|
Abstract
|
Sprague and colleagues (Sprague et al., 2022) investigated the temporal link between stimulus checks and opioid deaths in Ohio, US. As forecasted by public health experts, the authors found that opioid-related deaths have risen since the onset of COVID-19 in the US (Ochalek, Cumpston, Wills, Gal, & Moeller, 2020; Slavova, Rock, Bush, Quesinberry, & Walsh, 2020).
We have three areas of concern in Sprague et al. (2022) and provide potential remedies. The first area is the overreliance on causal language and interpretations that do not align with the study design for causal inference. While the authors state that their findings do not provide a direct causal link between economic incentive payments and opioid overdose deaths, they use language suggesting otherwise. For example, the last sentence in the Introduction says the authors set out to, "investigate the timing of the COVID-19 economic impact payments and the increases in opioid overdose deaths that have been reported during the COVID-19 pandemic (Vieson, Yeh, Lan, & Sprague, 2021)." This is misleading because the methodology from their time series analyses focuses on the temporal relationships in a single time series (opioid overdose deaths in Ohio) and no data from the COVID-19 pandemic (e.g., case rates, hospitalizations) or economic impact payments (e.g., number of eligible individuals, timing of payment receipt) are presented in the analysis. Another example is the use of the Granger causality test. Here the authors acknowledge that the test does not indicate causality, but that is not quite accurate...
Language: en |