This story summarizes an important study whose findings could affect the way rehospitalizations are handled. It suggests that returning to the same hospital where a surgical procedure was performed has tangible clinical benefits for patients.
But the story focused on the study’s more sensational number — a 26 percent mortality drop — while making no reference to the more conservative 8 percent figure that emerged after the authors attempted to reduce confounding error.
It also missed an opportunity to go further than the study did by providing some estimate — even an imperfect one — of the dollars and lives that might be saved if the study findings were confirmed and the authors’ recommendations were put into practice.
Efforts to improve the quality of care provided in the hospital include reducing readmissions. It is known that these are costly and may represent inadequate care during the hospital stay or post-discharge planning. Indeed, Medicare now has penalties for hospitals that have high readmission rates. It is in this context that the current study suggests that when readmissions occur after surgery, whether the readmission occurs in the same hospital or not is associated with the likelihood of dying. The 26% lower death rate reported in the story probably overstates the benefits of returning to the same hospital. That’s because the comparison reflects some level of bias. For example, if the patient is too sick to travel across town to the original hospital, the ambulance may be forced to take that critically ill person to a different hospital — the death isn’t due to the other hospital but to the seriousness of the patient presenting there. Nevertheless, even a small reduction in death rates with readmission to the same hospital — if shown to be real — represents an important finding. The challenge is what policy levers to use to encourage readmissions to the same hospital. This is not necessarily an easy task.
Alas, though the story contains some interesting information, it does not explore or speculate as to what the cost would be of returning most patients to an index hospital. It’s possible there would be downstream savings.
Interestingly, in the U.S. hospitals may face a penalty from Medicare if they have high readmission rates. In that sense, they could have an incentive to discourage patients from coming back to the same institution.
The story describes the authors’ finding of a 26 percent drop in 90-day mortality.
But it fails to mention that the authors’ instrumental variable analysis — intended to reduce confounding — yielded just an 8 percent drop in mortality risk for those patients who were readmitted to the original hospital. Though less dramatic, it is still a potentially important finding.
The story makes no mention of any downside that might be caused by returning patients requiring rehospitalization back to the hospital where the original work was done. It’s conceivable that policies encouraging people to go back to their original hospital could delay treatment of problems that can get worse without immediate care. For example, if someone travels for a procedure, getting back to the hospital for readmission could take longer than going to a local hospital. The story does note that going to the original hospital often results in more timely care, but the flip side was worth mentioning, too.
The story gives a nice description of how the study was performed. However, it doesn’t cite any potential limitations to the work, and thus may give readers a misleading impression of the study’s strength. Here are a few limitations that were acknowledged by the study researchers and editorialists:
Disease mongering was not an issue.
The story included a quote from the editorial that accompanied the study. But this quote is not very informative.
“Patients readmitted after surgery almost always have a postoperative complication, either medical or surgical.”
The point of this criterion is to encourage critical evaluation of the research, so we’re hesitant to give credit for a quote that’s mostly superfluous. The editorialists did comment on some of the important limitations raised in the quality of evidence domain above, but the story didn’t use any of that text.
The alternatives here are going to the same hospital or not. The piece emphasizes that this may be important, so we’ll give credit. The story could also have attempted to address what to do about these findings. What are the implications? The story mentions the ambulance driver or the patient traveling to another city for a procedure at a big name facility. The article implies that this situation could be changed, but doesn’t state how. This would require an initiative where insurers and payers decide that this is something important enough to address through policy changes.
As noted above, seeking to ensure that patients who have undergone major surgery would be readmitted to the hospital where the initial surgery was performed would appear to be logistically difficult, at best. The story did not address that concern.
The story doesn’t establish whether this is the first study of its kind or whether it builds upon previous research. The study authors noted that they found no prospective or retrospective studies that were applicable to the subject and were published between 1990 and February 1, 2014. But they also say that, while their data were being analyzed, two observational studies were published — containing conflicting results. The story doesn’t provide this context.
The story also could’ve met the standard here by noting that there are policies in place that attempt to reduce readmissions. So the idea of focusing on readmissions as a way to improve care is not novel, although this study provides important new data on the issue.
The story includes quotes from a telephone interview, so we can be sure it went beyond any news release.