130. Effects of COVID-19 on a Complex Behavioral Intervention to Improve the Diagnosis and Management of UTI in Nursing Homes

Abstract

BackgroundHalf of all urinary tract infections (UTI) are probably unnecessary. We conducted a cluster-randomized trial in which a toolkit to enhance the diagnosis and treatment of UTIs was introduced in study NHs via usual implementation versus an enhanced implementation approach based on external facilitation and peer comparison reporting.MethodsThirty Wisconsin NHs were randomized to each treatment arm in a 1.5:1 ratio. NHs used an online portal to report urine culture and antibiotic treatment data over a 6-month pre-intervention period (Jan-June 2019), a pre-COVID 8-month post intervention period (July 2019-Feb 2020) and an 8-month post-COVID intervention period (Mar-Oct 2020). Study outcomes included urine culture (UC), antibiotic start (AS), and antibiotic days of therapy (DOT) rates per 1,000 resident days. A generalized estimating equation model for panel data was used to assess differences in study outcomes between treatment arms before and after onset of the COVID-19 pandemic. STATA 16.1 was used for all analyses.ResultsA total of 802 UCs (457 pre-COVID, 345 post-COVID), 724 AS (401 pre-COVID, 323 post-COVID), and 6,454 DOT (3553 pre-COVID and 2901 post-COVID) were reported over the 16-month intervention period. No significant differences in the study outcomes were observed during the pre-COVID intervention period, however, UC rates in NHs assigned to the usual care arm of the study increased while those in the enhanced arm declined following onset of COVID-19 (Figure 1). AS and DOT rates followed a similar pattern although the differences between the study arms were not statistically significant.Figure 1. Post Implementation Periods ConclusionOur findings suggest that NHs assigned to usual implementation regressed in their diagnosis and treatment of UTIs during the COVID-19 pandemic while those receiving external facilitation and peer comparison reports were more resilient to the effects of COVID-19.Disclosures All Authors: No reported disclosures