The implementation of infection prevention and control measures and health care utilisation in ACF-supported health facilities during the COVID-19 pandemic in Kinshasa, Democratic Republic of the Congo, 2020.

ABSTRACT

BACKGROUND: Infection prevention and control (IPC) was a central component of the Democratic Republic of the Congo's COVID-19 response in 2020, aiming to prevent infections and ensure safe health service provision. OBJECTIVES: We aimed to assess the evolution of IPC capacity in 65 health facilities supported by Action Contre la Faim in three health zones in Kinshasa (Binza Meteo (BM), Binza Ozone (BO), and Gombe), investigate how triage and alert validation were implemented, and estimate how health service utilisation changed in these facilities (April-December 2020). METHODS: We used three datasets IPC Scorecard data assessing health facilities' IPC capacity at baseline, monthly and weekly triage data, and monthly routine data on eight health services. We examined factors associated with triage and isolation capacity with a mixed-effects negative binomial model and estimated changes in health service utilisation with a mixed-model with random intercept and long-term trend for each health facility. We reported incidence rate ratios (IRRs) for level change when the pandemic began, for trend change, and for lockdown and post-lockdown periods (Gombe). We estimated cumulative and monthly percent differences with expected consultations. RESULTS: IPC capacity reached an average score of 90% by the end of the programme. A one-point increase in the IPC score was associated with +6% and +5% increases in triage capacity in BO and Gombe, respectively, and with +21% and +10% increases in isolation capacity in the same zones. When the pandemic began, decreases were seen in outpatient consultations (IRR 0.67, 95% confidence interval (CI) [0.48-0.95] BM&BO-combined; IRR 0.29, 95%CI [0.16-0.53] Gombe), consultations for respiratory tract infections (IRR 0.48, 95%CI [0.28-0.87] BM&BO-combined), malaria (IRR 0.60, 95%CI [0.43-0.84] BM&BO-combined, IRR 0.33, 95%CI [0.18-0.58] Gombe), and vaccinations (IRR 0.27, 95%CI [0.10-0.71] Gombe). Maternal health services decreased in Gombe (ANC1 IRR 0.42, 95%CI [0.21-0.85]). CONCLUSIONS: The effectiveness of the triage and alert validation process was affected by the complexity of implementing a broad clinical definition in limited-resource settings with a pre-pandemic epidemiological profile characterised by infectious diseases with symptoms like COVID-19. Readily available testing capacity remains key for future pandemic response to improve the disease understanding and maintain health services.