Medicare Pays More for Lower-Quality Surgeries

When patients were treated at high-quality hospitals, Medicare spent $2,700 less in the first 30 days than it did for patients at low-quality hospitals, researchers have found.

High-quality hospitals might cost Medicare more on the front end but they save money in the long term researchers have concluded.

Researchers at Harvard's T.H. Chan School of Public Health examined Medicare costs and outcomes data between 2011 and 2012 for five major surgical procedures—coronary artery bypass grafting, pulmonary lobectomy, endovascular repair of abdominal aortic aneurysm, colectomy, and hip replacement.

High-quality hospitals were identified by 30-day surgical mortality rates and patient satisfaction scores.

Beyond the initial $32,000 average cost for the surgeries, the researchers calculated costs of the procedures and post-surgical care at both 30- and 90-day periods among 110,625 and 93,864 Medicare beneficiaries, respectively.

When patients were treated at high-quality hospitals, Medicare spent $2,700 less in the first 30 days than it did for patients at low-quality hospitals, and $2,200 less at 90 days after accounting for all the differences in patient populations, according to the study, which appeared in Health Affairs.

Nearly two-thirds of Medicare's savings were driven by lower use of post-acute care services by patients at high-quality hospitals compared with those at low-quality hospitals.

Study senior author Ashish Jha, MD, MPH, spoke with HealthLeaders Media about the findings. Jha is a practicing general internist at the VA, a professor at the Harvard Medical School, and a member of the Institute of Medicine. The following is an edited transcript.

HLM: What prompted this study?

Jha: We have found that delivering high-quality healthcare costs more money. So we thought, "maybe surgery is different. When you have a surgical procedure and it goes well, because you are at a good institution that does a good job, maybe you end up saving a lot of the costs of complications that follow."

HLM: That was the motivation for this study: Does that turn out to be true?

Jha: Of course, there is one countervailing force. If you are a high-quality hospital, you are going to keep more patients alive.

That means there are some people who are really sick who would have died at another institution that survive at yours. Those people end up costing a lot of money.

That actually made us worry when we began this project: Would the costs associated with that swamp any savings we might get from being generally good quality?

FL Health System on Path to Save Millions by Standardizing Surgical Practice

We found that no, they would not. High-quality hospitals have fewer complications, they have fewer people coming back for readmissions, and less post-acute care services.

All of that ends up saving the system a lot of money. Of course, it is also much better for patients.

HLM: Your findings seem to support the old notion that "you get what you pay for."

Jha: Yes, and no. We are finding that it's better to be in a Mercedes than a Kia, but at the same time Medicare is paying both the same amount, and that is a problem.

Medicare will come back and say, we have some programs that on the margins award hospitals for these measures. But it's very little compared to the benefit that Medicare is getting.

Readmission Location Linked to Post-Surgical Mortality Rates

Medicare saves $2,000 to $3,000 every time a person goes to a high-quality hospital instead of a low-quality hospital. [But] let me assure you that the high-quality hospital isn't getting paid $3,000 extra for that procedure.

HLM: Were you able to factor out the socio-demographics of the patient mix?

Jha: We were worried about that, so we adjusted for a series of things, including the underlying patient population differences between the high- and low-quality institutions, just to make sure that what we were picking up was something different than a population that is served by these hospitals.

We did find that low-quality hospitals tended to have more poor patients and we wanted to make sure what we were picking up was really quality and not poverty. So we adjusted for that the best we could. Is that adjustment perfect? No. Does it get at most of that difference? I think it does.

HLM: How did patient experience play a factor in this study?

Jha: There is now good evidence that institutions that have good patient experience tend to do a lot of things well.

If you think about patient experience questions , [such as about how doctors and nurses communicate and respond to a patients needs and issues,] institutions where physicians and nurses are communicating well with patients are probably also communicating well with each other.

They are part of an institution where communication and information flow is generally quite good. That probably helps out in terms of making sure that complications are identified early.

HLM: Are there any alternative explanations for your findings?

Jha: Those issues of underlying patient population differences are still out there and they are not fully settled. The only way to fully settle those issues is to do a randomized trial. Send some people randomly to high- and low-quality hospitals and see what happens. Well, that is not going to happen.

Barring that, we are left with these other techniques that don't get us all the way there, but get us a large chunk of the way here. It's hard for me to imagine that it is just patient population differences that explain these fairly sizable differences in spending.

HLM: What would you like to see done with these findings?

Jha: This study should be a strong impetus for Medicare and other payers to reward high-quality institutions more than giving them a tiny bonus or penalty.

We should be rewarding high-quality institutions with more financial resources, and at the same time, really penalizing low-quality institutions, because not only do people do badly there, they end up spending more money.

The system should take that into account. Right now we are paying the same for a Mercedes versus a Kia.