Predictors of survival among hospitalized COVID-19 patients in the Gambia, 2020–2022

In The Gambia, in-hospital mortality among laboratory-confirmed COVID-19 patients rose markedly, from 28.2% in 2020 to 48.4% in 2022. This mortality metric reflects proportion of deaths among admitted patients, distinct from the population-level case-fatality rate. Evidence on predictors of survival among hospitalised patients remains limited. We aimed to estimate in-hospital survival probability and identify independent predictors of time to death among COVID-19 patients admitted in The Gambia from March 2020 to December 2022. We conducted a retrospective cohort study of 705 laboratory-confirmed COVID-19 patients admitted to Edward Francis Small Teaching Hospital and the Medical Research Council Clinic. The primary outcome was time (days) to in-hospital death, with medical discharge as a competing event and admission date as time origin. Fine-Gray competing-risks regression estimated sub-distribution hazard ratios (SHRs) for mortality (95% CI; p < 0.05). Sensitivity analyses assessed the impact of excluding patients discharged against medical advice. Overall, in-hospital mortality was 33.3% (235/705), with an incidence rate of 40.1 deaths per 1,000 person-days. Most deaths occurred within the first 21 days. Advanced age was the strongest predictor (adjusted SHR: 3.54 for 50–59 years; 6.96 for ≥ 70 years). Independent comorbidities included chronic liver disease (SHR 3.20), cardiac failure (SHR 2.15), chronic kidney disease (SHR 1.91), and diabetes mellitus (SHR 1.78). Hypertension lost significance upon multivariable adjustment (SHR 0.92). COVID-19 vaccination was associated with a lower risk of mortality, although the adjusted association did not reach conventional statistical significance (SHR 0.58, 95% CI 0.32–1.06, p = 0.079), possibly due to limited power given only 11 deaths among vaccinated patients. Sensitivity analyses supported this directional consistency. In-hospital COVID-19 mortality in The Gambia was substantial, with advanced age and specific comorbidities independently predicting increased risk. The observed age-comorbidity interaction warrants further investigation for future risk-stratification tools. Although our vaccination estimate did not reach significance, its consistent protective direction combined with robust global evidence, supports continued vaccination scale-up in The Gambia and sub-Saharan Africa, where coverage in this cohort remained very low (8.2%). Prospective studies with standardised clinical data are needed to refine mortality prediction in this setting.