Predictors for cause-specific and timing of deaths in patients with COVID-19: a cohort study in Taiwan.

ABSTRACT

BACKGROUND: This cohort study determines the predictors for cause-specific and timing of deaths in patients with COVID-19 in Taiwan. METHODS: Patients with laboratory-confirmed COVID-19 admitted to Taipei City Hospital from January 1 to July 31, 2022, were recruited in this cohort. All patients were followed up until death, discharge from the hospital, or August 31, 2022. Early deaths within the first 2 weeks were recorded, and the cause of death was confirmed by the death certificate database of Taiwan. Predictors of cause-specific and timing of deaths of patients with COVID-19 were determined using multinomial Cox proportional hazards regression analysis. RESULTS: Of the 195 (8.0%) patients who died during hospitalization, 147 (84.0%) had COVID-19-specific deaths. Moreover, 54.9% of the deceased patients had early death. After controlling for other covariates, patients aged ≥ 65 years had a higher risk of COVID-19-specific, non-COVID-19-specific, early, and late deaths [adjusted hazards ratio (AHR) 3.85, 6.45, 3.33, and 6.57; 95% confidence interval (CI) 1.91-7.78, 1.17-35.68, 1.51-7.36, and 2.18-19.76, respectively]. Fully vaccinated patients had a lower risk of COVID-19-specific (AHR 0.68; 95% CI 0.47-0.98) and early deaths (AHR 0.54; 95% CI 0.35-0.84), whereas comorbid patients with chronic obstructive pulmonary disease had a higher risk of non-COVID-19-specific deaths (AHR 5.43; 95% CI 1.73-17.03). CONCLUSIONS: This study suggests that prioritizing COVID-19 vaccination and carefully monitoring comorbid patients during hospitalization can reduce the risk of COVID-19-specific and early deaths and non-COVID-19-specific mortalities, respectively.