Making hospitals pick up the tab for patients who are readmitted because of infections contracted in those facilities on a previous visit doesn’t seem to improve outcomes, says a study of a policy implemented by the Centers for Medicare & Medicaid Services several years ago.
“Effect of Nonpayment for Preventable Infections in U.S. Hospitals,” published in the Oct. 11, 2012 issue of the New England Journal of Medicine, says that the CMS had not been able to force hospitals to cut infection rates for central-catheter-associated bloodstream infections and catheter-associated urinary tract infections.
“Our models showed significant decreases in rates of central-catheter-associated bloodstream infections during the periods before implementation of the policy (4.8 percent per quarter) and after implementation (4.7 percent per quarter), with no measurable effect of the CMS policy on either the trend (incidence-rate ratio, 1.00; P=0.97) or the intercept (incidence-rate ratio, 0.95; P=0.75) in the post-implementation period versus the pre-implementation period,” the study says.
The CMS policy did not affect the catheter-associated urinary tract infections much either. The authors used data collected from 398 hospitals in 44 states from January 2006 to March 2011. “We found no evidence that the 2008 CMS policy to reduce payments for [the two types of infection] had any measurable effect on infection rates.”
The nonpayment initiative came about partly because of arguments that pay for performance doesn’t seem to work either. “… [D]espite widespread adoption of pay-for-performance programs by health plans over the past decade, the evidence that they improve patient outcomes, either in primary care settings or hospital settings, is mixed.”
They also question reports of P4P’s cost-effectiveness. “There are lingering concerns that pay for performance may lead providers to avoid the most seriously ill patients, which may mitigate any intended beneficial effect of these programs.”
In other words, P4P programs contain a perverse incentive.