Determinants of treatment outcomes among hospitalised patients with skin and soft tissue infections: a prospective observational study.

Abstract

The study aimed to determine treatment outcome and factors affecting treatment outcomes among hospitalised patients with skin and soft tissue infections (SSTIs) at the University of Gondar Comprehensive Specialised Hospital (UOGCSH) in Ethiopia. An institution-based prospective observational study. UOGCSH, Northwest Ethiopia. 423 patients from all age groups with clinically diagnosed SSTIs from 25 June to 25 December 2023 at the UOGCSH were included. Primary treatment outcomes were early apparent clinical response within 48-72 hours and treatment failure after 72 hours of optimal antibiotic therapy. Secondary treatment outcomes included hospital length of stay (HLOS) and in-hospital mortality. Multiple linear regression assessed factors influencing the HLOS, and multivariable logistic regression identified predictors of treatment failure. The average HLOS was 13.46±3.01 days. Of the patients, 39.3% had an early clinical response within 48-72 hours, whereas 34.4% had treatment failure. At 0.7%, the in-hospital death rate was modest. Living in a rural area (adjusted OR (AOR) 5.54, 95% CI 2.67 to 11.37), having concurrent illnesses (AOR 2.11, 95% CI 1.10 to 4.07) and starting antibiotics later than 12 hours (AOR 0.08, 95% CI 0.04 to 0.17) were significantly associated with treatment failure. Concomitant disorders and complex comorbidities were also associated with longer HLOS, whereas higher socioeconomic level, oral step-down therapy, early antibiotic initiation and early clinical response were linked to better results and shorter hospital stays. Timely antibiotic initiation, efficient source control, patient comorbidities and socio-economic considerations affect the treatment course for SSTIs. Prolonged treatment and the frequent use of 'watch' and 'reserve' antibiotics underscore the need for improved antimicrobial stewardship. In this situation, optimising clinical results and minimising HLOS requires prompt clinical evaluation and customised antibiotic therapy. However, the single-centre design and potential residual confounding may introduce bias.