Abstract
BACKGROUND: In this longitudinal study, we aimed to determine factors influencing survival outcomes among patients with stroke at three tertiary hospitals over a 12-month period. The investigation sought to uncover influential determinants to enhance the precision of prognostic assessments and inform targeted interventions for individuals affected by stroke. METHODS: Employing a longitudinal study design, participants were observed for 12 months from baseline, censoring survivors at the endpoint. The dataset originated from a comprehensive study involving stroke patients treated at three referral hospitals in Zimbabwe: Parirenyatwa, Sally Mugabe, and Chitungwiza Central Hospitals. The primary outcome variable, the duration of survival until death, was measured in days from the initiation of stroke treatment. Gompertz parametric regression analysis was utilized for data modeling following AFT model diagnostics. RESULTS: In our study, 188 stroke patients were enrolled at baseline. However, 51 patients were excluded from the analysis due to either missing information or loss to follow-up. Among the remaining 137 patients who were tracked over a 12-month period, 42% were censored, and 58% were deceased. Individuals utilizing 'Free Service (older than 65/pensioners/retirees/social welfare)' hospital bill payment methods showed a decreased risk of death, (adjusted hazard ratio; aHR: 0.4, 95% CI: 0.20, 0.80), suggesting a protective effect compared to cash paying patients. Those who had attained a secondary school level education displayed a significantly lower risk of death (aHR: 0.4, 95% CI: 0.24, 0.79) compared to those with primary level education. Age was a significant risk factor, with individuals aged 45-65 and those over 65 years showing higher adjusted hazard ratios 3.4 (95% CI: 1.42, 8.36) and 3.7(95%CI:1.44, 9.36), respectively, relative to those below 45 years of age. Housing status revealed a protective effect for those residing with parents/relatives (aHR: 0.4, 95% CI: 0.20, 0.64). Total functional outcome demonstrated significantly lower hazards for individuals with mild or moderate (aHR: 0.2, 95% CI: 0.09, 0.40) and severe outcomes (aHR: 0.2, 95% CI: 0.10, 0.46) compared to those with very severe outcomes. CONCLUSION: The study findings demonstrate that hospital bill payment methods, housing status and staying with relatives, educational attainment, functional outcome, and age significantly affect survival outcomes among stroke patients. This highlights the need to consider socio-demographic and clinical variables in the development of prognostic assessments and targeted interventions for individuals recovering from stroke.