Mortality over time among COVID-19 patients hospitalized during the first surge of the pandemic: A large cohort study.

BACKGROUND: Capacity strain negatively impacts patient outcome, and the effects of patient surge are a continuous threat during the COVID-19 pandemic. Evaluating changes in mortality over time enables evidence-based resource planning, thus improving patient outcome. Our aim was to describe baseline risk factors associated with mortality among COVID-19 hospitalized patients and to compare mortality rates over time. METHODS: We conducted a retrospective cohort study in the largest referral hospital for COVID-19 patients in Sao Paulo, Brazil. We investigated risk factors associated with mortality during hospitalization. Independent variables included age group, sex, the Charlson Comorbidity Index, admission period according to the stage of the first wave of the epidemic (early, peak, and late), and intubation. RESULTS: We included 2949 consecutive COVID-19 patients. 1895 of them were admitted to the ICU, and 1473 required mechanical ventilation. Median length of stay in the ICU was 10 (IQR 5-17) days. Overall mortality rate was 35%, and the adjusted odds ratios for mortality increased with age, male sex, higher Charlson Comorbidity index, need for mechanical ventilation, and being admitted to the hospital during the wave peak of the epidemic. Being admitted to the hospital during the wave peak was associated with a 33% higher risk of mortality. CONCLUSIONS: In-hospital mortality was independently affected by the epidemic period. The recognition of modifiable operational variables associated with patient outcome highlights the importance of a preparedness plan and institutional protocols that include evidence-based practices and allocation of resources.

Adverse events and death in stroke patients admitted to the emergency department of a tertiary university hospital.

OBJECTIVES: To identify the occurrence of adverse events in stroke patients presenting to the emergency department of a tertiary university facility, and to disclose the categories of adverse events associated with death. METHODS: This matched case-control study enrolled 468 patients admitted with stroke to the emergency department from March 1996 to September 1999. The cases comprised 234 consecutive deaths and the controls 234 discharged patients, matched for primary diagnosis and admission period. Adverse events, detected by chart review, were classified according to the degree of severity, immediate causes, and professional category. The association with death was analysed by conditional logistic regression. RESULTS: Adverse events totaled 1218 and occurred in 295 patients: 932 events (76.5%) in 170 cases and 286 (23.5%) in 125 controls. Major adverse events equaled 54.1% of all events (659 episodes): 538 events in 143 cases and 121 in 65 controls. Diagnostic or therapeutic procedures and nursing activities accounted for 55.2% of events. Nursing (38.4%) and medical (31%) adverse events represented the most common related professional categories. A significant association with death was found for major adverse events, medical adverse events, and nosocomial infections, with adjusted odds ratio estimates of 3.74 [95% confidence interval (CI) 1.64-8.54], 3.71 (95% CI 1.61-8.53), and 3.22 (95% CI 1.21-8.59), respectively. CONCLUSION: Adverse events, mostly severe, predominated among deceased patients, resulting mainly from diagnostic or therapeutic procedures and nursing activities. In spite of limitations concerning the observational retrospective nature of this study, we found that severe adverse events, medical adverse events, and nosocomial infections were significantly associated with death in stroke patients.