Summary: While employers have long engaged in efforts to improve the quality of health care services, the range of quality improvement–related activities they have pursued has until recently been limited to promoting national accreditation, performance measurement, and targeted disease management. But some employers, typically working through regional coalitions, have begun to pursue improvement initiatives that may have a more lasting impact on cost and quality. These efforts involve benefit redesign, value-based purchasing, payment reform, and aggressive efforts to engage employees in their own care. Going forward, health reform provisions such as health insurance exchanges may strengthen the relationship between employers and the health care system, leading to more advanced approaches to quality improvement.
By Brian Schilling
As health care spending continues to rise, these costs place a major burden on U.S. firms by straining budgets and undermining international competitiveness.1 Some employers have responded by limiting coverage or eliminating health care benefits altogether. Just 69 percent of employers provide health insurance today, down from 80 percent in 1998.2 Others, including those committed to providing health insurance benefits, have become increasingly involved in efforts to improve health care quality as a means of controlling costs.
The notion that employers might use their purchasing power to help improve the quality and efficiency of the health care system was articulated as early as the 1960s by Kerr White, a professor at Johns Hopkins University. But for more than a decade thereafter, employer efforts remained loosely organized.
Then in the early 1980s, Walter McClure of the Center for Policy Studies, a think tank in Minnesota, proposed the deceptively simple "buy right" principle, which proposes that purchasers reward efficient, high-quality providers "by giving their beneficiaries the two things they don't have now: the means, and the incentives, to identify and choose good providers over poor providers."3
The idea caught on, most notably in Pennsylvania, where employers, physicians, hospitals, and insurers all embraced the concept. That support quickly translated into data-gathering demonstration projects across the state. The effort was credited with helping to bring health care inflation into line with general inflation.
McClure's "buy right" principles—focusing on cost and quality and giving consumers the information to assess both—served as a unifying philosophy for other employers and employer coalitions. As a result, by the early 1990s, employer efforts to improve health care quality began to coalesce around external accreditation and performance measurement, two concepts that focused on the macro (or health plan) level of the system. Accreditation, it was believed, would provide a certain assurance that health plans were properly designed and managed and were doing things such as credentialing physicians, addressing complaints, and following reasonable utilization review practices. At the same time, performance measurement was seen as a means of objectively comparing health plans on how effectively they did such things as immunize children, deliver prenatal care, or screen for breast cancer.
A number of large employers—General Electric, IBM, and Xerox among them—supported these types of activities, as did private foundations, which led to the creation in 1990 of the National Committee for Quality Assurance, the nation's first major health plan accrediting body.
Today, collaborative efforts to measure health care quality and find market-based strategies to promote improvement are on the rise (for summaries of the work of 10 different coalitions, see Regional Health Improvement Collaboratives: Essential Elements for Successful Healthcare Reform http://www.nrhi.org/downloads/RegionalHealthImprovementCollaboratives.pdf by the Jewish Healthcare Foundation). Among the more than 50 established regional business coalitions on health, about two-thirds are currently engaged in some sort of collaborative performance measurement and/or reporting effort.4 In some instances, employers are collaborating directly with providers to improve quality and reduce costs. For an example of such a program, see the case study in this issue of Quality Matters.
Employers are increasingly engaged in a range of innovative approaches to improving health care quality and controlling costs that directly address some of the underlying flaws in the health care system, including: pay-for-volume approaches to compensation; the failure to engage patients in their own care; barriers to accessing care; a lack of coordination among providers; and counterproductive benefit designs.
The Patient-Centered Medical Home
The patient-centered medical home (PCMH) is a delivery system model designed to enhance coordination by ensuring that each patient's care is coordinated by a single provider.5 While the model traces its roots back to 1967, it did not gain widespread support among employers until 2006, when IBM and other organizations started the Patient Centered Primary Care Collaborative to promote it. By 2009, some 500 large employers, insurers, consumer groups, and doctors had joined the group.
A large number of employer-led pilots of PCMHs have been launched nationwide since then and language promoting the model in the Medicare program was woven into last year's health care reform law. While the model is not easy to implement, its appeal lies in the increased continuity and intensity of primary care services it allows. Its advocates hope this focus will prevent or delay the onset of costly acute and chronic conditions.
PCMHs may be particularly helpful for the sickest patients. It is widely observed that the sickest 20 percent of patients consume 80 percent of all health care resources and that the sickest 5 percent of patients consume 50 percent of all health care resources.6
In a recent issue of The New Yorker, Atul Gawande, M.D., wrote about a PCMH-like initiative in Atlantic City, N.J., that serves two large, disparate groups of employees—casino workers and the staff of the area's only hospital. Both self-insured groups sought to lower their rising health care costs as a means of keeping their health benefits intact. Their solution was to establish the Special Care Center in 2007.7
The Special Care Center offers patients unlimited same-day access (for the acutely ill), no copayments, and a staff that is attuned to their needs. At daily meetings, all staff participate in a review of each patient's status, and outstanding issues, such as missed appointments or prescription problems, are assigned to specific individuals for follow up. A customized electronic information system tracks whether patients are meeting their goals. Eight full-time "health coaches" work with patients to help them navigate the system and adopt healthy behaviors.
The results, at least in terms of the health status of participants, speak for themselves: of 503 patients with high blood pressure, only two were in poor control. Among patients with high cholesterol, the average drop in LDL levels was 50 points. Sixty-three percent of smokers with heart and lung disease quit smoking. Consumption of health services dropped as well—emergency department visits and hospital admissions by over 40 percent and surgical procedures by 25 percent.
According to Gawande, an independent economist estimated that, among the highest-cost patients, the program cut costs by about 25 percent. That same economist, however, cautions that the results are still preliminary and based on a small sample.
Early reports from a dozen more formal pilot projects suggest that PCMHs save money and improve care quality. For instance, a Boeing PCMH pilot project that ran from 2007 to 2009 reportedly cut per capita health care costs among 740 high-cost/high-risk participants by 20 percent, primarily due to a reduction in emergency department visits and hospitalizations.8 Other measures of the health of program participants were similarly encouraging: physical functioning scores improved 15 percent; mental functioning scores improved 16 percent; and patient-reported work days missed declined by over 56 percent.9
But according to one of the pilot's chief architects, Arnold Milstein, M.D., M.P.H., medical director of the Pacific Business Group on Health, the jury is still out on whether medical homes not exclusively focused on older and/or sicker patients can actually reduce per capita health spending. It's not hard to see why: for a healthy person, it may not be cost-effective to introduce a more labor-intensive model of care coordination when there's little or no care to coordinate.
Another approach to improving health care quality that is popular among large employers is to bring health care in house by offering employees (and typically dependents) access to a workplace clinic. Nationally, only about 10 percent of American workers have access to such a health center, but at employers with 500 or more employees at a given site, 31 percent of employers offer a clinic of some sort.10
Besides start-up costs (which can easily top $1 million), privacy concerns are perhaps the chief obstacle to a successful workplace clinic. Some employers have dealt with this by ensuring an arms-length distance between the clinic and the employer and/or hiring a third-party vendor to manage the clinic.
For instance, Glatfelter, an Ohio-based paper company with 1,400 employees, established an independently managed workplace clinic in 2008 across the street from its production operation. Convenience was a major factor in establishing the clinics, says Gregory Paradiso, Glatfelter's director of compensation, benefits, and health.
To encourage Glatfelter employees to use the facility, copayments are half as much in the clinic as they are in the broader community. In addition, employees that use one of the staff doctors as their primary care physician (PCP) get an average $100 per-month discount on premiums. As a result, about 90 percent of employees at the Ohio location use the center and the majority have selected one of its two physicians as their PCP. "I believe our employees get better care at the center than they can get out in the community. Our doctors are salaried, so there's no incentive to overtreat patients or order tests that aren't necessary," Paradiso said.
Paradiso validated his assumption about salaried doctors ordering only needed care by comparing the in-clinic and out-of-clinic rates of referrals to specialists and inpatient admissions. "We consistently found that through spending extra time with patients, clinic rates for specialist referrals and inpatient admissions are half that of physicians in the community," he said. Paradiso calculated that this saves, annually, over $1 million in specialist fees and hospital costs. Paradiso also studied generic prescribing rates and found a similar dynamic: in-clinic rates were 78 percent versus 70 percent for community physicians. Based on typical prescription drug use among Glatfelter employees, Paradiso says that difference alone saves the company more than $500,000 per year.
Though peer-reviewed studies on the subject are sparse, various in-house analyses show that workplace medical clinics are a good investment. A Pitney Bowes analysis from 2006 found that the cost per episode of care was $276 at an in-house clinic, compared with $645 for treating a similar patient in the community. Another analysis of 15 employers' in-house clinics found an average return of $2.55 for every dollar spent to support an in-house clinic, but savings estimates do vary widely, depending on how they're calculated.11 When increased productivity and other factors are included, several employers found returns in excess of three-to-one.
An analysis of the Glatfelter Family Medical Center in 2009 found a roughly two-to-one savings. But there are undoubtedly also significant savings in terms of increased productivity, better health, and greater employee retention.
Consumer-Directed Health Plans
Employers are also increasingly turning to what are known as consumer-directed health plans (CDHPs)—essentially health insurance with a high deductible that is combined with a health savings account (HSA) and decision support tools—as a means of controlling costs. According to the U.S. Government Accountability Office, there were only about 3 million people enrolled in such plans in 2005. Last year, 22 million people were covered by CDHPs.12
The thinking behind CDHPs is straightforward behavioral economics: by increasing deductibles and/or copayments, employees will make more prudent choices about what care they need. Less clear is whether or not the CDHP model will, as some predict, encourage people to forgo needed care because "early" costs would come out of their own pocket due to the high deductible.
Perhaps the best known example of a company successfully shifting employees to a CDHP is Safeway, which began its effort in 2006. The national grocery story chain affected a near-100 percent changeover of its previous menu of health HMO and PPO plans in favor of a CDHP with a $1,000 deductible. Safeway also instituted a more aggressive formulary to encourage the use of generics and gave people an incentive to take a health risk assessment.
In the five years since then, overall costs for the CDHP-covered population have remained steady. Safeway estimates that it has saved somewhere between $150 million and $230 million as a result of the changeover. Some part of that savings is attributed to efforts by employees to select lower-cost services. Safeway caps the allowable reimbursement for certain procedures such as colonoscopies that, according to Safeway's survey of providers near its headquarters, cost anywhere from $900 to $8,000. Employees can use any provider they choose, but Safeway reimburses only up to $1,500. Employees who choose a more expensive site pay the difference.
Promoting Patient Engagement
Among the core issues in health care is the question of how to get people more actively engaged in their own care. The issue is significant—patients who are passively or only nominally involved in their own care can undermine their treatment in any number of ways: by missing appointments; by failing to refill medications; by not exercising or eating right; by refusing the support of disease management programs; or simply by ignoring obvious signs that a trip to the doctor might be warranted. Such disengaged patients can be expensive to treat or to employ.13
To encourage patient engagement, the Illinois-based Council 31 of the American Federation of State, County and Municipal Employees (AFSCME) in 2006 took an unusually direct approach to encouraging its 400 employees, retirees, and dependents to get involved in their own health care using a formal, written contract.
The contract required employees to fill out a health risk assessment annually, participate in comprehensive care management programs, establish relationships with primary care doctors, undergo regular biometric screenings, take regular e-learning lessons, and in time become more adept using interactive online tools to help manage serious health conditions. Employees were not required to participate, but the alternative is to pay substantially more for health care. Today, 90 percent of employees have signed the contract.
Since 2006, AFSCME's overall average monthly medical costs have gone down by about 10 percent. By contrast, health care costs nationally have gone up about 40 percent over the same period.14 As important, the overall health of the participant population is steadily improving—aggregate measures of smoking, hypertension, and cholesterol are all declining. At the same time, patient-reported health status is up.
Value-Based Benefit Design
As early as 1982, research established a clear link between consumers' out-of-pocket costs and their use of health care.15 As costs went up, consumption went down, and, importantly, vice versa. Leading employers did not fully articulate a benefit design strategy based on this finding for some time thereafter, but today are leveraging incentives and disincentives to discourage employees from seeking unneeded tests, point them to quality providers, and help them better manage chronic conditions through value-based benefit design.
A 2007 Mercer survey found that among employers with at least 10,000 workers, 80 percent planned to adopt this approach in the next five years. Among firms with 500 or more workers, nearly 20 percent already charge employees less for "high value" health care services, such as prescriptions needed to effectively manage chronic diseases.16
Whether or not employers deliberately move to adopt a value-based approach to benefit design, some aspects of it will be woven into future benefit plans. Last year's health care reform legislation incorporates value-based insurance principles, including the requirement that health plans provide free recommended preventive services such as mammograms and colon cancer screenings.
Work by the National Committee for Quality Assurance (NCQA) may soon give employers significantly more data with which to fine-tune their benefit plans. The organization recently added new "relative resource use" measures to its Quality Compass tool, showing how efficiently plans in a given market treat particular illnesses.
NCQA's relative resource use data are detailed enough to show why plans score the way they do: Are they especially good at controlling pharmacy costs? Efficient providers of inpatient care? Or do they effectively manage outpatient costs? Relative resource use data will show not only which plan in a given region is the most efficient at treating patients with diabetes, but also how efficient they are at different aspects of diabetes care such as pharmacy, inpatient care, and outpatient surgery. Employers will be able to use these data to design benefits and quality improvement efforts.
In 2010, about 600 health plans reported the relative resource use measures nationally for the first time and NCQA expects that many more will do so in the years ahead.
Coordination with Medicare on Payment Reform
With a budget of $468 billion in 2010, Medicare is by far the largest purchaser of health care in the U.S. and the only purchaser other than Medicaid (which under reform may ultimately exceed Medicare in total spending) with sufficient clout to effect real change in the health care system on its own. With last year's health care reform law, the Medicare program received a mandate to work toward overhauling the still prevailing pay-for-volume approach to provider and hospital compensation. The mandate comes in the form of a new Center for Medicare and Medicaid Innovation, which will—among other things—develop, test, and implement new payment approaches to streamline the adoption of concepts such as the medical home. Explicit in the legislation is that these efforts to revise payment practices should serve as models for commercial insurers.
Employers have taken note. In October 2010, seven large employer purchasers including one state joined together to launch Catalyst for Payment Reform (CPR), an organization that seeks, among other things, to "ensure that the private sector voice and experience is considered in corresponding changes to public sector payment." That is, the group plans to provide thought leadership and coordination on payment reform among large employers and to align with Medicare's payment reform efforts where its approaches are applicable and transferable to the private sector.
"The way we pay for health care in this country greatly contributes to waste, overuse, and high costs," said Suzanne Delbanco, Ph.D., executive director of CPR. "Revamping payment practices in the private sector and aligning with Medicare where feasible are both critical to bringing the cost curve under control."
CPR's efforts also include a nascent national payment reform scorecard which would track, among other things, what percent of payments to hospitals and physicians are performance-based. Currently, that amount is somewhere in the low single digits.
"The amount of talk today is disproportionate to action," said Robert Galvin, M.D., chief executive officer of Equity Healthcare, which manages the health benefits and costs for nearly 300,000 employees and dependents of 34 companies. "For all the papers, speeches, and conferences, less than 3 percent of payments to doctors and hospitals have anything to do with performance."
The Health Exchanges of 2014
Also looming over the health care landscape is the pending launch of 50 state health exchanges in 2014. The exchanges, mandated under the health reform law, were designed to offer tens of millions of Americans—primarily uninsured or self-insured individuals, plus employees of small firms—a new, streamlined way to obtain health coverage. But uncertain is whether mid-size or even large employers will drop coverage and force employees into the exchanges as well (employers with more than 50 employees would not be eligible to participate in the exchanges directly until, at least, 2017).
Surveys today show that not all employers are decided on whether to participate in the health exchanges. A slim majority (52%) of 1,400 employers surveyed in 2010 expect to continue with their current health plans in 2014 even if the exchanges offer competitive rates. But many others (33%) weren't sure and fully 12 percent expected to drop coverage altogether.17
Come 2014, both large employers and the exchanges will presumably have the same interest in promoting wellness and quality and could in theory be effective allies where these interests overlap. But the specific structure, regulatory role, and negotiating power of the exchanges will vary from state to state and employer participation will likely vary too—their impact on the market remains to be seen.
As costs continue to rise, it is inevitable that employers committed to providing health care benefits will pursue ever more creative and effective ways to improve health care quality and control costs. Increasingly, these efforts target the core, underlying problems with the health care system rather than simply measure and report on quality at a macro level. "I see a future in which employers will leverage their purchasing power to essentially make it hard for providers at all levels of the system not to adopt certain proven delivery system practices and accept new payment reforms," said Andrew Webber, president of the National Business Coalition on Health, an association of employer-based health care coalitions.
1) In 1980, the U.S. spent $253 billion on health care. Today, we spend more than eight times that amount ($2.3 trillion as of 2008). U.S. Health Care Costs at Kaiseredu.org, accessed 2/5/11 at http://www.kaiseredu.org/Issue-Modules/US-Health-Care-Costs/Background-Brief.aspx.
2) Henry J. Kaiser Family Foundation/Health Research and Educational Trust survey of employer health benefits. Reported in "Health Benefits in 2005: Premium Increases Slow Down, Coverage Continues to Erode," Health Affairs, September 2005 24(5):1273–80 and available online at http://content.healthaffairs.org/content/24/5/1273.full.
3) J, K. Iglehart, "A Conversation with Walter McClure: Competition and the Pursuit of Quality," Health Affairs, 1988 7(1):79–90.
4) Personal communication with Carly McKeon, director of membership and communications, National Business Coalition on Health; January 30, 2011.
5) The PCMH focuses on the following core principles: Every patient should have an ongoing relationship with a primary care doctor; a patient's primary care doctor should coordinate all his or her care; access should be streamlined, through open scheduling and e-mail communication; integrated, advanced data systems should support the care process and help reduce errors; patients should be active participants in their own care; and revised payment systems should reward primary care physicians for coordinating care.
6) R. E. Herzlinger, Market-Driven Health Care: Who Wins, Who Loses in the Transformation of America's Largest Service Industry, Basic Books, 1999; M. W. Stanton, "The High Concentration of U.S. Health Care Expenditures," Research in Action 19, available at http://www.ahrq.gov/research/ria19/expendria.htm.
7) A. Gawande, "The Hot Spotters: Can We Lower Medical Costs by Giving the Neediest Patients Better Care?," The New Yorker, January 24, 2011.
8) A. Milstein and P. Kothari, "Are Higher-Value Care Models Replicable?," Health Affairs, Oct. 24, 2009, available at http://healthaffairs.org/blog/2009/10/20/are-higher-value-care-models-replicable/.
9) Mercer's National Survey of Employer-Sponsored Health Plans, January 2009. Reported in Workforce Magazine and available at http://www.workforce.com/section/news/article/survey-work-site-clinics-gaining-popularity.php.
10) Corporate Research Group, Best Practices of On-Site Employee Health Clinics: Strategies for Success (New York: Corporate Research Group, 2008). There are no widely accepted guidelines about what size facility can support an on-site clinic. Some vendors suggest that firms with as few as 300 employees at a single location may generate the patient volume necessary to make a center practical, but 700 to 1,000 may be a more realistic number and, even then, only if spouses and dependents are also eligible to use the facility.
11) Employee Benefit Research Institute's 2010 Consumer Engagement in Health Care Survey.
12) National Survey of Employer-Sponsored Health Plans, Mercer 2010.
13) One recent meta-analysis of various studies found that workers with poorly controlled diabetes missed six more days of work per year than workers with well-controlled diabetes. Another found that direct treatment costs were $940 higher, per year, for diabetics whose illness was poorly controlled vs. well controlled. Although the numbers differ, this same pattern—of poor control costing more than good control—holds for asthma, heart disease, hypertension, depression, and alcohol abuse.
14) Note: 40 percent figure is based on a reported 7 percent increase in health care costs for 2007 as reported by Towers Perrin in the 2007 Health Care Cost Survey and increases in 2008 (9.9%), 2009 (9.2%), and 2010 (9%) as reported by PriceWaterhouseCoopers in Behind the Numbers: Medical Cost Trends 2010.
15) The RAND Health Insurance Experiment, 1982.
16) National Survey of Employer-Sponsored Health Plans, Mercer, 2007.
17) National Survey of Employer-Sponsored Health Plans, Mercer 2010.