Physician Gainsharing programs reduce hospital spending

May 13 2008

Giving physicians cash payments for reduced hospital spending can help control costs without sacrificing quality or access to care, researchers report in a study released today in the May/June 2008 issue of the journal Health Affairs. The issue is a thematic volume on health reform supported by a grant from the Robert Wood Johnson Foundation.

In a five-year study of more than 220,000 patients, Arizona State University researchers showed that "gainsharing" programs, in which physicians are paid for reducing hospital spending, cut costs by 7.4 percent or $315 per patient. If these experiences are representative, nationwide use of gainsharing would cut hospital costs for coronary stent patients by about $195 million a year, the researchers say.

Jonathan Ketcham and Michael Furukawa, assistant professors at the Arizona State University School of Health Management and Policy, looked at six cardiac catheterization labs that implemented gainsharing programs and compared them to 123 non-gainsharing labs in 31 states. They examined the effect that financial incentives had on the cost of devices and drugs, as well as the volume of patients per physician in each hospital. They also measured whether gainsharing programs led doctors to select healthier patients.

They found that the majority of savings from gainsharing programs could be attributed to lower prices for coronary stents. The researchers found that these savings came without altering referral patterns via cherry picking, steering, or increased caseloads. They also found that the gainsharing programs did not increase the risk of in-lab complications and were associated with significant decreases in three specific types of complications.

"We found no evidence that gainsharing prompted physicians to avoid patients with existing health problems or pick the healthiest patients," said Ketcham. "Gainsharing reduces costs for coronary stent patients while apparently leaving quality and access unharmed," the authors write. Ketcham adds that analyzing gainsharing's effects on additional quality measures and understanding how gainsharing influences physicians' decisions would be valuable. The authors also note that widespread adoption of gainsharing could alter how device makers decide which types of products to develop and market.

Other issue highlights:

Study Shows No Malpractice Premium Crisis For Massachusetts Physicians. Although Massachusetts has the fourth-highest median malpractice settlement payments in the nation, and therefore should have the fourth-highest premiums, nearly all Bay State physicians paid lower inflation-adjusted malpractice premiums in 2005 than in 1990. The study -- which provides the most comprehensive analysis of premiums to date -- clashes with popular perceptions and assumptions underlying legislative proposals to cap damages awards.

Suffolk University Law School researchers Marc Rodwin and colleagues analyzed malpractice premiums from 1975 to 2005 using data from the state-regulated mutual insurer, known as Pro-Mutual. In 2005, malpractice premiums were $17,810 for the coverage level and policy type that physicians most frequently purchased, compared with $17,907 in 1990. Despite premium increases since 1995 or 2000 for all physicians, premiums were still lower in 2005 than in 1990, when they reached a 30-year peak. Mean premiums increased only in three specialties comprising 4 percent of physicians: obstetrics, neurology, and orthopedists performing spinal surgery.

The study furthermore documents increased variation among premiums paid within each practice specialty since 1990, when insurers began to adjust rates for each practice specialty by discounting low-risk physicians and surcharging those with high risks. By 2005, there was a threefold difference in premiums for physicians within OB-GYN, the highest-risk specialty, as a result of rate discounts and surcharges based on an individual physician's risk factors. As a result, although mean OB-GYN premiums increased significantly since 1990, nearly one-third of physicians in OB-GYN paid lower premiums in 2005 than in 1990.

The authors note that OB-GYN premiums are higher than for most other physicians because infants injured during birth sometimes require lifelong custodial care. They therefore recommend that patient safety and quality efforts should focus on OB-GYNs and the two other high-risk specialties to reduce injuries. When similar efforts were undertaken in anesthesiology in the 1990s, injuries fell dramatically, and premiums did as well.

The authors also propose alternative means to compensate injuries. For infants injured during birth, they recommend no-fault compensation systems such as those used in Virginia and Florida. Alternatively, they suggest shifting liability from physicians to hospitals for all injuries that occur in hospitals. Both proposals would reduce the malpractice premiums for high-risk physicians while still compensating patients.

Study Examines Relationship Between Health Care Spending And Physician Practice Patterns. Primary care physicians in high-spending health care regions see patients back more frequently for return visits, are more likely to recommend screening tests of uncertain benefit, and opt for more resource-intensive management strategies than physicians who practice in low-spending health care regions, a survey of more than 800 physicians found.

Researcher Brenda Sirovich of the Veterans Affairs Medical Center in White River Junction, Vermont, and colleagues measured local health care spending using Medicare data. They then compared the spending figures with the physicians' survey responses about how often they typically see patients back with well-controlled hypertension; whether they routinely screen patients for three types of cancer; and how likely they are to recommend tests, referrals, or hospitalization for patients with three common clinical conditions.

Primary care physicians in high-spending regions were more likely to recommend discretionary treatments than were physicians in lower-spending areas, the researchers found. Compared with physicians practicing in low-spending regions, those in the highest-spending regions would recommend an additional 80 hypertension follow-up visits per year, 14 screening spiral computed tomography (CT) scans (for smokers), 25 echocardiograms (for exertional chest pain), 24 cardiac care unit admissions (for end-stage congestive heart failure) and 29 gastroenterology referrals (for heartburn) per 100 patients seen. Physicians in high- and low-spending regions were equally likely to recommend guideline-supported interventions.

"Current policy efforts to improve the quality of care and address disparities in spending have focused largely on fostering adherence to clinical guidelines," the researchers write. "This study suggests that greater attention to clinical judgment-and to the local factors that are likely to influence physician practice-will be required," they conclude.