EVERY FEW YEARS a new health care structure emerges in response to market vacuums or the industry's excesses. In the 1970s, third party administrators arose to provide transactional and medical management services for self-funded employer health plans (Department of Labor Employee Benefits Security Administration) newly established under the Employer Retirement Income Security Act. The Affordable Care Act provides funding for Consumer Oriented and Operated Plans (James, 2013)—health plans exclusively owned and managed by their members rather than commercial health plans.
Worksite primary care clinics are an increasingly popular service for mid-sized and large employers interested in promoting better population-level health, reduced health costs, and increased productivity. In most current configurations, they sit in front of, are independent from, but influence use of the employer's health plan.
Their financial impacts accrue through 2 broad mechanisms. Most provide lower replacement costs for services—for example, office visits, drugs, labs and x-rays–-that previously were provided through the health plan network. But the bigger opportunity is to drive appropriateness, cutting through the current system's perverse incentives, and changing patients' care and cost patterns throughout the care continuum. Other authors have described similar models. In the New Yorker, Atul Gawande described (Gawande, 2011) how Rushika Fernandopulle produced better care at lower cost with groups of high-cost chronic patients. In the April/June issue of this journal, Jerry Reeves and Brian Kapp wrote (Reeves & Kapp, 2013) about their population health management successes using strong carrots and sticks.
The most progressive vendors in this space establish comprehensive primary care group (and possibly occupational) health services and then enhance that footprint with a range of management approaches that address a wide range of health care clinical and financial risks. This approach aligns with the interests of the patient and purchaser rather than the vendor and provides a fresh spin on the best health care management lessons of the past 35 years. Vendors with this orientation are likely to pursue two goals: (1) Facilitating better care and health for patients and (2) Protecting the financial interests of purchasers.
A core premise here is that a tremendous percentage of health care activity is inappropriate and has evolved to exploit patients and purchasers. A 2008 PwC study (PricewaterhouseCoopers, 2008) estimated that 54.5% of all health care expenditures (about $1.5 trillion annually in 2013) provides zero value, a figure that, working in the field, my firm's team has come to believe may be low.
After all, over decades, health care's players have devised scores of ways to extract large sums that they are not legitimately entitled to. Aside from the more well-known mechanisms—egregious unit pricing and excessive overutilization of lucrative services (like cardiac stenting [Stergiopoulos & Brown, 2012] and complex spinal surgeries [Deyo et al., 2010])—there are many more subtle and less well-appreciated devices. A few examples:
- Health plans buy stakes in pharmacy benefit management companies, significantly increase the pricing on generic drugs, use the margins as a revenue stream, and then tell their clients they are managing their costs.
- Health systems put primary care physicians into the field, but incentivize referrals into the system for diagnostics and procedures at much higher reimbursements than can be obtained in ambulatory settings.
- Health plans take on responsibility for management of high-cost cases but devote little real energy to oversight, enforcing higher costs.
- Health plans promote the “choice” inherent in yellow pages provider networks, failing to discriminate between high and low performing physicians and services, as though the right to go to providers that deliver poorer outcomes at higher episodic cost benefits patients.
The market opportunity, then, is to establish mechanisms that disrupt the structural supports of health care abuse and distortion. Employers seek solutions that, over time, improve quality and return more money than they cost.
KEY DESIGN ELEMENTS
The mechanisms listed below are not exhaustive but are intended to give a sense of changes in practice that can yield dramatic improvements in performance.
Outside fee-for-service reimbursement
Several clinic vendors have abandoned fee-for-service reimbursement as a way of demonstrating to their employer clients that they have stepped away from financial conflict in the delivery of care.
In this approach, all operational costs—for clinicians, medical and office supplies, drugs, labs, insurance, utilities—are passed through without a markup, and a per employee (or per member) per month management fee is charged. This means that clinicians have no financial incentive to deliver unnecessary care (or to deny necessary care). Clients judge performance, based on measurable changes to population health outcomes and cost, which means that the only incentive is devising mechanisms that drive appropriateness, not only in the clinic, but downstream throughout the continuum.
In other words, unlike most of health care, in which providers are financially incentivized to deliver as many products/services as possible, these clinic vendors are paid to manage a process. It is a profoundly liberating and potentially disruptive innovation.
Eliminating barriers to primary care access
If one is trying to optimize care and cost, then access is crucial. Conventional wisdom notwithstanding, it makes little sense to establish barriers to primary care access.
To that end, many clinic vendors have established clinic services that are “free” to patients. Patients using the clinics pay nothing for office visits, standard drugs, laboratories, or x-rays. This encourages participation and allows the clinic staff to engage patients, manage their care, and advocate for them throughout the health system. Strong incentives can translate to two thirds or more of employees and dependents participating in the first year, with rates becoming even stickier over time. These participation levels are crucial to get traction over larger health plan costs.
Clinical care led by physicians and supported by nurse practitioners
Many vendors offer clinics led by nurse practitioners, claiming that they are “cheaper but just as good” as physicians. In fact, there is evidence (Horrocks et al., 2002) that nurse practitioners can provide equal or better “routine care” than doctors. But physicians, with much deeper training, are better at identifying and managing complex patients, who typically generate the greatest costs. Some clinic vendors are migrating to a combination of physicians and extenders, segmenting risk and allocating patients according to clinical need, but with sensitivity to continuity.
Empowered primary care
Over the past 20 years, primary care's decreasing valuation by the American Medical Association's Relative Value Scale Update Committee in Medicare (Klepper, 2013), combined with commercial health plans' declining reimbursement, have materially altered primary care practice. Physician patient panels now routinely exceed 2500 patients, which translates to average visit times of 12 minutes or less. Rushed visit durations have doubled specialty referrals (Barnett, 2012) over the past decade, almost certainly fueling unnecessary diagnostics and procedures, with associated increases in patient risk and purchasers cost.
Stepping outside fee-for-service allows patient panels of 1600 patients and 20 minutes, changing this dynamic, with reductions in unnecessary specialty visits, outpatient diagnostics, and procedures. Patients are exposed to less unnecessary risk and purchaser pay less for improved outcomes.
Evolved health IT infrastructure
In addition to a fully-functional electronic health record, effectively managing population-level health requires systems that can seamlessly receive (and send) health risk/biometric profile data, medical/surgical claims records from health plans, medication claims records from pharmacy benefit managements, and utilization/disease management information from medical management vendors. Clinicians and patients can benefit from analytical tools that identify and segment risk levels of patients with chronic disease and those who are at high risk for an acute event. A host of other clinical data—for example, care gaps, other services accessed by the patient—can inform diagnostic and treatment plan decisions. Selected information from the clinical record should also be reflected into a personal health record that is easily available to the patient through a Web portal.
A BROADER APPROACH TO THE MANAGEMENT OF HEALTH CARE CLINICAL/FINANCIAL RISK
Employers investing in clinics do not merely seek reduced primary care costs, but lower overall health plan costs, accompanied by improved health status and productivity outcomes.
As important as empowered primary care is, it still is not adequate to address many of these issues. Any meaningful effort to address health care excesses must also focus on the avoidance of unnecessary care, on management of high cost cases, more efficient acquisition of high value products and services, development of data-driven high performance networks, strong patient incentives to adhere to approaches that are proven to work, and other mechanisms. Primary care provides the foothold, but the activity must then be extended to a variety of other clinical and business practices.
The following charts represent actual case studies, representing one clinic vendor's financial impact. (Quality outcomes impact will be covered in a separate article.) Employers range from 440 to 3000 employees and have had clinics deployed between 18 and 36 months. Notes on the methodology are as follows:
- The analyses show per employee per month cost over time, including medical/surgical claims and drug claims prior to clinic implementation, then adding clinic operational and drug costs post implementation.
- The analyses use a 12-month rolling average of per employee per month costs. This approach smoothes volatility in the claims experience, but may hide some patterns.
- All data are presented without catastrophic claims, or “shock losses.” In this case, shocks are episodic claims of $50,000 or more that accrue to an individual patient over 1 year. Unless a clinic vendor is allowed to manage high-cost claims, it is difficult to positively impact shocks in the first 3 years, since trauma cannot be avoided and chronic events have been percolating for years. After year 4, shock losses should decline, because of more aggressive chronic disease/lifestyle management.
- Cost data are framed against the Milliman Medical Index (Mayne et al., 2013) for 2010-2012, which averaged 7.2% nationally for total health expenditures for a family of 4, including employer contribution, employee contribution, and employee out-of-pocket expenses.
Case Study 1. Manufacturer–Union
Currently 1239 employees.
Clinic implemented April 2010, 37 months of postimplementation claims data.
Case Study 2. Assembly Plant
Currently 2956 employees.
Clinic implemented August 2011, 21 months of postimplementation claims data.
Case Study 3. Manufacturer
Currently 658 employees.
Clinic implemented November 2010, 30 months of postimplementation claims data.
Case Study 4. Local Government
Currently 440 employees.
Clinic implemented October 2011, 19 months of postimplementation claims data.
Note that, except for case study 3, all analyses show the same curve, with costs rising for 6 to 14 months postclinic implementation and then plummeting. This is due to a release of pent-up demand. All client groups have coverage with copays, so a significant portion of each population has health needs, but it avoids care for fear of cost. By contrast, patients come in droves to a free clinic. Unmet needs are diagnosed, followed by specialty visits, drugs and, sometimes, procedures.
In other words, the clinic vendor walks into a situation that has been festering for years and must work to get control of it. After some period of time, costs drop precipitously and consistently move to far less than projected using the very conservative Milliman Medical Index.
The same medical management model was used in all 4 cases, and it appears to work. The cost change curves are consistent across cases, yielding significantly lower cost. There are also improved quality outcomes, though those data are not shown here.
These financial impact data are the tip of the iceberg. Health care is brimming over with inappropriate care and cost. The model described here has demonstrated that cost can be lowered significantly with associated quality improvements. There is an opportunity to exploit these market vacuums much further, for great savings.
US health care cost drivers are a multiheaded monster, driving excess and inappropriate care and cost. The recent emphasis on primary care is well-founded, but not enough. Managing risk requires broad, multivector risk management, based in primary care, but extending out through multiple vectors. This approach can prove transformative and holds promise in any environment in which a health plan sponsor is at risk: for example, Medicare Advantage, Medicaid Managed Care, Indigent Care.
Barnett M. L.. ( 2012;). Trends in physician referrals in the United States, 1999-2009. Archives of Internal Medicine. , 172:(2), 163–170.
US Department of Labor, Employee Benefits Security Administration. ( 2013;). Understanding your fiduciary responsibilities under a group health plan. Washington, DC: Employee Benefits Security Administration.
Deyo R. A., Mirza S. K., Martin B. I., Kreuter W., Goodman D. C., Jarvik J. G.. ( 2010;). Trends, major medical complications, and charges associated with surgery for lumbar spinal stenosis in older adults. JAMA. , 303:(13), 1259–1265.
Gawande A.. ( 2011;, January 24). The hot spotters. The New Yorker. , pp. 41–51.
Horrocks S., Anderson E., Salisbury C.. ( 2002;). Systematic review of whether nurse practitioners working in primary care can provide equivalent care to doctors. BMJ. , 324:, 819
Mayne L., Girod C. S., Weltz S. A.. ( 2013;, May 22). Milliman medical index. Seattle, WA: Milliman Inc.
PricewaterhouseCoopers. ( 2008;). The Price of Excess: Identifying Waste in Healthcare Spending. Washington DC: PwC Health Research Institute.
Reeves J., Kapp B.. ( 2013;, April/June). Improved cost, health, and satisfaction with a health home benefit plan for self-insured employers and small physician practices. Journal of Ambulatory Care Management. , 36:(2), 108–120.
Stergiopoulos K., Brown D. L.. ( 2012;, February). Initial coronary stent implantation with medical therapy vs medical therapy alone for stable coronary artery disease: Meta-analysis of randomized controlled trials. Archives of Internal Medicine. , 172:(4), 312–319.