Citation

Ertle AR. Ann. Intern Med. 2020; 173(12): 949-955.

Copyright

(Copyright © 2020, American College of Physicians)

DOI

unavailable

PMID

unavailable

Abstract

I read the article by Ranney, et. al. with interest. Their approach is an episode-of-care analysis even though they do not call it that. This methodology was first pioneered by Schneeweiss and Rosenblatt in 1983 where they called it a diagnosis cluster.1 There has been ample research and numerous software development efforts to create complex episode groupers (e.g., Cave Grouper), but this is fraught with issues and limitations. For example, one can include all respiratory infections in a group, but a common cold is far different than pneumonia. Likewise, should a gunshot to the leg with a.22 caliber handgun be grouped with a purposeful gunshot to the head? The authors point out that it is also hard to know for sure which claims are related to the firearm injury or to something else that happens to temporally coincide. They also point out that it is difficult to define the end of the episode, and they chose an arbitrary six-month post-injury cost analysis. But what is more important to me is what new information their article might convey. As the authors stated, not surprisingly healthcare costs for both the insurer and the patients and family go up. The same analysis and findings could be performed for car accidents, skiing accidents, cancer diagnosis, or virtually anything else where one's health suddenly changes for the worse. The burden of everything we do, and even just aging, on our healthcare system is large.

DOI: 10.7326/M20-0323 see comment section

Language: en