The stark difference in how medical care is delivered from one region of the country to the next is central to the national health reform debate. For many years, CHCF has participated in the funding of variation research, both by the Dartmouth Atlas of Health Care and other researchers. Below, we take a look at the implications and questions posed by this body of research and identify some areas ripe for further exploration.

• Dartmouth Atlas work. The Dartmouth research, led by John E. Wennberg, MD, MPH, and Elliott S. Fisher, MD, MPH, has been seminal in identifying and communicating the significant and unexplained variation that exists in medical care across the country. By examining the causes of such variation, actions can be taken to improve the quality of care (providing care that works), to reduce the cost of care (not providing care that has minimal benefit), and to make it more patient-centered. (The Dartmouth Atlas of Health Care)

  Are patterns of use in Medicare fee-for-service (FFS) predictive of those for other coverage types (Medicare HMO, Medicare, private HMOs, and private PPOs)? Though the Dartmouth work focuses on traditional, FFS Medicare patients, published studies have demonstrated that variation exists in other coverage types as well, although the patterns are not exactly the same. (Health Affairs Web Exclusive, February 2008, "Variations in Hospital Resource Use for Medicare and Privately Insured Populations in California")

• The California experience. In California, the Dartmouth research shows significant variation among and within regions. Areas with fragmented delivery systems — and an associated high level of competition — tend to have higher care utilization and more aggressive treatment patterns (as measured by more doctors, more inpatient days, and more specialist visits). An in-depth look at Los Angeles shows that significantly different treatment patterns can occur at hospitals a few blocks apart. (The Dartmouth Atlas of Health Care 2008, Chapter 4)

  CHCF-funded research by Laurence Baker, PhD, of Stanford University, examined regional variation in California for different types of coverage. Baker found the greatest variation among Medicare FFS and private FFS/PPO plans. HMO plans — whether Medicare or private — tended to show less variation. This leads to consideration of how other factors, such as the level of care integration, and perhaps financial incentives, are able to reduce utilization variation. (CHCF, Same Disease, Different Care: How Patient Health Coverage Drives Treatment Patterns in California, April 2008)

  CHCF-funded research found that practice patterns for certain surgeries vary dramatically from place to place. The data maps present elective surgery and other procedure rates across California relative to state averages and show widely varying rates in spite of data adjustments for characteristics of the population and other factors. While some geographic variation in surgical rates is expected due to differences in the prevalence of disease, much of the variation cannot be explained by illness rates. The consistency of results across all payers makes this a significant contribution to the literature documenting regional variation. The reports are based on estimates developed by Laurence Baker in collaboration with Maryann O’Sullivan, JD, and the staff and leadership of the Campaign for Effective Patient Care. (CHCF, All Over the Map: Elective Procedure Rates in California Vary Widely, September 2011)

• Factors of managed care. Physicians who practice in a managed care environment — and the patients who select such HMOs — may have different perspectives on aggressive treatment than those who opt for fee-for-service coverage. There may also be differences in the disease burden of patients choosing different types of plans. Those with chronic illness and comorbidities may prefer to stay in FFS Medicare.

• The CalPERS example. The California Public Employee Retirement System (CalPERS) fund has long been concerned that its aggregate health plan costs are much higher in Northern California than in Southern California. Yet the Dartmouth Atlas work shows that Sacramento providers are more efficient. How can Sacramento be a low-cost market for Medicare and a high-priced market for CalPERS? One explanation may be in price differences. Medicare prices are formula-driven and therefore more consistent within a region. However, health plans negotiate individually with each hospital, resulting in much greater differences in price per inpatient day. This means there may be more days of care provided in Los Angeles, but since each day carries a lower cost, it does not result in higher overall costs.

• Will accountable health care organizations help? To a large extent this depends on how closely aligned the physicians are and what financial incentives are at play. Dartmouth Atlas research found that even leading systems, such as the Mayo Clinics, continue to show variation in treatment among Medicare FFS patients, although the degree of variation lessened in locations where hospitals and physicians were more tightly affiliated. (The Dartmouth Atlas of Health Care 2008, Chapter 3)

  Laurence Baker's research with Kaiser Permanente in California found that although variation continued to exist between hospitals in Northern California and Southern California, the differences were not as great as those seen north and south with non-Kaiser hospitals. (CHCF, Same Disease, Different Care: How Patient Health Coverage Drives Treatment Patterns in California, April 2008)

• Higher hospital resource use and mortality rates. A recent CHCF-funded study by the University of California's five medical centers and Cedars-Sinai Medical Center in Los Angeles examined variations in medical treatment, cost, and patient outcomes among hospitals, and has some surprising findings to add to the national health reform debate. The study examined the wide variation in hospital utilization among Medicare beneficiaries with congestive heart failure. Researchers found lower mortality rates at facilities that used more health care resources, compared with those at hospitals that used fewer resources. The findings suggested that more resource-intensive care may prolong life among certain patients with heart failure. Further work is underway to examine quality-of-life issues and potential interventions to ensure care is beneficial and warranted. ("Looking Forward, Looking Back: Assessing Variations in Hospital Resource Use and Outcomes for Elderly Patients with Heart Failure," M. K. Ong et al., Circulation: Cardiovascular Quality and Outcomes, October 13, 2009)

• MRI scanners and surgeries. Magnetic resonance imaging (MRI) is a technology frequently used to evaluate low back pain, despite evidence that challenges the usefulness of routine MRI and the surgical interventions it may trigger. In a Health Affairs Web Exclusive, Jacqueline Baras and Laurence Baker analyzed the relationship between MRI supply and care for FFS Medicare patients with low back pain. More scanners appear to translate into more scans and more surgeries for enrollees. ("Magnetic Resonance Imaging and Low Back Pain Care for Medicare Patients," Health Affairs, October 2009)

Research referenced here is available through the links above.