Cost sharing for health care--whose skin? Which game?

n engl j med 358;4 www.nejm.org january 24, 2008 411 In this issue of the Journal, Trivedi and colleagues1 examine the effect of cost sharing on the use of screening mammography among women enrolled in Medicare managed-care plans from 2001 to 2004. Focusing on more than 350,000 women between the ages of 65 and 69 years, the authors show that cost sharing — either in the form of copayments (in which patients pay flat fees when they receive services) or coinsurance (in which patients pay a fixed percentage of the cost of those services) — reduces the number of women who undergo mammography. The effects are large relative to the modest cost burdens the plans impose. The authors estimate that cost sharing on the order of $10 to $20 reduces by 8% the proportion of women who undergo mammography. Their findings are robust, with similar findings in unadjusted analyses and in multivariable analyses adjusted for potential demographic and regional confounders. Likewise, they are consistent between their cross-sectional comparisons (i.e., between plans) and longitudinal comparisons (i.e., the change over time within a plan before and after cost sharing was instituted). Their findings are also broadly consistent with other analyses of cost sharing for cancer screening.2 The authors conclude that cost sharing should be waived for mammography, essentially because mammography is beneficial, and therefore reducing its use by imposing out-of-pocket costs is against the interest of public health.3 Their conclusion raises a challenging health policy question: How, if at all, should cost sharing be incorporated into the design of health insurance? Specifically, what is the best way to structure financial incentives so that patients use health care services wisely but not excessively — or, in colloquial terms, what kind of “skin in the game” best serves the interest of patients within the fiscal constraints of the health care system? 4