Predictors of mortality among older patients in the medical wards of a tertiary hospital in Nigeria.

ABSTRACT

BACKGROUND: Older people face the biggest challenges in the overburdened healthcare services in Nigeria especially when hospitalized. There is no reliable data on the predictors of mortality in this population. AIMS: To determine the predictors of mortality among older patients on admission in the medical wards of University College Hospital, Ibadan. METHODS: Using a prospective cohort design, we investigated 450 patients (> 60 years) from the day of admission to death or discharge. Variables assessed included sociodemographic, family dynamics, lifestyle habits, healthcare utilization, quality of life, frailty, anxiety, depression, cognition, functional disability and anthropometric parameters. Kaplan-Meier method and Log-rank test were used to estimate and compare survival functions, respectively. Cox proportional hazard regression analysis was used to determine the predictors of mortality. RESULTS: The mean age of the subjects was 71.5 ± 8.0 years and 234 (52.0%) were females. Overall, there were 99 (22.0%) in-hospital deaths. The median survival time (MST) was 36.0 ± 3.0 days [females = 40.0 ± 3.5 days vs males = 31.0 ± 4.5 days (p < 0.001)]. There was a significant negative correlation between MST and age (r = - 0.931). Predictors of mortality on Cox's proportional hazard regression analyses were male sex HR = 2.03 (95% CI 1.27-3.24), severe frailty HR = 2.07 (1.02-4.20), cognitive impairment HR = 1.90 (1.14-3.17) and having ≥ 5 morbidities HR = 1.94 (1.14-3.30). CONCLUSION: There was a high mortality among older patients particularly the frail, male or those with multiple morbidities. Prompt and holistic management of morbidities and targeted interventions for cognitive impairment and frailty are needed to improve survival during hospitalization.