Infection prevention and control adherence among SARS-CoV-2 infected healthcare workers in Addis Ababa, Ethiopia: lessons for future outbreaks.

BACKGROUND: Health care worker (HCW) infections during outbreaks of high-consequence pathogens, from SARS-CoV-2 to Ethiopia's 2025-2026 Marburg outbreak, highlight persistent Infection Prevention and Control (IPC) gaps. We evaluated IPC adherence and its determinants among infected HCWs in Addis Ababa, Ethiopia, to inform future outbreak preparedness. METHODS: This census-based cross-sectional study (March 2020-March 2021) evaluated 1152 SARS-CoV-2-infected HCWs with clinical exposures 14 days pre-diagnosis. Environments were classified via an adapted OSHA pyramid into: High-Pathogen Designated Areas isolation and treatment units, General High-Risk Areas [intensive care units, emergency departments, and operating theaters], Medium-Risk Clinical Areas [inpatient and outpatient departments], and Low-Risk Support Areas [administrative and non-clinical zones]. Adherence was defined as correct practice in > 50% of interactions. Bivariate chi-square tests and multivariable logistic regression identified independent determinants for donning/doffing (n = 742) and N95 seal-checks (n = 350), reporting adjusted odds ratios (aOR) with 95% confidence intervals at P < 0.05. RESULTS: Only 19.2% of infected HCWs worked in High-Pathogen Designated Areas. Mask (85.7-97.7%) and glove (68.2-98.0%) adherence was high, but fell below 50% for gowns, coveralls, head caps, and shoe covers. Exceptions included face shields/goggles (52.8%) and N95 use during aerosol-generating procedures (AGPs) (66.5%). N95 seal-check compliance was 34.9%; only 1.1% used fit-tested respirators. Hand hygiene rose from 49.3% pre-interaction to 82.5% post-fluid exposure. Chi-square analyses linked health care professionasl and High-Pathogen Designated Area status with significantly superior adherence. For donning/doffing, lower compliance was predicted by non-High-Pathogen Designated Area placement (General High-Risk Areas aOR = 0.12 (0.07-0.19); Medium-Risk Clinical Areas aOR = 0.08 (0.04-0.14); Low-Risk Support Areas [aOR = 0.07 (0.03-0.17)] and support staff [aOR = 0.35 (0.14-0.88)]; AGP involvement [aOR = 3.61 (2.33-5.58)] and IPC training [aOR = 1.59 (1.05-2.41)] were positive predictors. N95 seal-check odds were lower in General High-Risk Areas aOR = 0.53 (0.30-0.93); Medium-Risk Clinical Areas aOR = 0.33 (0.16-0.66); Low-Risk Support Areas [aOR = 0.25 (0.08-0.77)]. Positive predictors included communal living [aOR = 2.37 (1.29-4.35)], IPC training [aOR = 1.71 (1.03-2.83)], and AGP involvement [aOR = 1.71 (1.05-2.80)], while age > 30 [aOR = 0.57 (0.34-0.98)] and stress [aOR = 0.46 (0.28-0.77)] were negative determinants. CONCLUSIONS: This study underscores IPC gaps among HCWs especially in non-designated areas, where most infections occurred, highlighting the need for standardized IPC across all healthcare environments.