A shift in hospital-physician affiliations

Many more physicians now work directly for hospitals. What does this mean for them?.

Back in 2003, the typical U.S. hospital didn't have any physicians directly working for it. More than 40% of hospitals were structured this way, while 29% employed physicians and 27% had some kind of looser contractual affiliation.

That situation has changed significantly, according to a study published online by Annals of Internal Medicine on Sept. 20. As of 2012, 42% of hospitals employed physicians and 38% had a contractual relationship, while only 19% had no employment relationship with any physicians. In addition to measuring changes in physician and hospital affiliations, researchers also tried to determine whether these trends had affected patient care. They found no significant effects.

To understand what these findings mean for physicians, hospitals, and patients, ACP Hospitalist recently spoke to study coauthor Ashish K. Jha, MD, MPH, a general internist and professor of medicine and health policy at Harvard University in Cambridge, Mass.

Q: What motivated this research?

A: What we've heard anecdotally is that over time, hospitals have increasingly started employing physicians. Physicians have always had a very interesting arms' length relationship with hospitals...where physicians went and applied their trade and were really independent, but that relationship has been changing. People who are advocates...say this will be good for patients: As hospitals begin to employ physicians, they will be able to get physicians to change clinical practice behaviors and really enact changes that will lead to better patient outcomes. We wanted to test 2 questions: Are we really seeing an explosion of these tight relationships between physicians and hospitals, and if we are, is this good for patient care? We found that while it was true that tight affiliation—the hospitals increasingly employing physicians—has taken off, it has had no impact whatsoever on patient outcomes.

Q: How did those findings compare to your expectations?

A: Certainly the advocates of this new model have argued that it's going to be good for patient care. I assumed on some level that it probably would have an impact. I was surprised that there was absolutely no effect whatsoever. It's an important reminder that at the end of the day, for most physicians, whether they're affiliated or employed or independent, they're making their best efforts to try to take care of patients. Simply moving from an affiliated to a tightly employed relationship doesn't alter the dynamic or the kind of care that patients receive in hospitals.

Q: Your study didn't analyze physician employment by specialty, but do you think the physicians moving into hospital employment are hospitalists, primary care, or subspecialists?

A: I think it's all of them. Some of it is hospitalists, certainly. Increasingly what we have seen from the data, from other stories out there, is hospitals buying up primary care practices [and] specialty practices. A physician who might have been an independent primary care physician who came in and took care of her patients in the hospital, may now be employed by the hospital. I think that too is happening, and our findings reflect that as well.

Q: What should physicians who are considering these employment options take from this study?

A: If you're a physician who is thinking about this, you're doing it primarily for whatever effect it has on your finances, whatever effect it has on your business or your practice, but you shouldn't do it for the patient care. There's no reason to think that this will either have a beneficial or a negative effect on patient care.

There are other studies [finding] that when this kind of employment relationship happens, prices on the private side tend to go up. Hospitals, once they begin employing physicians, can often demand higher prices from insurance companies, just because they have more market power.

Q: Are there other potential consequences of these affiliations that should be considered?

A: This is the first time in modern American history that hospitals employing physicians is the most dominant model of hospital care. Understanding the repercussions of that, not just in terms of short-term quality and cost but in the long-term impact on physician autonomy and physician happiness, is going to be very important, and I think we need to continue to monitor this trend. We need to ask ourselves if there is a trend that is reducing physician autonomy, leading to higher prices, and not having better quality, is that a good thing? Should we really have policies that encourage that? Should we let that happen, or should we try to think about how to clamp down on that? That's a policy question.

Q: Do you expect such concerns to affect the observed trend toward physician employment?

A: Some of this trend has been going on because people have been arguing it's going to improve patient care. Hopefully this study offers a little dose of reality on that. Whether that alone slows down this trend or if the fact that hospitals can actually drive prices much higher will continue to push the trend forward, I don't know.

Q: What else should physicians take from this study?

A: For me, it's a reminder. The whole model of why employment would help is based in the thought process that somehow physicians are not adequately motivated to work on quality and maybe being employed by the hospital can push them to become motivated. I think studies like ours could hopefully dispel some of that mythology. Improving quality and improving outcomes is difficult and just switching your affiliation isn't going to do the trick. It's really the hard work of changing the way you deliver health care.