Postdoctoral Fellow The University of Manitoba Winnipeg, Canada
Background: The COVID-19 pandemic and associated restrictions had a major impact on healthcare delivery, including medications use, which could have been exacerbated due to socioeconomic inequalities that affect vulnerable population groups.
Objectives: To investigate the impact of COVID-19 pandemic on medication use among populations of different socioeconomic status (SES) and area of residence.
Methods: Administrative health databases from Manitoba, Canada, were used to conduct a population-based cohort study before (third quarter 2016 to first quarter 2020) and during (second quarter 2020 to first quarter 2021) the pandemic. We examined the overall use of 7 common medication classes (respiratory, cardiovascular, anti-diabetic, psychotropic, opioid, corticosteroid, and NSAID medications). Individuals with at least one prescribed medication were considered exposed within a quarter, with a denominator of total population within that quarter. Autoregressive models were used to estimate the impact of restrictions on quarterly medication use and to quantify the change in level and slope (second quarter of 2020 as the interruption point). Stratified models by SES (higher vs. lower) and area of residence (urban vs. rural) were conducted.
Results: We examined 1,353,485 to 1,411,630 individuals during the study period. Over 58% were among the higher SES group and more than 60% lived in urban areas. The pandemic restrictions were associated with a 10.4% (p=0.0792) and 8.6% (p=0.0144) relative reduction in overall medication use among residents in urban and rural areas, respectively. We observed a 9.7% relative decrease in medications use in the higher SES group (p=0.0763) and a 9.93% relative decrease in the lower SES group (p=0.0203). During the pandemic period, non-significant rebound increases were observed in lower SES (β3=0.337, p=0.5224) and higher SES (β3=0.494, p=0.5198) groups. Among those living in rural areas, we found a non-significant rebound increase per quarter (β3=0.585, p=0.4885), and the same pattern was seen for individuals living in urban areas (β3=0.156, p=0.6969).
Conclusions: Temporary declines in medication use – at different magnitudes – were observed within different classes of SES and area of residence, which can be attributed to a possible differential disruption in routine healthcare. Our findings highlight a key aspect of the influence of the pandemic on healthcare access disparities. Further investigation is needed to assess the potential effect of changes in medications use on health outcomes, with a focus on vulnerable and marginalized groups.
