The 2008 Medicare policy to withhold payment for treating certain hospital-acquired infections failed to decrease infection rates in U.S. hospitals, according to a report published online Oct. 10 in the New England Journal of Medicine.
In a study involving 398 hospitals or medical systems across the country, implementing a Centers for Medicare and Medicaid Services policy of nonpayment for the treatment of preventable catheter-associated bloodstream infections and catheter-associated urinary tract infections appeared to have no impact at all on the acquisition of those infections, according to Dr. Ashish K. Jha of the department of health policy and management, Harvard School of Public Health, Boston, and his associates.
Dr. Ashish K. Jha
"As CMS continues to expand this policy to cover Medicaid through the Affordable Care Act, require public reporting of National Healthcare Safety Network [NHSN] data through the Hospital Compare website, and impose greater financial penalties on hospitals that perform poorly on these measures, careful evaluation is needed to determine when these programs work, when they have unintended consequences, and what might be done to improve patient outcomes," Dr. Jha noted.
Dr. Jha and his colleagues assessed data from the NHSN, a public health surveillance program for monitoring health care–associated infections across the country. A total of 1,166 nonfederal acute-care hospitals report their infection rates to this Centers for Disease Control and Prevention–sponsored network every month.
Dr. Jha and his colleagues assessed NHSN data on three types of infection at 398 of those hospitals in 41 states. They examined central catheter–associated bloodstream and catheter-associated urinary tract infections because these are the two hospital-acquired infections for which CMS currently does not pay. They also looked at ventilator-associated pneumonia, which is not targeted by the CMS policy, as a control.
Rates of central catheter–associated bloodstream infections were already decreasing at the time the CMS policy was implemented, likely because the federal government, national organizations, and accrediting agencies had already focused attention on preventing these nosocomial infections. The rate of these infections was 4.8% per quarter before the policy was implemented and 4.7% afterward, a nonsignificant difference, the investigators said (N. Engl. J. Med. 2012 [doi:10.1056/NEJMsa1202419]).
This pattern also was seen with catheter-associated UTIs, in which there was a small, nonsignificant increase in the infection rate after implementation of the CMS policy. For the control condition of ventilator-associated pneumonia, the infection rate was 7.3% before implementation and 8.2% after implementation of the policy, also showing no significant impact on infection rates.
These findings were consistent across all hospital types, regardless of size, regional location, type of ownership, or teaching status.
To assess whether any benefit of the nonpayment policy may have been offset by strategies to lower infection rates, such as mandatory reporting, the researchers performed a separate analysis involving only the hospital units located in states that didn’t have mandatory reporting. Again, no demonstrable effect on infection rates was seen.
To allow more time for hospitals to adapt to the policy change, the investigators performed a sensitivity analysis comparing infection rates 2 years after implementation with those before implementation. Again, they found no further decreases in the rates of any infections.
A possible explanation for these findings is that the amount of this financial disincentive was quite small. "Reductions in payment may have been equivalent to as little as 0.6% of Medicare revenue for the average hospital," Dr. Jha and his associates said. "Greater financial penalties might induce a greater change in hospital responsiveness to the CMS policy."
The study results are particularly important given the increasing use of financial disincentives to improve the quality of healthcare. There is very little evidence that this strategy, or other pay-for-performance strategies, actually improves patient outcomes, they said.
This study was supported by the Agency for Healthcare Research and Quality. No financial conflicts of interest were reported.