|
Abstract
|
The study by Holland et al1 reports trends in emergency department (ED) visits involving mental health and substance use, including deliberate self-harm, unintentional overdose and other injuries, and assault, before and during the COVID-19 pandemic. The authors report 2 metrics for each time period, using data from the National Syndromic Surveillance Program (NSSP) at the US Centers for Disease Control and Prevention (CDC): the absolute number of ED visits of a given type and the rate of ED visits of a given type per 100 000 ED visits of all types.
We share the authors' interest in assessing the possible effects of the pandemic on behavioral health and injury outcomes. But we find it difficult to draw clear inferences from the present findings for 2 main reasons. First, as the authors mention briefly, the number of hospitals reporting to the NSSP rose across the study period. The study did not adjust for this by restricting to a common set of hospitals, as the authors suggest future research should do, or by standardizing for or at least reporting for readers the number of hospitals participating each month. This makes it difficult to interpret trends in absolute visit counts, particularly for outcomes such as deliberate self-harm, for which absolute counts remained below prepandemic levels for most or all of the study period (despite the number of hospitals reporting to NSSP rising).
Second, we see no clear way to draw inferences between the type of rate the study reports--for example, ED visits for self-harm per 100 000 total ED visits--and the rates of most immediate public health interest, that is, ED visits and events associated with self-harm per unit population. In particular, while it might have been plausible before March 2020 to assume that general patterns of ED care-seeking were mainly consistent over time, this has definitely not been the case during the COVID-19 pandemic--not for ED visits overall...
Language: en |