When the quality improvement movement first started in medicine, I was all for it. The principle that you can’t fix what you can’t measure resonated with me. I felt that individual physicians and systems should be accountable to the people paying the freight and our patients.
Doubt should have started creeping in my mind when experts on QI in medicine started saying that we shouldn’t measure outcomes such as injury and death rates. A physician’s group or clinic didn’t have enough cases to enable a valid comparison, much less an individual physician. Therefore we should measure processes, because if the processes are better the outcomes must improve. Or at least that was the assumption.
A study of the VA system reminds us this is not necessarily the case. The VA has spent billions of dollars on electronic medical records and other institutional and cultural changes in the last few decades. It claims to provide high quality care, because it has data showing that regular VA patients are more likely to receive appropriate services such as measurements of cholesterol and long-term sugar levels for diabetics. According to this recent analysis, none of this ultimately mattered. In spite of spending more for these processes, the overall mortality rate was no different between VA and other patients.
A few other studies outside the VA found similar realities. One recent study of the time it takes to rush a patient having a heart attack back to the cath lab to open up the blocked artery (door-to-balloon time) found that among hospitals in Michigan, the times improved significantly, but the mortality rates did not improve. Another commentary showed that improved process measures among Medicare patients hospitalized with heart failure has not led to improved 30-day readmissions or death rates.
This blog post is not intended to be a comprehensive review of this subject of quality improvement in medicine. I know there are other studies that have found a connection between process outcomes and actual health outcomes that patients care about. However, as politicians and the business community continue to push for accountability in what they pay for health care, these studies should serve as a reminder that measuring things that are easy to measure and creating report cards from that information often are merely exercises in pleasing the pencil pushers and bureaucrats.
I’m also concerned that the quality improvement movement has created more administrative waste. Employees at both healthcare facilities such as hospitals and the payers spend their time collecting and reporting data and don’t directly care for patients. When people talk about the high cost of administration in healthcare, I think many of these quality reporting efforts contribute to this problem.
The unfortunate reality is that healthcare is often too complex to be reduced to a few simple measures, especially in primary care. I don’t mind being held accountable to the people who pay an awful lot for healthcare, but we should all humbly accept that often a clean straightforward measure of quality for many aspects of healthcare delivery simply do not exist.