Last night, I posted on a study that appeared in Health Affairs last month showing that Medicare patients typically receive a lower quality of care within hospitals than privately insured counterparts. I now have access to the study, so I’m going to share a little more detail on it here tonight.
For one thing, the authors of the study affirm my suspicion that some of the difference in care received by Medicare beneficiaries versus beneficiaries of private insurance is due to the fact that many physicians do not accept Medicare:
There may be differences in the characteristics of physicians who predominantly serve patients with a certain insurance status. A study of hospitals in Florida has found some evidence that, compared to other patients in the same hospital, uninsured and Medicaid patients are treated by lower-quality physicians.
Patients in the same hospital may not be served by physicians with the same level of experience and expertise…some patients are more likely than others to have surgery performed by slow, less experienced surgeons.
Compared to patients in the other payer groups, privately insured patients may be more likely to have physicians who are able to monitor their care more carefully, perhaps through improved electronic health record systems or more personalized treatment, which could lead to lower mortality rates. These patients may also receive newer, more expensive treatments than comparable publicly insured and uninsured patients.
I mentioned this in last night’s post but again, for the record, I am not necessarily going to draw conclusions one way or the other as to whether or not the physicians who accept Medicare and Medicaid patients somehow are of a “lower quality” than those who do not. That’s not an issue I feel particularly qualified to comment on, and both categories are so broad that it’s hard believe there’s not a ton of variance in there anyway, and I can think of a few confounding variables just off of the top of my head. That being said, it’s also not unreasonable to assume that there may be some class differences on average between the physicians who accept Medicare and Medicaid patients and those who do not.
The study, which looked at data from a total of 1,434 hospitals in 11 different states, also found that patients who had private insurance had lower “risk-adjusted mortality” for all of the procedures considered–including esophageal resection, pancreatic resection, hip replacement, abdominal aortic aneurysm repair, coronary artery bypass graft, carotid endarterectomy, craniotomy, and percutaneous transection of the coronary artery. At the same time, Medicare patients had higher-than-expected mortality for several different medical conditions–acute myocardial infarction (heart attack), congestive heart failure, stroke, gastrointestinal hemorrhage, hip fracture, and pneumonia.
The story, however, is not black and white. Although in many cases patients with private insurance did much better as a group than patients with Medicare, some procedures gave a slight edge to those who were not privately insured. For instance, in the case of stroke, pneumonia, and congestive heart failure, beneficiaries of Medicare and Medicaid, as well as “low-payment groups,” actually did better than the privately insured. At the end of the day, however, it seems that privately insured patients fared best overall in the study:
Our study suggests that within-hospital differences in quality exist across payer types. In particular, patients with Medicare appear to receive notably worse care than patients with private insurance on the majority of Inpatient Quality Indicators.
Is the conclusion that should be drawn from this study that private insurance is somehow definitively “better” than Medicare and Medicaid? No, it’s not. (That may or may not be the case, but this study doesn’t “prove” anything, one way or the other). The take-away point here is that insurance type does seem to make a difference as far as risk-adjusted health outcomes go, and that it’s unclear to which extent that’s a result of the insurance form itself versus the different approaches that are inspired by the form of insurance as some sort of signaling mechanism.
In other words, in those cases where Medicare beneficiaries did worse than the privately insured, is it because something about Medicare directly failed them, or is it because the very nature of being a Medicare patient negatively impacts how a hospital and its team of physicians received and treated them? I don’t think there’s a clear cut answer, but it’d be prudent to keep these sorts of issues in mind when we talk about expanding programs like Medicare and Medicaid.