605 - Child Physical Abuse Following COVID-19 Relief Payments

Friday, April 28, 2023

5:15 PM – 7:15 PM ET

Poster Number: 605
Publication Number: 605.104

Yilu Dong, Medical College of Wisconsin, Wauwatosa, WI, United States; Linda Moua, Medical College of Wisconsin, Milwaukee, WI, United States; Jenna Le, Medical College of Wisconsin, Wauwatosa, WI, United States; Wendi-Jo Wendt, Medical College of Wisconsin, Mequon, WI, United States; Danny G. Thomas, Medical College of Wisconsin, MIlwaukee, WI, United States

Presenting Author(s)

Yilu Dong, PhD

Assistant Professor
Medical College of Wisconsin
Wauwatosa, Wisconsin, United States

Background: Child physical abuse (CPA) has been linked to economic hardship. During COVID-19, the US issued three rounds of Economic Impact Payments (EIPs): 1) $1,200 in April 2020, together with an unemployment insurance (UI) expansion of $600/week through July 2020, 2) $600 in December 2020, and 3) $1,400 in March 2021. It is unknown whether these payments were associated with lower CPA rates.

Objective: To determine if COVID-19 relief payments were associated with lower CPA rates among the low-income families.

Design/Methods: This study used multi-group interrupted time series design to identify the association between each EIP payment and CPA incidence. All possible CPA encounters at 42 hospitals nationwide from Jan. 2019 to Dec. 2021 were extracted from the pediatric health information system (PHIS) and categorized into low-, mid-, and high-income groups by patient primary payers. The outcome was total number of CPA encounters among each group, per hospital/month. As identification of the associations required that the slope of the CPA rate (i.e., trend, not the constant gaps) of the high-income (control) group would mimic that of the low-income (treatment) group in absence of the interventions, only hospitals for which the parallel CPA trends held pre-COVID were used for the study (p > 0.1). Prais-Winsten regression was used to account for autocorrelation in errors after detecting an AR (1) process.

Results: 32 hospitals were eligible for the analysis. Durbin-Watson statistic of the transformed model = 2.09. Prior to COVID-19, the aggregated low- and high-income group shared similar CPA trends (p = 0.485) (Figure 1). Upon release of EIP 1, there was an immediate average reduction of 15.7 CPA encounters per hospital in the low-income group compared to its high-income counterpart (p < 0.001), followed by a significant average increase in slope of 4.9 CPA encounters per hospital/month (p < 0.001). In December 2020 (EIP 2), there was another relative decrease of 16.9 CPA encounters per hospital (p = 0.001), with an insignificant slope change (p = 0.401). In March 2021 (EIP 3), an average reduction of 12.7 CPA encounters per hospital was estimated, despite being insignificant at conventional levels (p = 0.14). Afterwards, the low-income group experienced a significant downward CPA trend with an average reduction of 9.8 encounters per hospital/month compared to the high-income (p = 0.004).

Conclusion(s):

COVID-19 relief payments were associated with lower CPA incidence immediately among the low-income families, but the influences were not sustained.