Skip to main content

Advanced Search

Advanced Search

Current Filters

Filter your query

Publication Types



Other Publication


Understanding Variation in Hospital Quality

Patients, like all consumers wanting to make sure they get the highest quality for their money, want to be cared for by the best hospitals. Commonwealth Fund–supported research, however, challenges the notion that there are "good" and "bad" hospitals. A widely cited 2005 Harvard University study in The New England Journal of Medicine found that, based on data reported to the Centers for Medicare and Medicaid Services (CMS) under the Hospital Quality Alliance (HQA) program, the quality of hospital care varies widely—not only by geographic region and type of hospital, but also across medical conditions within individual hospitals. We spoke to Ashish Jha, M.D., the study's lead author and an assistant professor of health policy at the Harvard School of Public Health, about why this variation occurs, how hospitals can improve their performance, and the power of patients to bring about change.

Your 2005 study of hospital quality discovered variations not only among various regions of the country, but within hospitals themselves. Why would a hospital score well for treatment of one condition, like congestive heart failure, but not for another, like pneumonia?
Ashish Jha: It has a lot to do with where hospitals have put their energies and attention. An interesting finding in our study was that both congestive heart failure and acute myocardial infarction care seem to go very closely together. This suggests that hospitals that have focused on cardiac care do well on all the different cardiac care conditions but may not put the same attention to other clinical services. It underlines the notion that hospitals are not necessarily naturally either all good or all bad.

What does the fact that there are no "good" and "bad" hospitals mean for consumers?
Jha: It makes being informed all that much more important, because you can't take word-of-mouth information or anecdotal evidence about one type of care and use it to make judgments about an entire hospital. I think it also means that consumers have a really important role to play in motivating change. They can use that information to go to their local hospital or provider and demand they provide better care in areas that are important to them.

How would people go about doing that?
Jha: Just by speaking to their doctors and nurses. They don't necessarily need to track down the hospital chief executive; they can ask their doctor, "Why don't we do better in pneumonia care? Why aren't we better on congestive heart failure?" I think when physicians, nurses, and administrators start receiving that kind of feedback, it will motivate them to put the processes in place to ensure more consistent care.

You found that geographic region had some bearing on hospital quality. Why is that important?
Jha: We found that Boston, as a local region, or the Northeast, as a broader region, tended to do better. It's not really known why regions like this do better than others. It may have to do with the culture of quality in certain places. People who train in one area generally tend to stay and practice medicine in that area. So, I think over time certain regions end up building a cadre of providers focused on providing evidence-based care. Other factors that people often think about, like having high-technology tools and hospitals that can do major invasive procedures, did not seem to make a difference. That's important because it signifies that a hospital doesn't need to be big or very wealthy. Good quality is achievable by everyone.

How can hospitals act upon these findings?
Jha: The biggest thing hospitals can do is to start paying attention to the care that is delivered. That means paying attention to the quality of care in areas being measured, and also in those not currently being measured. The next step is to put in quality improvement programs. And then the more complicated, but potentially really important, thing they can do is use health information technology. When hospitals have electronic health records and access to high-quality data, they'll really be able to focus on quality.

The Hospital Quality Alliance only reports on a few indicators of quality. Is that enough?
Jha: It's clearly not enough. The three conditions that we looked at [acute myocardial infarction, congestive heart failure, and pneumonia] are important and pretty common, and there's widespread agreement on treatment. But, at the end of the day, they represent only a small fraction of care in hospitals. And we've seen that how you do in one area does not necessarily predict how you'll do in others. So, the next step is to expand to other conditions and other indicators—looking at things like patients outcomes.

How about looking at indicators that measure safety, or patient-centeredness?
Jha: CMS and the Agency for Healthcare Research and Quality have developed a survey called the Hospital Consumer Assessment of Healthcare Providers and Systems, which is in the middle of being piloted and will be nationally administered this fall at almost every hospital in the country. The goal is to measure patient-centeredness and patients' experiences in hospitals. Safety is obviously extremely important. I haven't really seen any adequate national efforts to collect data on safety systematically across all hospitals.

Is the future of quality measurement one where we will have hundreds of indicators to look at, or can we conceive of an aggregated, composite measure of performance?
Jha: I don't think any of us know the answer to that right now. We have to start measuring more than what we do now, but that will lead to issues. How do you evaluate a hospital on 80 different measures? There's going to have to be a lot of methodological work, taking data from different sources and pulling them into composite measures.

Do you think that, in the future, patients will take this kind of information in hand before they visit a doctor or hospital?
Jha: I hope so. Especially in the context of health savings accounts and consumer-directed health care. It's interesting that this push is going on, but there's no way we can ask consumers to take on a greater portion of their health care without also providing more transparent data on what they buy with their dollars. As people pay more out-of-pocket, they will want more information. The most interesting thing is that we now have this national reporting system in which almost every hospital is participating. And it's information that's available to every single American [See].

What are your plans for future research?
Jha: We're examining different ways of grading hospitals. HQA is supposed to give you information about how individual hospitals perform. But there are other sources, like the Leapfrog Group, that make information publicly available. We're working on how consumers can use all these sources of information rationally to maximize their outcomes. Because as much as people care about these quality indicators, what patients really care about is having the best chance of surviving when they go into a hospital with a heart attack.

March 2006

Publication Details