News Flash: "Trickle-Down" Does Not Save Lives

This week the New England Journal of Medicine published an evaluation of Premier HQID, a Medicare pay-for-performance program that provides monetary incentives to hospitals to improve quality of care. Hospitals that perform well on certain measures get 1-2% bonuses and hospitals that underperform are penalized by the same amount. 

In the study, a huge number of hospitals, 252 in the Premier program and 3363 not, submitted data to Medicare on 33 quality indicators for common conditions such as heart failure, pneumonia and joint replacement.  Data from over 6 million patients  went into the analysis. 

The headline-making conclusion went like this:

"Tying financial incentives to performance, often referred to as pay for performance, has gained broad acceptance as an approach to improving the quality of health care....We found no evidence that the largest hospital-based pay-for-performance program led to a decrease in 30-day mortality. Expectations of improved outcomes for programs modeled after Premier HQID should therefore remain modest."

To me, and I imagine others who were not consulted when this program was planned, this finding comes as little surprise.  Let me explain why:

•  Financial incentives in the program did not necessarily go to the people who prevent hospital deaths. They went to hospital administrators who had the freedom to use the dollars as they wished.  So our tax dollars - make no mistake -  might have paid for nurse education or better staffing, and they might have bought a fancy new office suite for the CEO, the benefits of which would presumably trickle down to the little people.

• Almost all of the Medicare quality indicators concern processes of care, not patient outcomes.  Processes (like which drugs are given and when) are only thinly linked to patient outcomes because of the complexity of human beings (whether patient or health care provider) and big organizations like hospitals.  If the right antibiotic is given at the right interval before surgery, is there certainty that the patient will not have an infection post-operatively? Of course not. It helps, but the maddening and wonderful complexity of humans and organizations means the process can only boost the odds in the patient's favor, not ensure the desired outcome. In short, there are no guarantees.

All the hospitals in the study report their data on Medicare's Hospital Compare website; this characteristic was equalized because public reporting is widely understood as a way to improve hospital quality. However, this understanding is based upon assumption, not fact, and some emerging data suggest otherwise.

This new study's findings about pay-for-performance are not unique.  I did find some evidence of success of the Premier program on (surprise!) Premier's website, where one also finds the list of Premier and other health care executives as well as the cadre of physicians who comprise the Premier board of directors.  Again excluding the largest group of health care providersin the country.

Lest this post appear as a rant from a burned-out nurse, I will add that nearly 1 in 5 older adults discharged from a hospital is re-admitted within 30 days. About half of these patients saw a physician in the interim, so it appears doctor visits did not prevent readmission.  Moreover, errors in health care are deadly - to nearly 200,000 people a year. And studies from across the globe show that nursing care saves lives. We understand the scope o the problem, and we know at least part of the solution. When will we stop making up voodoo quality strategies and start implementing what works?