To the Editor Joshi et al1 claim that for the 3 target conditions of acute myocardial infarction, heart failure, and pneumonia, most of the decline in excess readmissions is due to regression to the mean (RTM) when applying the statistical model described by Linden2 and Davis3 to the excess readmission ratios (ERRs). Unfortunately, Joshi et al1 did not assess the plausibility of the underlying model assumptions. Davis3 pointed out that the underlying statistical model to estimate the effects of RTM requires that the distributions of the ERRs during the 2 evaluation periods be gaussian with equal means and variances. To check this assumption, we ran the Levene test for the equality of variances for each of the 3 conditions and found that the P values for both pneumonia and acute myocardial infarction were less than .01. This suggests that the model assumptions underlying the RTM effects estimation by Joshi et al1 for these 2 conditions are violated. Consequently, the magnitude of RTM’s effects reported by Joshi et al1 may be inaccurate.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Du C, Zhou G, Li S. More Considerations on Both Model Assumptions and Results Interpretations—Evaluating Readmission. JAMA Intern Med. 2019;179(11):1599. doi:10.1001/jamainternmed.2019.4688
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: