Search

Objectives: To assess outcomes of patients admitted to hospital with COVID-19 and to determine the predictors of mortality., Setting: This study was conducted in six facilities, which included both government and privately run secondary and tertiary level facilities in the central and coastal regions of Kenya., Participants: We enrolled 787 reverse transcriptase-PCR-confirmed SARS-CoV2-infected persons. Patients whose records could not be accessed were excluded., Primary and Secondary Outcome Measures: The primary outcome was COVID-19-related death. We used Cox proportional hazards regressions to determine factors related to in-hospital mortality., Results: Data from patients with 787 COVID-19 were available. The median age was 43 years (IQR 30-53), with 505 (64%) being men. At admission, 455 (58%) were symptomatic with an additional 63 (9%) developing clinical symptoms during hospitalisation. The most common symptoms were cough (337, 43%), loss of taste or smell (279, 35%) and fever (126, 16%). Comorbidities were reported in 340 (43%), with cardiovascular disease, diabetes and HIV documented in 130 (17%), 116 (15%), 53 (7%), respectively. 90 (11%) were admitted to the Intensive Care Unit (ICU) for a mean of 11 days, 52 (7%) were ventilated with a mean of 10 days, 107 (14%) died. The risk of death increased with age (HR 1.57 (95% CI 1.13 to 2.19)) for persons >60 years compared with those <60 years old; having comorbidities (HR 2.34 (1.68 to 3.25)) and among men (HR 1.76 (1.27 to 2.44)) compared with women. Elevated white cell count and aspartate aminotransferase were associated with higher risk of death., Conclusions: The risk of death from COVID-19 is high among older patients, those with comorbidities and among men. Clinical parameters including patient clinical signs, haematology and liver function tests were associated with risk of death and may guide stratification of high-risk patients., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)