subsidies that are clearly related to providing health benefits for employees and retirees at Penn State.


VI.  Summary


This section has provided an analysis of many potential paths forward, including a summary of the state of knowledge of each option as well as a description of the questions that would need to be addressed before an informed decision could be made regarding the applicability of each to the situation faced by the Penn State community.   It should be noted that these options are not necessarily mutually exclusive and some options can work in conjunction with existing benefits offerings while others would replace existing programs.  The list of options examined is of course not exhaustive as there are many alternatives that have not been examined; the goal has been to consider the most salient possibilities.  Nor has the Task Force "prioritized" this list, as there is a great deal of uncertainty involved with all of the options, and with some more than others.  It is difficult to obtain truly accurate numbers for making an informed decision, and the process would necessarily involve detailed discussions with vendors, TPAs, and program implementers.  The goal has been to provide a starting point for an informed dialogue between the University's stakeholders on the possibilities, and also to provide a framework to address the tradeoffs that such a discussion will necessarily entail.

Report of the Communications Subcommittee

The Communications Subcommittee of the University Health Care Task Force was charged to “[i]nvestigate improvements to communication, understanding, consultation, transparency, and employee participation for current and future health care programming and health care benefits.”


As inputs for this report, we relied on the following sources:


·         Commentary by the Faculty Council of the College of Education

·         Focus Group with three University Park faculty members

·         Focus Group with four senators from Commonwealth campuses

·         Complaints received by TF about Cost of Health care and administration of Health Savings Account

·         Our exchange with Value Subcommittee

·         Social Media Chatter

·         AAUP discussions

·         Published articles and book


We should note that the views from employees obtained by the subcommittee are not necessarily representative of the views of all Penn State faculty and staff. As is often the case with controversial issues, those with negative or dissenting opinions are likely to be prominently represented, but these opinions may or may not represent the majority. We would not know whether they did or not unless we were to assemble a representative sample and conduct a scientific study of the opinion climate surrounding the healthcare issue at Penn State. In the absence of such a study, the subcommittee decided to examine whatever opinions came our way and also reached out to those who had been vocal in their discussion of the healthcare plan rollout last year. We deliberated on the issues raised and discussed them with the entire taskforce.


A central conclusion of our deliberations is that, whenever a new initiative of this magnitude is contemplated at Penn State, it is critical to involve employees at all times, from the earliest stage of planning and coordination to final decision-making and implementation. Communications with faculty and staff should be consultative in nature and not simply educational or informational. With this in mind, we recommend six strategies for improving communications between the administration and the employees:


I.  Communication of Context and Constraints.


First and foremost, it is important for Penn State to address clearly and forthrightly the reasoning behind any proposed health care initiative. It can do so by clearly outlining the larger policy context and the resulting constraints faced by the University. Examples of such communications include:


A.   An explication of the Affordable Care Act, including aspects that could benefit employees' health, and its impact on Penn State.


B.  The demographic and health characteristics of Penn State employees with implications for health care expenditures.


C.   Regional competition among health care providers to offer services to Penn State employees.


D.  Competitive practices among providers in different parts of the state, creating the need for diversity of plans within the Penn State system.


E.   An outline of the health economics at Penn State, including a discussion of the wage-benefit tradeoff.


F.   The rationale behind the cost-containment strategy adopted by the senior leadership.


G.   The constraints imposed by the 10-year contract with Highmark.


H. Evidentiary basis for emphasizing wellness—do the insurers mandate it or does PSU have evidence to suggest that wellness programs are indeed beneficial to employers and employees?



II.  Communications to improve Transparency and Consultation in Decision-Making.


An essential feature of an effective communication policy is improving the transparency of the administration’s decision-making process and providing the faculty and staff with opportunities to shape and test programs before they are rolled out as a “done deal.” The following actions would fall under category.


A.   A flow chart explaining how Penn State makes decisions affecting its employees, along with the time-line for consultations and approval. Presumably, the Office of Human Resources (OHR) goes through a planning cycle before launching any major initiative, keeping in mind the various deadlines for consulting with the Senate benefits committee, informing the board of trustees, and so on. At the earliest possible phase of this cycle, the University community should be made aware of the movement of the proposed initiative through these various steps.


B.   Active efforts to seek input from all stakeholders at critical stages. A reasonable period of time should be set aside for input from faculty and staff so that meaningful changes can be made to policy in response. The goal should not be to simply obtain “buy in” from employees, but give them an opportunity to shape the proposal so that they have a sense of ownership.


C.   Greater involvement of elected representatives of employees. The Faculty Senate Benefits Committee is our best structured, standing group for reviewing policy changes.


D.  Pre-policy townhall meetings, both physical and virtual, and other mechanisms for people to ask questions, voice concerns and make suggestions.  For example, some Colleges are holding town-hall meetings to discuss the upcoming Gen Ed reform.  OHR could do the same thing during the initial stages of planning employee-related initiatives.


E.   Gauging community reactions to potential policy directions, through blogs and other social media (e.g., the recent initiative by the Gen Ed taskforce at


F.   Assessing support in the community for specific policy options with the aid of employee focus groups and other research methods when there are clear choices to be made.


G.   Timing the rollout of initiatives to occur during the middle of Fall and Spring semesters (when they are likely to be noticed by most people), not during summers or holidays.



III.  Building a Culture of Open, Honest and Frequent Communications.


HR communications relating to the welfare of employees and their communities ought to reflect the spirit of dialogue and be a part of a culture of openness in information exchange rather than shrouded in secrecy. Suggestions include:


A.   Year-round communications about benefits-related issues instead of being focused around a short period in the Fall.


B.   Constant/frequent communications. Silence on the part of the administration can breed suspicion and doubt, leading to rumors and conspiracy theories.


C.   Two-way communication forums are sorely needed in the Penn State community, so that employees can not only hear back from the administration but also be exposed to the voices of fellow Penn Staters.


D.  Better communication vehicles about HR matters, e.g., the monthly ITS newsletter, that have the flavor of water-cooler conversations about matters of common welfare and serve to build a sense of community.


E.   More detailed information about the expenses and management of health care and retirement plans, e.g., details of “revenue sharing” and rationale for high “expense ratio” of funds offered in the health savings accounts (HSAs).


F.   Tips to employees about how to navigate through the maze of the health care system, just as there are tips about leading a healthy lifestyle.


G.   PSU News media and press releases should reflect views of faculty and staff, not just those of the administration.


IV.  Recognizing Individual Sensitivity and Respecting Personal Privacy.


HR must communicate that employees have an absolute right to privacy of their personal information, which means they have a real choice in determining who has access to it and who does not. Of course, information is routinely collected in the natural course of patient diagnosis, treatment and payment by insurers and health care providers, and this information is clearly protected by HIPAA[28] privacy rules. And, while employees may be comfortable with HR handling certain kinds of sensitive information, such as salaries for example, they may not want to share most other aspects of their personal life with their employer. Penn State must respect individual freedom in all health-related decision-making and make every effort to provide employees with the most current information to aid that decision.


A.   Employees should be given choice and an opportunity to consent before their personal health information is collected by Penn State for administrative purposes.  If employees opt out of the use of this data, however, it could make analysis (for purposes of changing plans, or reducing costs, for example) difficult to implement.  A firewall, or handling of the data outside of the University through an entity that satisfies the HIPAA privacy requirements, might be an acceptable alternative.


B.   Motives for collecting personal health data about employees should be made explicit.


C.   Given major security breaches of confidential information at the highest levels of government and corporate sectors, Penn State must understand that it is of paramount importance to protect the security of the private information collected about its employees.


D.  Employees should not be asked to part with their privacy for a price/fee.


E.   Employees should not be asked to pay a price for protecting their privacy.


F.   Employees should be given opportunities to customize their health care choices, i.e., self-tailoring, rather than being asked to reveal information about themselves for the system to personalize their choices for them, i.e., system-tailoring.  Some employees are upset about WebMD automatically issuing health warnings and diagnoses based on private and/or incorrect/outdated information imported from other systems (biometric data or pharmaceutical records, for example).  Employees should be provided options for specifying the nature of collation of information from the various systems.


G.   Be aggressive in protecting employees’ right to privacy of their health information by employing claims-based data whenever possible. For example, PSU could obtain data on utilization of preventive visits or services directly from Highmark rather than requiring enrollees to undertake a wellness survey and undergo biometric screening.


V.  Communications to Build Credibility.


Given the negative reaction to last year’s rollout of the health initiative and the fallout from it, Penn State must work hard to regain the trust of its faculty and staff. The following should be key guidelines governing all communications between the administration and the employees.


A.   Embrace a culture of participation. All major decisions should be based on input from the relevant stakeholders.


B.   Practice shared governance. Compositions of committees and taskforces should involve actual stakeholders and ought not to be driven by administrators in a top-down fashion.


C.   Make use of the disciplinary expertise of our faculty when we design new programs. For example, faculty experts could have helped the University by bringing to bear the scholarly body of literature on wellness programs before the University adopted Highmark’s preventive-health plan.


D.  Embrace principles of fairness and equity. For example, the administration should go out of its way to explain why salary-based indexing of health care costs is fair to employees in all salary bands.


E.   Embrace principle of equality. Any and all instances of differential treatment of employees (e.g., health care cost differences based on their smoking status or the employment status of spouses) ought to be explicitly addressed by the administration.


F.   Fully disclose how employee data are being used and will be used in the future.


G.   Provide real choice to stakeholders and explain the rationale for incentives. For example, incentivizing employees to choose one health plan or provider over another will be seen as limiting choice unless the rationale for the policy is clearly explained.


VI.  Communications to Demonstrate Goodwill and Collegiality.


Penn State should follow its own dictum of “Take Care of Your Health” by taking care of its employees. The administration needs to consider the “human face” of its employees and express goodwill in ways that serve to boost their morale. Some suggestions:


A.   Offer rewards for pursuing wellness rather than punishments for not doing so. Do not try to frame the latter as the former.


B.   Actions should convey that PSU is looking out for the welfare of its employees rather than simply pursuing a cost-cutting agenda.  Communicating the evidence of benefit (such as improved health outcomes) for specific types of behavior change programs and for clinical preventive services is important for helping employees understand why these activities are offered or incentivized.


C.   Provide sufficient lead time, with adequate opportunity for consultations and clarifications, whenever employees are required to sign up for a new initiative. (e.g., the dependent verification program in 2013 worked better in part because of the long lead time).


D.  Change management protocol for all new initiatives should include a trial period before actual implementation and/or a staggered rollout in stages.


E.   The language of University communications about employee benefits should be designed to reflect employees’ interests rather than those of our vendors. There was a great deal of resentment last year about PSU administration and OHR using the “talking points” given to them by Highmark, as if the University was representing the insurance provider rather than its employees.


F.   Provide just-in-time information about their choices to help employees see the consequences of their decisions at the time that they make them. For example, several employees were surprised by the steep increase in their premiums this year. This could have been avoided if they were fully informed about the implications of their choices at the time they made them last Fall.  Of course, the faculty and staff do need to pay attention to the information provided if it is to be useful.


G.   Demonstrate its benevolence at every opportunity. The University should communicate the dollar value of the benefits that it provides for employees. For example, more online tools in the Employee Self-Service Center could be provided for individual employees to assess the value of the benefits they receive from the University.


In sum, the taskforce calls for a major shift in the nature, tone and structure of communications between the administration and the employees at Penn State with respect to health benefits, both in planning and implementation of major initiatives. The culture of behind-the-scenes orchestration of such important initiatives should be replaced with a culture of open and free flow of information and frequent consultation, creating the climate for transparent decision-making and shared governance. Instead of being coercive and punitive, communications about new initiatives should provide greater agency (control and choice) for individual employees in matters relating to their personal welfare. Above all, communications should be seen not as information transmission from the administration to the employees, but as a dialogue between the two.

Report of the Health Care Trends Subcommittee

I.  Background

The health care delivery system in the United States is a complex multifaceted web that includes four main categories of interdependent players:  health care providers, patients, payers and employers.  The complexity of the system has developed as the system has evolved, from one in which the patient had a direct relationship with a single physician who provided the majority of care needed, with the patient paying for the services, to one in which care is delivered by a multitude of providers across a variety of settings, often without communication between/among those providers; payment is provided through insurance provided by employers or government based on employment status, age or income; patients remain ignorant of and protected from the actual costs of the services being provided; and providers receive payment based on the amount of work, rather than the outcomes of that work.

The costs of health care are staggering; in 2007, spending on health care services in the U.S. totaled more than $2.2 trillion, or 16.2% of GDP, with projections for health care spending to exceed 19% of GDP by 2017 (Bipartisan Policy Center, 2009).  While the rate of growth has decreased in recent years, it is widely suspected that much of this decrease has resulted from patients delaying needed care because of economic uncertainty – a situation that may result in significantly higher spending as the economy improves and consumers become more confident (Kaiser Family Foundation, 2012). 











Finding ways to reduce the rate of growth in health care spending – “bending the cost curve” has been a focus of the federal government over the last several years.  A variety of interventions have been tried, ranging from the introduction of price controls for health care providers, to the development of Medicare Advantage Plans and the passage of the Patient Protection and Affordable Care Act, with a myriad of ideas and policy changes.  And yet, spending on health care services is concentrated in a relatively small number of patients.  It has been estimated that just 1% of the U.S. population accounts for more than 20% of total health care spending, the top 5% account for almost 50% of spending and the top 15% account for almost 75% of all spending ("Better Care at Lower Cost: Is It Possible?," 2013; Kaiser Family Foundation, 2012).  This concentration of spending may suggest that broad policy changes are not the best way to reduce health care expenditures.

These costs are ultimately paid by all of us.  Health insurance premiums have consistently outpaced inflation since 2000 and the experience at Penn State is no exception.  These increased costs may be felt through higher employee cost sharing, either as a larger share of premiums or increased deductibles and co-payments; or increased premium costs on the part of employers, which may limit wage increases (Kaiser Family Foundation, 2012).  The Kaiser Foundation estimated that the median employer contribution for health care insurance premiums was at 12.8% of payroll expenses in 2010 (Kaiser Family Foundation, 2012).  And even with employers bearing so much of the cost, individuals, especially those with multiple chronic conditions, continue to bear relatively high costs.  In 2008, patients with 3 or more chronic conditions were far more likely to pay 10% or more of their income for health care related costs (including health care insurance premiums) than patients without multiple chronic conditions.

II.  Factors Impacting Health Care Costs

Although a comprehensive discussion of all the factors that contribute to increases in health care costs is beyond the scope of this document, a number of particularly important factors follow.

A.  The increase in chronic conditions

Due in part to the aging of the U.S. population, the prevalence of a number of chronic conditions is increasing.  Heart diseases, diabetes, and obesity, among others, are not only costly, but also place significant stress on a system that is not designed to provide comprehensive, coordinated care over time, including for those with multiple chronic conditions.  This lack of coordination of care results in the provision of medical services that may be unnecessary, with services duplicated from one setting to the next, or patients receiving less than optimal follow-up care after discharge from the hospital, resulting in readmissions.  

B.  Waste and abuse

These are also a major factor in health care costs, with waste including overtreatment, unnecessary or inappropriate care, medical errors, excessive administrative costs, and actual fraud.  The overuse or inappropriate use of medical treatments and technologies not only contributes to rising health care costs, but also brings additional risks and complications which themselves cause increased spending. It has been estimated that as much as 30% of all health care dollars are wasted.

C.  Increased prevalence of health care insurance

Health Insurance impacts the growth in health care costs by making it more likely that patients will access care, although the additional dollars spent on care provided by physicians for prevention and treatment services may be more than offset by reductions in the use of emergency services for acute care or increased costs related to delays in accessing care.  Nonetheless, the fact that insurance covers more of the population may encourage the development of new, additional technologies and treatments since a payment source exists (Kaiser Family Foundation, 2012).

D.  Price increases associated with drugs, medical devices and hospital care

Pricing is one of the significant drivers of  overall health care cost increases (Moses MD et al., 2013).    This occurs, at least in part, because of the introduction of new, innovative technologies and the development of treatment options for conditions that have previously been untreatable or that may have had limited options, such as cancer or HIV (Kaiser Family Foundation, 2012).  It has been estimated that almost half the overall increase in health care spending over the last several decades is due to the development of new technologies, which have had a substantial impact on care provided (Bipartisan Policy Center, 2009).

An important consideration in these price increases is the relatively limited impact that they have had on the ultimate consumer – the patient.  Since 1980, patients have had personal out-of-pocket spending decrease, with payers now covering more than 90% of all hospital and physician costs; this protection from the actual costs of services is believed to significantly impact patients’ use of services (Moses MD et al., 2013).

We now face a crisis in health care – care that is too costly, focuses more on acute illness or injury than on maintaining health, is fragmented, delivers uneven outcomes, is not accessible by large numbers of patients in need and is not well prepared for the large numbers of aging, chronically ill patients that are threatening to rapidly overwhelm our existing system.  As a society, the costs of health care are forcing many families to delay or cut back on needed health care services.  The Kaiser Family Foundation, in a 2011 health tracking poll, found that 50% of Americans said they had to cut back on medical care in the prior 12 months because of cost concerns (Kaiser Family Foundation, 2012).

III.  Challenges facing the health care system

There are large, complex and difficult issues that must be addressed as we attempt to retool the health care system to achieve better outcomes at lower costs.  A better understanding of the challenges that providers face is an important step in recognizing the actions that are likely to occur over the next few years. The following list, while not comprehensive, discusses the 6 most significant issues that will have to be specifically addressed over the next several years at Penn State, as well as nationally.

A.  Increased Incidence of Chronic Medical Conditions

While we may not attribute recent growth in health care costs to demographic changes, it is clear that the aging of the U.S. population is likely to have a significant impact on the system moving forward.  The increased incidence of chronic medical conditions will require a fundamentally different approach to care than the acute episodic system of care that we currently have.  Not only will we have to begin changing how we provide medical care, but we will also have to address some of the economic, cultural and societal factors that impact the need for health care services, such as the number of uninsured, our cultural bias towards “more is better”, the high rates of obesity, drug use, and other lifestyle choices that impact the need for health care services.

B.  An Undersupply of Primary Care Physicians

It has been estimated that we will need approximately 261,000 primary care physicians in 2025, up from just under 210,000 in 2010.  Given the limited number of medical students who choose primary care as their specialty, bridging this gap may prove difficult. The development of new team models of care, the current work to ensure that all providers work at the highest level of their training, the growth in physician extenders such as physician assistants and nurse practitioners and the increasing use of telemedicine may all help to offset these shortfalls.  These changes will, however, require changes in both reimbursement policies and, equally important, acceptance of changes in the traditional physician-patient relationship.

C.  Slow Adoption of Evidence Based Medicine

There is growing acceptance that evidence based medicine (EBM) – clinical treatment approaches and protocols based on the results of rigorous study of comparative approaches and outcomes, could positively impact both the cost of health care services and outcomes of care.  Nonetheless, adoption of EBM is slow and subject to individual practitioner’s judgments and acceptance of these protocols.  Research studies have shown that, overall, just over half of the recommended treatments for acute, preventive and chronic conditions are consistently delivered.

D.  Problems with Fee-for-Service Payments

The current fee for service payment methodologies are a barrier to the development and adoption of new, innovative treatment and care systems that may break through the current fragmented systems of care.  The number and variety of parties who are involved in payment and regulation of the health care industry introduces both complexity and rigidity as providers struggle to meet the standards and provide services consistent with payment incentive systems.

E.  Changes in Care Delivery Settings

The shift from inpatient to outpatient care, or the movement from institutional settings to the home, introduces a greater level of complexity as medical professionals struggle with understanding how to deliver care in these new settings, often with regulatory requirements that may be inconsistent with the new delivery options.

F.  A Growing Focus on Value

Value, which is defined as a function of quality, satisfaction and cost, is conceptually important but poses some unique challenges.   Defining and measuring critical aspects of quality, especially those focusing on outcomes rather than the process of care; understanding the interplay between individual risk factors and socioeconomic context on an individual’s health and response to medical care; and determining the best ways to present and explain what is often complicated information so that patients can make informed decisions – these all remain works in progress.

Providers are reacting to these, and a myriad of other challenges, in many ways.  Some examples include investments in infrastructure such as information technology, care management and care coordination and the development of new team models of care will help with some of the challenges.   Additionally, there is a significant trend of hospital consolidation into regional and national systems, as well as the growing employment of physicians by hospitals and systems.  These latter two actions have potentially significant import for the costs of health care.  Whether they will play out as helping to reduce health care costs through the achievement of cost efficiencies due to scale and scope, the development of sufficient size to successfully enter into new delivery organizations such as ACOs and Clinically Integrated Networks and the conversion of struggling inpatient facilities into critically needed outpatient centers, or whether these trends will protect providers from the forces driving towards change is yet to be seen.

IV.  Patient Protection and Affordable Care Act (PPACA)

In response to rising concerns about the costs of health care, a lack of consistent high-quality outcomes, the problem of the uninsured and perceptions about the fundamental structural flaws of the current delivery system, the Patient Protection and Affordable Care Act was passed in 2010.   PPACA is intended to increase the availability of affordable health insurance coverage, reduce the numbers of uninsured and reduce the costs of health care, for individuals and the government. PPACA has five major objectives, including:

·         Improved coordination of care across the continuum;


·         Enhanced prevention and addressing the drivers of chronic disease;


·         Increased quality and value of care;


·         Improved patient experience; and


·         Reduced overall health care costs. 

Although there are a number of items included in the law that specifically target cost reductions, such as reducing update factors for hospital services or decreasing payments for Medicare Advantage plans (Cutler, 2010), these are generally included as a way of paying for many of the changes that are designed to address the efficiency of care provided.  Key provisions address prohibitions on pre-existing conditions as a reason for denying insurance coverage, changes in community rating processes, establishment of a set of minimum/essential benefits, mandated first-dollar coverage of preventive services, requirements for individuals to have health insurance, subsidies for individuals based on income criteria, requirements for businesses to provide insurance or pay penalties, the establishment of a set of insurance exchanges to purchase health care insurance, expansion of Medicaid eligibility (States can choose whether or not to participate), new provider payment pilot projects such as bundled payments, and a series of new fees and payment reductions intended to help pay for the overall costs.

Of particular interest are some of the Act’s sections intended to encourage innovation and change in how health care services are organized, delivered and paid for, including the development of medical homes for patients with multiple chronic conditions, bundled payments, accountable care organizations, increased primary care funding, enhanced prevention activities, and the development of the Center for Medicare and Medicaid Innovation.

Two areas in particular are important to the discussion of the Penn State University health care spend – the increased emphasis on health care prevention and the opportunities for innovation in care delivery and payment.

A.   Prevention Focus of the Affordable Care Act

The 2010 Patient Protection and Affordable Care Act (PPACA) contains a focus on prevention that is unprecedented in U.S. health care policy.  Prior to passage of PPACA, there was no mechanism to standardize provision of preventive services in private sector health plans, and the degree of cost-sharing for preventive services in private plans was plan-specific.  Under PPACA, employer-sponsored group health plans and private health insurance plans that are not “grandfathered” are required to provide a standard set of clinical preventive services with no deductibles or copays.  This includes employer-sponsored health plans such as that at Penn State.

Several provisions in the law – particularly in Title IV, Prevention of Chronic Disease and Improving Public Health – address prevention programs and health insurance coverage for preventive services (Koh & Sebelius, 2010).  Most relevant to our purposes are two provisions:  (1) mandated coverage of a standard set of clinical preventive services without cost sharing in private health plans, and (2) promotion of workplace wellness programs.  As these provisions are expected to increase consumer demand for primary care services, the law also authorizes investment in developing a primary care workforce to promote prevention in underserved areas.  It should be noted that the set of biometric screenings that is often included in workplace wellness programs (e.g., measures of body mass index, blood pressure, blood lipids, glucose levels) is not equivalent to the set of clinical preventive services mandated for first-dollar coverage.  This discussion focuses on the clinical preventive services mandate.

B.  Why Clinical Preventive Services?

Clinical preventive services include recommended screenings, immunizations, counseling, and other services that prevent health problems from developing (primary prevention) or detect diseases at early stages when they are more amenable to treatment (secondary prevention). Counseling services are included because modifiable risk factors – such as smoking, nutritional deficits, and sedentary lifestyle -- account for a substantial proportion of mortality and morbidity in the U.S. (see for example: (Danaei et al., 2009)). Yet studies show that many recommended clinical preventive services are underused.  American adults receive only about 55% of recommended preventive services (such as cancer screenings and immunizations) and only 18% of recommended preventive counseling services such as tobacco cessation counseling (McGlynn et al., 2003).  Increasing the use of recommended evidence-based clinical preventive services to 90% from current levels could result in estimated total savings of $3.7 billion (Maciosek, Coffield, Flottemesch, Edwards, & Solberg, 2010).  Thus a goal of U.S. health policy is increasing the provision of underused recommended clinical preventive services.  

The logic of offering the recommended clinical preventive services without cost-sharing is that out-of-pocket costs have been shown to be a barrier to consumers’ use of preventive care, especially among those with lower incomes; removing these costs is expected to increase utilization of the recommended preventive services and, over time, to improve the health of the population.  A healthier population is expected to consume fewer treatment-related services, resulting in lowered overall health care expenditures.  There is some evidence from Massachusetts – where health care reform was enacted in 2006 – that between 2001 and 2011 utilization increased for clinical preventive services (i.e., cholesterol screening, cervical cancer screening, and colorectal cancer screening) and health status improved, relative to other New England states (Van der Wees, Zaslavsky, & Ayanian, 2013).

C.  Clinical Preventive Services Covered Under PPACA

ACA identifies a specific set of clinical preventive services for coverage without cost sharing.  The preventive services that are covered in Section 2713 of PPACA follow.

1.  Those receiving a rating of A or B by the U.S. Preventive Services Task Force (USPSTF). 

The USPSTF is an independent panel of nonfederal experts that develop what are considered to be the “gold standard” for evidence-based primary care practice.  USPSTF recommendations are specific as to age, sex, and appropriate periodicity for the service (e.g., annually, biannually). Ratings of A and B are the two highest ratings for preventive services and indicate that there is sufficient evidence of benefit with regard to health outcomes to justify a recommendation to offer or provide the service in primary care practice (U.S. Prevention Services Task Force, 2008).  USPSTF recommendations are reviewed periodically to account for the changing evidence base.  Because of its stringent evidence requirements, the USPSTF’s recommendations may differ from those based on the expert opinions of professional groups or advocacy organizations.


The USPSTF recommendations may be viewed at:

One departure from the current USPSTF recommendations is noteworthy.  PPACA covers mammography screening every 1-2 years beginning at age 40 based on the 2002 USPSTF recommendation, rather than screening every 2 years for women ages 50-74 who are not at increased risk for breast cancer, as recommended in 2009.  (The 2009 recommendation is now under review by the USPSTF.)    

2. Immunizations recommended by the Advisory Committee on Immunization Practices (ACIP) and issued by the Centers for Disease Control and Prevention.

These recommendations may be viewed at:


3.  For children and adolescents, services recommended in guidelines of the Bright Futures Steering Committee of the American Academy of Pediatrics (Pediatrics, 2008). 


These recommendations may be viewed at:


4.  For women, preventive services recommended by the Health Resources and Services Administration (HRSA) based on the Institute of Medicine’s Committee on Preventive Services for Women (IOM, 2011). 


The HRSA recommendations may be viewed at:

Examples of clinical preventive services for adults from these combined sources include the influenza vaccine, tobacco counseling and cessation interventions, FDA-approved contraceptive methods and counseling for women, and colorectal cancer screening for men and women over age 50 ( It is important to recognize that a number of high-profile preventive services are not covered by PPACA because the USPSTF recommends against their provision; a noteworthy example is prostate cancer (PSA) screening.

D.  Potential Impact of PPACA

Although it is too early to tell what impact PPACA will have on health care utilization and health outcomes among the privately insured, an immediate impact is likely to be greater demand for primary care office visits to obtain clinical preventive services.  Given the number of clinical preventive services covered without copays, more than one preventive visit may be needed to obtain the covered preventive services for one’s age/sex group and must be provided without cost-sharing.  This possibility is specifically recognized in HRSA’s recommendations for women’s preventive care under PPACA (

Due to the different periodicity for specific clinical preventive services, primary care preventive visits could increase initially as patients “catch up” with the recommendations for their age/sex group, but later level off.  In monitoring the impact of PPACA on visits, it will be important to distinguish between preventive visits (which are expected to increase) and visits for acute care and treatments (which may decrease over time as prevention improves).

In addition, it will be important to monitor appropriate use of clinical preventive services under PPACA.  The covered clinical preventive services are accompanied by recommendations specifying the applicable age and sex groups, risk status (e.g., average risk or high risk based on some indicator), and the frequency of provision.  Adherence to these recommendations is measured as percent of the eligible population (defined by age and sex and, if applicable, risk status) that receives the service within the specified time period.  As noted, most clinical preventive services are underused at the population level and the policy focus has been on increasing the use of these services. There are, however, concerns that PPACA coverage of clinical preventive services without copays could lead to overuse of the recommended preventive services if payments to providers are not tightly linked with the specific recommendations.  Overuse is a concern because of increased financial costs incurred, increased risk of “false positives,” and the associated burden of suffering with follow-up testing and possible unnecessary treatments.

A key issue for employers such as Penn State is ensuring that health plan enrollees are informed about the covered clinical preventive services so that they can make informed health care choices to optimize their health.

E.  Innovative Delivery Reform Focus of the PPACA.

PPACA includes several new models of care and/or payment that are intended to foster a shift in the health care delivery system.  Specifically, increased payments for primary care, new models of care coordination for dual eligible beneficiaries, programs to provide care at home for beneficiaries with high needs and the development of accountable care organizations, bundled payment programs and value based payment mechanisms are all included in PPACA.  While these different programs target a variety of patients using different delivery and payment mechanisms, the underlying assumption is that by combining new approaches to organizing and delivering care with payments designed to support those changes, we can increase quality while reducing costs.  Three particular programs appear to have some evidence of efficacy in achieving these goals – management of chronic disease, improved care coordination and bundling payments.


1.  Chronic Disease Management 


As previously stated, a very small percentage of the overall population is responsible for a disproportionately large amount of health care spending.  Putting aside high-cost, single cause episodes such as trauma or cancer, one of the largest drivers of health care spending is for the management of chronic disease.  Approximately three quarters of the spending in the U.S. is for care related to chronic disease (Bipartisan Policy Center, 2009), with considerable evidence that some large percentage of the effective treatment options are not followed.

Management of chronic diseases generally fall into two categories: (i) those targeted at populations who have or are at risk for a particular chronic disease and (ii) specific disease management programs aimed at specific high risk patients.  The evidence would suggest that specifically targeting the highest risk patients and providing a multi-pronged, longitudinal approach to managing their particular condition is most likely to result in improved quality and reduced costs. 

2.  Care Coordination

One of the most frustrating aspects of the current health care system is the degree of complexity and fragmentation that exists.  Not only does this result in difficulty for patients and their families as they try to navigate through a system that is not designed to be patient friendly, but it often results in duplicative tests and lost opportunities for coordination of care among providers.  Care coordination is designed to overcome these flaws in the system by focusing on effective transitions from one level or setting of care to another and by facilitating the exchange of information from one provider to another.  The development of the Patient Centered Medical Home embodies the concept of care coordination, focusing on the primary care environment as the coordinating body for ongoing care.  Evidence suggests that enhanced care coordination, targeted specifically at high-risk patient groups, can result in improved quality and reduced health care costs (citation).


3.  Bundled Payments 


The concept of bundling payments is relatively simple.  A provider assumes financial risk for all care provided to a patient within a single episode of care, typically defined as the period of hospitalization for a specific, pre-defined condition, plus a period of up to 90 days post-hospitalization.  This financial risk includes not only the care from the provider at risk, but all care received from any provider (with certain exceptions for specific, pre-defined events).  Because the provider is financially at risk, there is considerable incentive to make sure that the patient receives all care necessary to ensure that the patient is compliant with care plans, receives and takes medications, avoids readmissions, etc.  The at-risk provider is suddenly faced with a need to understand and manage the myriad of factors that may traditionally cause patients to be non-compliant or to require high cost care interventions.

While significant evidence demonstrates the efficacy of bundled payments in achieving cost savings while maintaining or improving outcomes, the underlying premise of greater incentives to providers for improving outcomes and reducing costs – essentially by shifting the risk to the provider, is present in the concepts of value based purchasing, accountable care organizations and clinically integrated networks.  In each of these, the provider is given the financial incentive to move outside their traditional framework of care and develop new interventions that may be effective in managing a patients’ condition either well beyond the timeframe normally under their control or by using interventions that are not allowed or paid for under traditional payment systems.

F.  Best Options Moving Forward?

In reviewing those reform activities that appear to have the greatest potential for positively impacting quality, outcomes and costs, three characteristics appear to be important.  First, focus on improving the delivery of primary care services, including prevention.  This includes not just a sufficient number of primary care physicians, but an organized primary care delivery system that includes physicians, physician extenders and support systems, such as a robust information technology system.  Second, focus on reorganizing and redesigning care, incorporating evidence-based medicine and targeting high-risk patients/conditions.  Finally, focus on redesigning financial payment structures to give providers incentive to improve and consequences for not doing so.

Based on these principles, five key activities would appear to be a good starting point for changing the health care delivery system at a local level.

1.  Patient-Centered Medical Homes 

This involves a focus on comprehensive, coordinated care designed around the needs of the whole patient, providing responsive care that emphasizes quality and safety (Peikes, Genevro, Scholle, & Torda, February 2011).  PCMHs have demonstrated cost savings, quality improvements  and enhanced patient satisfaction (Reid et al., 2010) (Grumbach & Grundy, November 2010).


2.  Clinically Integrated Delivery Networks


Groups of providers working together to develop approaches to care redesign and delivery intended to improve quality, reduce costs and impact a population’s overall health status (American Hospital Association, 2010b).


3.  Comprehensive case management 


Research has consistently demonstrated that focusing care management efforts on targeted patient populations, particularly those who are high-risk, high-cost patients, can have a dramatic impact on both quality and cost of care (Academy Health, 2012)


4.  Bundled episodes. 


The Center for Medicare and Medicaid (CMS) has tried a number of approaches to reducing costs of care for beneficiaries.  The concept of “bundling”, that is, providing economic incentives for a provider or group of providers to redesign care associated with specific episodes of care with the goal of reducing costs while improving quality, has proven effective (American Hospital Association, 2010a) (Hussey, Mulcahy, Schnyer, & Schneider, 2012)


5.  Pay for Performance.


The concept behind the myriad of existing pay for performance (P4P) programs currently in existence is simple.  Rather than paying providers for the volume of work they produce, they will be paid at least in part on the outcomes of that work.  The idea of P4P has gained significant momentum, with both commercial payers and the federal government developing programs.  To date, the percentage of overall payments based on P4P is relatively small, although CMS is now penalizing hospitals that do not meet benchmarks against peers (the federal program is designed so that it is budget neutral; there must be “winners and losers”).  Challenges in the P4P world include how to best measure performance, how to account for differences in patient populations so providers are not incentivized to avoid certain difficult patients and the lack of consistently positive results from these programs.  Nonetheless, it appears that P4P is here to stay (Cromwell, Trisolini, Pope, Mitchell, & Greenwald, 2011) ("Health Policy Brief: Pay-for-Performance," 2012).


V.  Potentially Impactful Future Developments

The rate of change we are seeing in health care today is significant.  And the impact of health care reform, not the passage of PPACA or any other legislation, but the real demand for change in spending on health care services combined with improvements in outcomes, is going to shape the health care delivery system over the next several decades.  In addition to the factors identified above that will certainly impact Penn State University’s spending on health care services, a number of other potential changes should be monitored.  These include:

A.  Insurance Exchanges.

How insurance exchanges develop over time and, in particular, whether the ability of individuals to purchase health care in an open marketplace moves health insurance along the same path as retirement benefits have followed in many/most companies – from a defined benefit to a defined contribution.  If this shift should occur, it could have significant implications for the University, including its ability to recruit and retain faculty and staff, the cost of care vs. wages and the ability to more accurately manage overall health care spending through determination of contributions.

B.  Electronic Medical Records.


Health care is a data-intensive industry.  The development of electronic medical records over the last few decades has been tremendously expensive, time consuming and disruptive.  The benefits of “big data” are only just beginning to be realized in an improved ability to understand and impact health care utilization, faster dissemination of the results of research and development of best practice guidelines and incorporation of standardized safety practices into routine medical care.  As we move from capturing data electronically to integrating information across providers and over time, we will begin to broadly understand not only how effective various medical interventions are, but what factors may indicate that an intervention is not cost effective or may not have a positive impact on quality of life.  Patients, providers, insurers, employers, all of society is likely to be faced with difficult discussions about how and when to spend resources; discussions which may be difficult but possible on a population level but much less acceptable when applied to individuals. 


C.  Human Genome Issues.


Progress in understanding the human genome and particularly in how our genetic background may impact the risk and course of disease is astonishing.  Yet we’ve really only scratched the surface of what may be possible.  The concept of personalized medicine – identifying and shaping care within the context of an individual’s genetic background—offers  tremendous possibilities for improved outcomes and reduced costs.  Yet, as with every advance, questions will have to be addressed.  What level of potential benefit must exist before a costly treatment or intervention is justified?  Should an individual’s genetic makeup be a factor in medical decision making, especially at a time when we have so many remaining questions about the human genome and the interplay between genetics and the environment?


D.  New Technologies for Interacting with Patients.


We are quickly becoming a “mobile” society – the advent of smart phones, almost universal access to the internet and increasing expectations of virtually instant response will certainly impact health care, as it has other industries.  The development of telemedicine holds out the potential for improved access to specialty care even for the most remote.  But the development of the technology and the workforce to provide these services is still relatively early.  Uncertainty about funding/payment sources will continue to exist.  Still, the implementation of new ways of interacting with patients, especially in settings outside of high cost institutions, could help reduce spending while increasing satisfaction, especially for those with chronic conditions who must engage with the system frequently.


E.  An Aging Population.


Finally, the aging of the population will not only impact health care spending, but will also impact the availability of the necessary work force.   Already, there is expected to be a shortage of primary care physicians, given the compensation gaps between primary care and specialty physicians, differences in the work/life balance expected by younger physicians and an increasing pressure on residency slots as funding for graduate medical education decreases.  Shortages of nurses will be even more impactful as more than a third of all nurses are already older than 50 and will begin retiring in large numbers over the next decade.  Similar patterns among other health care professionals are expected. The health care industry has not yet developed a solution to what could be significant gaps between required and available health care workers.

VI.  References

1.  Academy Health. (2012). Research Insights: What Works in Care Coordination? Activities to Reduce Spending in Medicare Fee-for-Service. from


2.  Report: Bundled Payment. In American Hospital Association (Ed.), 2010 Committee on Research. Chicago.


3.  American Hospital Association. (2010b). Trendwatch:  Clinical integration --The key to real reform. from


4.  Better Care at Lower Cost: Is It Possible? (2013).   Retrieved 3/12/2014, 2014, from


5.  Bipartisan Policy Center. (2009). Improving quality and value in the U.S. Health Care System. Washington, DC: Bipartisan Policy Center.  Cromwell, J., Trisolini, M. G., Pope, G. C., Mitchell, J. B., & Greenwald, L. M. (Eds.) (2011). Pay for Performance in Health Care: Methods and Approaches. Research Triangle Park, NC.

6.  Cutler, D. (2010). How Health Care Reform Must Bend The Cost Curve. Health Affairs, 29(6), 1131-1135. doi: DOI 10.1377/hlthaff.2010.0416


7.  Danaei, G, Ding, EL, Mozaffarian, D, Taylor, B, Rehm, J, Murray, CJL, & Ezzati, M. (2009). The preventable causes of death in the United States: Comparative risk assessment of dietary, lifestyle, and metabolic risk factors.  


8.  Grumbach, K, & Grundy, P. (November 2010). Outcomes of Implementing Patient Centered Medical Home Interventions: A Review of the Evidence from Prospective Evaluation Studies in the United States, from


9.  Health Policy Brief: Pay-for-Performance. (2012). Health Affairs.  Retrieved March 10, 2014, from


10. Hussey, PS, Mulcahy, AW, Schnyer, C, & Schneider, EC. (2012). Bundled Payment: Effects on Health Care Spending and Quality. Closing the Quality Gap: Revisiting the State of the Science. In Agency for Healthcare Research and Quality (Ed.). Rockville, MD.


11. IOM, Institute of Medicine. (2011). Clinical Preventive Services for Women: Closing the Gaps. In Institute of Medicine (Ed.). Washington, D.C.


12. Kaiser Family Foundation. (2012). Health Care Costs:  A Primer.  Key information on health care costs and their impact. California: The Henry J. Kaiser Family Foundation.


13. Koh, HK, & Sebelius, KG. (2010). Promoting prevention through the Affordable Care Act. NEJM, 363, 1296-1299.


14. Maciosek, MV, Coffield, AB, Flottemesch, J, Edwards, NM, & Solberg, LI. (2010). Greater use of preventive services in U.S. health care could save lives at little or no cost. Health Affairs, 29, 1656-1660.


15. McGlynn, EA, Asch, SM, Adams, J, Kessey, J, Hicks, J, DeCristofaro, A, & DKerr, E. (2003). The quality of health care delivered to adults in the United States. New England Journal of Medicine, 34, 2635-2645.


16. Moses MD, Hamilton, Matheson, David, Dorsey MD, E. Ray, George, Benjamin, Sadoff, David, & Yoshimura PhD, Satoshi. (2013). The anatomy of health care in the United States. JAMA, 310(18), 1947-1963.


17. Pediatrics, American Academy of. (2008). Bright Futures: Guidelines for Health Supervision of Infants, Children and Adolescents. In American Academy of Pediatrics (Ed.), (3rd Edition ed.). Elk Grove Village, IL.


18.  Peikes, D, Genevro, J, Scholle, SH, & Torda, P. (February 2011). The Patient-Centered Medical Home: Strategies to Put Patients at the Center of Primary Care.  AHRQ Publication #11-0029. from


19. Reid, RJ, Coleman, K, Johnson, EA, Fiahman, PA, Hsu, C, Soman, P, . . . Larson, EB. (2010). The group health medical home at year two: cost csavings, higher patient satisfaction, and less burnout for providers. Health Affairs, 29(5), 835-843.


20. U.S. Prevention Services Task Force. (2008). Grade definitions.   Retrieved December 16, 2013, from


21.  Van der Wees, PJ, Zaslavsky, AM, & Ayanian, JZ. (2013). Improvements in health status after Massachusetts Health Care Reform. Milbank Quarterly, 91, 663-689.


The Health Care Task Force was formed as a consequence of the controversy over the University’s “Take Care of Your Health” initiative that was introduced in July of 2013 and, in response to a substantial level of opposition by the faculty and staff, rescinded in September.  Those opposed to that wellness initiative articulated concerns in three primary areas.


The first was privacy.   One of the requirements of the wellness initiative was that faculty and staff had to participate in a biometric screening performed by an external provider hired by the University in conjunction with its third party administrator, Highmark.  Given that many people may find it difficult to speak frankly about their private health matters, even to a trusted family doctor, it is not surprising that to force one to submit to an assembly-line battery of tests performed in a public setting by complete strangers was viewed by some as unnecessarily callous.  Another requirement of the wellness initiative was that employees had to fill out an online questionnaire provided by WebMD.  The survey was not only long, but also intrusive, as it inquired about issues that most individuals would be quite reticent to discuss in any public forum.  While it may be advisable to have such an uncomfortable discussion with one’s family doctor, and necessary to share such information with one’s insurer who processes payment for treatment, some felt that it was another matter entirely to share the intimate details of one’s private medical history with WebMD, an entity known by most as a purveyor of free online medical advice.


The second concern was with process.  While it is common practice for those in the political arena to announce uncomfortable news on a Friday evening before a holiday weekend when nobody is paying attention, many of the faculty and staff at Penn State expect more forthright behavior from their University.  To announce major changes in something as important as the employee health care program, replete with significant financial penalties for noncompliance and without prior discussion or consultation in the middle of the summer when many faculty and staff are on vacation or otherwise not engaged, was widely viewed as unacceptable. 


Moreover, the financial penalties contained in the initiative--$100 per month for nonparticipation in the wellness initiative, $75 per month for those using tobacco products, and $100 per month for spouses enrolled in the Penn State plan who had access to healthcare from their employers—appeared to many to be arbitrary and coercive, not unlike those associated with traffic or parking tickets.  While there are strong economic arguments for incentivizing individuals not to engage in voluntary behaviors that increase their health care costs and which end up being borne by all plan participants, the magnitudes of these incentives need to be clearly tied to the actuarial costs to the plan of the behaviors in question.  If this analysis was performed, it was not communicated to the faculty and staff.


The third concern involved the anticipated effectiveness of the wellness initiative in lowering health care costs.   As noted in the Report of the Academic Research Subcommittee, there is a voluminous body of research examining the efficacy of a wide variety of wellness programs, and the results are decidedly mixed.    While the specifically focused disease management programs may be worth their cost, the evidence suggests that many of those whose goal is to change lifestyles are probably not.  These results are in the public domain and were well known to many in the University community prior to the University’s decision to implement the wellness initiative.  When pressed to provide evidence for the program’s likely efficacy, those speaking for the University cited three studies, two of which were authored by employees of Highmark, the program’s vendor.    To implement a costly, intrusive and unpopular mandate that the academic literature suggests would not attain the stated goals, and to justify the mandate by reference to potentially self-serving vendor studies, did not engender confidence among some that the necessary level of due diligence had been performed.


The unsuccessful roll out of the wellness initiative last summer provides a telling contrast to an earlier modification of the Penn State health care plan to include in-network deductibles and coinsurance.  Prior to 2011, the plan provided first-dollar coverage for in-network treatment, a level of generosity that was likely a contributor to the double-digit increases in the annual expenditures on health care that were being experienced by the University.   By this time, it was well known that exposing employees to part of the cost of consuming health care services through deductibles and coinsurance was an effective strategy to reduce overall health care expenditures, and this approach was being widely implemented in the public and private sectors. 


Recognizing the merits of this approach, and before implementing the plan change, the University embarked on a campaign of consultation and education, with town hall meetings and PowerPoint presentations designed to drive home the points (i) that Penn State was self-insured; (ii) that expenses for health care came out of the same budgetary pool as raises, travel, and research budgets, and (iii) that it was in the broad university community’s best interest to get the growth of health care costs under control before they consumed the entire budget.  Even though the changes in the health care plan resulted in a substantial increase in out-of-pocket costs to faculty and staff, the point was successfully made and the implementation of the new plan in 2011 with in-network deductibles and coinsurance went smoothly.   The bottom line is that the vast majority of the faculty and staff understand the underlying problem of increasing health care costs, and the opposition recently encountered was directed at the particular tool chosen—which was viewed by many as intrusive, coercive and punitive—as well as the manner in which it was chosen.


The other major change in the Penn State health care plan this year was the introduction of a qualified high deductible health care (“PPO Savings”) plan that is composed of a high deductible coupled with a health savings account to which the University contributes seed money.   That this did not generate substantial opposition could be a consequence of the cover provided by the vocal debate over the wellness initiative, so that many may have overlooked this change.  But this was also an option that was voluntary in nature, and one whose primary focus—a high deductible—was a natural extension of previous modifications to the University health care plan, and so it was not entirely unexpected.  Plus, those familiar with the academic literature on high-deductible plans knew that they were efficacious in reducing health care expenditures.


There has, however, arisen in the University community a substantial debate involving the salary indexing of premiums associated with the traditional, lower deductible (“PPO Blue”) plan.  Indexing of premiums to salaries has been a part of the Penn State health care landscape since 2011, and was presumably introduced for equity considerations: Employees at the lower levels of the salary scale were being hit with deductibles and coinsurance costs that were large relative to their incomes, and so the premium indexing spread some of the pain associated with the plan redesign to employees at higher salary levels as well.  While this policy may achieve some equity goals, and perhaps help to balance the University’s health care books, such indexing of premiums does nothing to reduce the growth of health care costs, since the premiums have already been paid and are therefore sunk costs to the employee when making decisions about the consumption of health care services.  


This year the indexing has been steepened, which has substantially increased the cost to higher-income employees of remaining in the lower deductible plan.  This is another example of a lack of consultation and communication, as this policy was simply announced in the summer 2013, bundled with the roll out of the wellness initiative and the noncompliance penalties.  A case can be made that to index premiums in this fashion encourages higher-salaried employees to opt for the high deductible plan, and that this could be an efficient way to bend the health care cost curve.  While this approach may have economic merit, the University has to date failed to make the argument cogently, effectively, or publicly.  And, the fact that only 16% of the workforce opted for the high deductible plan—approximately one-third less than what was projected—in spite of the fact that the current PPO Savings plan clearly dominates the alternative for higher income employees, provides yet more evidence that the level of communication on this issue has been deficient.


The goal of the Task Force in drafting this report is to provide the groundwork for an informed discussion between the various university stakeholders on the challenges confronting the Penn State community in the cost-effective provision of high quality health care.  To this end, the Report has presented detailed reviews of the academic literature on several polices of particular relevance to the University community:  High deductible health care plans (which are currently offered as an option), wellness initiatives (recently implemented, then rescinded), and tiered benefit plans (which are potentially on the horizon).  The Report also contains the results of a benchmarking survey with a relevant group of peer institutions, and an extensive discussion of the pros and cons associated with a wide range of potential policy paths forward, including guidance on the questions that would need to be answered and the data that would need to be gathered in order to make informed decisions on the merits.  The Report has also suggested guidelines for future communications on health care matters and, indeed, on all major OHR initiatives relating to employee benefits, stressing the importance of full disclosure, consultation and transparency in decision-making, sensitivity for privacy concerns, and providing a greater voice to employees. Finally, the Report has described, in some detail, the challenges that likely await the University as implementation of the Patient Protection and Affordable Care Act impacts the health care landscape.


Finally, as a land grant institution committed to both discovering and disseminating important scientific evidence to address pressing societal problems, Penn State has the opportunity to provide valuable lessons to other employers wrestling with the continued high cost of providing quality health care benefits to their employees.  As the analyses in this Report have illustrated, the evidence on many of the potential health care options available is mixed or incomplete due to a variety of factors.  Many of these factors are related to the research design employed, with most studies lacking the design characteristics or data availability to permit definitive answers to key questions, the answers to which are critical to informed decision making.  By recognizing the limitations in the existing literature, and incrementally testing different interventions, Penn State has an opportunity to lead the way in understanding what works in terms of more effectively providing efficient and affordable employment based health insurance.  In addition, Penn State Hershey has the opportunity to set itself apart from other health care providers, since it serves many other employers and health care payers, by using research, and clinical and behavioral science, to manage population health in a cost-effective way.  By taking a deliberate yet measured approach, and collaborating with the health and research expertise throughout the University, Penn State can provide significant benefits to society in this area.  

Appendix A:  Charge Letter*

Appendix B:  Task Force Membership*  

Appendix C:  Benchmarking Survey Results*

Appendix D:  Appendices to the Communications Subcommittee Report

(*If accessible documents for these .pdf appendices are needed, please contact the Senate office at 814-863-0221.)

D.I  Commentary by the Faculty Council of the College of Education


TO: Penn State University Health Care Task Force; Prof. Keith Crocker, Chair

CC: President Rodney Erickson; Prof. Brent Yarnal, University Faculty Senate Chair

DATE: January 16, 2014

The Faculty Council of the College of Education provides this public commentary on the health care issue that began with the “Take Care of Your Health” initiative. In response to President Rodney Erickson’s invitation to prepare commentary and consultation, the College of Education’s Faculty Council—an elected body of faculty representatives with the purpose of facilitating effective shared governance within the College—solicited comments and concerns from faculty and staff. A public town hall was held (October 9, 2013), members of the community were invited to provide comments to their representatives privately and via email, and online to the draft of this document. A good-faith effort was made to synthesize the collected feedback into this statement of our community’s principles and values that the Faculty Council believes are essential to protect in Penn State’s health care policies, as well as possible options that may be ethically considered to address the University’s fiscal considerations.

On behalf of all the faculty and staff who shared comments and concerns, the College of Education Faculty Council wishes to affirm that health care benefits provided by the University are a public good for many families in our community that must be based in transparency, fair treatment, equality, consent, and choice. The health care benefits vision represented by the “Take Care of Your Health Initiative” violates these basic rights, undermines the “shared risk pool” that is the entire purpose of insurance, and threatens the sense of trust between the University administration and its employees essential to Penn State’s stability and future vitality. Policies like those envisioned by “Take Care of Your Health” divide rather than unite and result in an adversarial employment climate. We found a hesitance among some staff and untenured or non-tenured faculty to make public comments that aren’t anonymous out of fear of potential retribution—which reflects how the actions of the University administration have eroded trust.

Penn State’s health care policies must treat all employees and their families fairly. This core principle is violated, for example, by the coercive and arbitrary $100/month penalty, without appeal, that the “Take Care of Your Health” initiative attempted to impose. We believe that any seizure of pay (or reduction of an individual’s benefits) by the University as employer without due process through an impartial appellate body threatens common standards of fairness and protection against arbitrary abuse. Due process and access to appeal become even more morally imperative when the University as employer seeks to coerce employee dependents too— even those not employed by the University. There is concern that if the University administration can coerce spouses and domestic partners that it could decide in the future to coerce employees’ children too.

Penn State’s health care policies must respect the principle of equality. This core principle is violated, for example, by the discriminatory and arbitrary $100/month penalty for employees with spouses or domestic partners who could receive any health coverage, even inferior and unaffordable, elsewhere. Differential treatment of employees based on the status of their spouses or domestic partners threatens the equality of all employees and hurts the most financially vulnerable employee households. This principle is also violated by the arbitrary $75/month against smokers. Differentiation of employees into favored and disfavored classes based on

personal lifestyle practices that are unrelated to job performance and not illegal threatens personal liberties, separation between work and private lives, and basic human dignity.
Targeting employees who smoke is problematic because of the “slippery slope” it represents. What are the limits to other private behaviors the University can choose to penalize in the future? Profiling employees based on personal characteristics unrelated to their job performance threatens the Penn State commitment against all forms of discrimination.

Penn State’s health care policies must respect the principle of consent and choice. This core principle is violated by punitive coercion of health practices, tests, or treatments—such as compulsory physicals and blood screening coerced by a $100/month penalty. It is further violated by punitive coercion to surrender personal medical data to the University or a third-party vendor like the WebMD survey and the ICH biometric screening. These unethical policies are a threat to the basic right to privacy and control over our own bodies. Mandatory or compulsory surrender of personal health data to an employer or any third party threatens the basic right to medical privacy guaranteed by law. The use or transfer of personal employee data by the University as employer (or a third-party vendor contracted by the University) without voluntary, informed consent is a threat to the most fundamental ethics of a research community and commitment to an ethical culture. Additionally, access to choice in health care is important. There is concern that the University’s continuing partnership with Highmark may disrupt employees’ access to Geisinger, the largest provider of full medical services in the area relied on by many employees—especially those with children needing pediatric care.

By permanently withdrawing the “Take Care of Your Health” initiative and abandoning the coercive and arbitrary measures it represented, Penn State can restore trust between its administration and employees. An adversarial climate of conflict is ultimately more costly to the University in lost efficiency and productivity than any long-term putative savings promised by the “Wellness” industry. To move forward, Penn State must restore trust and address the fear expressed by some faculty and staff that protest or appeal might be responded to with threat of termination, denial of tenure, or even veiled remarks to seek employment elsewhere if unhappy at Penn State. A University environment in which employees, particularly those employed at- will, feel afraid to publically protest against potential abuses or to seek redress through the system for fear of retribution threatens the essential mission of higher education to free inquiry and debate of ideas.

Recognizing that rising health care costs have fiscal implications for the University, there are possible cost-control measures that do not threaten the principles of fairness, equality, consent, and choice. Measures that subdivide the employee shared risk pool into favored and disfavored groups, target particular employees or households for discrimination, or coerce employees into surrendering their medical privacy and choice over their own bodies are unethical. Instead, the University should consider the range of possible measures that would be ethical:

     Transforming Penn State campuses into more healthful environments—such as completely Smoke-Free Zones

     Contributing to costs to employees to support wellness practices (e.g., low-cost healthy lunch options, membership fees to health clubs/facilities either on campus or in the community)

     Offering reasonable financial incentives for voluntary health behaviors supported by valid research—such as voluntary smoking cessation or weight management programs

     Increasing the premium contributions or deductibles of all employees to the shared risk pool at a rate commensurate with actual inflation—not an arbitrary amount like “$100/month”

     Altering the amounts of benefits for all participants in the risk pool

     Offering employee incentives for reducing costly medical procedures or overused services

    Changing or renegotiating with medical networks or health insurance providers However, all ethical measures would require the University to be more transparent about its fiscal condition and liabilities and to commit to using critically reviewed research to inform evidence-based policies on benefits. When the University administration simply insists it cannot afford employee benefit costs without providing transparent information, trust is jeopardized— especially as employees see the University able to spend large sums of money on other projects (like paying $100 to employees who had to submit to “Take Care of Your Health” or the “Penn State Lives Here” publicity campaign). We call on the Task Force to use the expert researchers Penn State has on its faculty to provide and review research germane to the issue, openly and transparently. In conclusion, the Faculty Council of the College of Education believes that respecting transparency, fair treatment, equality, consent, and choice in future benefits policies will restore trust and heal our Penn State community. We extend thanks to the President, to the University Senate, and to the members of the Task Force for giving units the opportunity to offer public commentary on the issue. Your time and consideration are greatly appreciated.

D.II   Focus Group with three University Park Faculty Members


Notes from the Communication Subcommittee on Our Meeting with Three UP-Faculty

January 22, 2014


J. Harwood, S. Sundar, and S. Johnson (unable to attend)




On Wednesday Shyam and John met with Brian Curran (A&A), Michael Berube (LA), and Mary Beth Oliver (Communications) for 90 minutes. The purpose was to gather their input on how communication about the “Taking Care Of Your Health” and the “Wellness Initiative” could have been improved.

We chose with these faculty because they were very visibly critical of the University’s efforts, both in content and communication.

We asked them to talk about 


1) Their concerns regarding PSU's health care rollout last Fall (a post-mortem of what went wrong in their opinion)


2) Important issues for the health care task force to consider while we deliberate on the best way forward for Penn State


3) Their ideas for charting a course forward for Penn State administration in terms of transparency and communication of its future initiatives involving faculty and staff.


This summary is not a transcription but a summary of the key points that we heard.

We have also attached documents that Brian Curran (chair of the newly formed AAUP chapter) gave us.




All three faculty pointed out that the events of last spring did not occur in a political or social vacuum. There was a lot of pent-up anger over several issues: the turmoil associated with the Sandusky affair (and the mandated training and testing program for reporting abuse); the constructive of the high-pressure pipeline through a residential neighborhood; and changes to the Bennett Center. In all three cases, faculty deplored the lack of transparency in reaching decisions and tone-deafness in communicating them to the communities that were affected. (Two of these issues affect only University Park faculty, staff, and students.) The pipeline and t