Risk factors for mortality in patients with Coronavirus disease 2019 (COVID-19) infection: a systematic review and meta-analysis of observational studies

Abstract

Purpose: Coronavirus disease 2019 (COVID-19) is an emerging disease that was first reported in Wuhan city, the capital of Hubei province in China, and has subsequently spread worldwide. Risk factors for mortality have not been well summarized. Current meta-analysis of retrospective cohort studies was done to summarize available findings on the association between age, gender, comorbidities and risk of death from COVID-19 infection.

Methods: Online databases including Web of Science, PubMed, Scopus, Cochrane Library and Google scholar were searched to detect relevant publications up to 1 May 2020, using relevant keywords. To pool data, random-effects model was used. Furthermore, sensitivity analysis and publication bias test were also done.

Results: In total, 14 studies with 29,909 COVID-19 infected patients and 1445 cases of death were included in the current meta-analysis. Significant associations were found between older age (≥65 vs <65 years old) (pooled ORs = 4.59, 95%CIs = 2.61-8.04, p < .001), gender (male vs female) (pooled ORs = 1.50, 95%CIs = 1.06-2.12, p = .021) and risk of death from COVID-19 infection. In addition, hypertension (pooled ORs = 2.70, 95%CIs = 1.40-5.24, p = .003), cardiovascular diseases (CVDs) (pooled ORs = 3.72, 95%CIs = 1.77-7.83, p = .001), diabetes (pooled ORs = 2.41, 95%CIs = 1.05-5.51, p = .037), chronic obstructive pulmonary disease (COPD) (pooled ORs = 3.53, 95%CIs = 1.79-6.96, p < .001) and cancer (pooled ORs = 3.04, 95%CIs = 1.80-5.14, p < .001), were associated with higher risk of mortality.

Conclusions: Older age (≥65 years old), male gender, hypertension, CVDs, diabetes, COPD and malignancies were associated with greater risk of death from COVID-19 infection. These findings could help clinicians to identify patients with poor prognosis at an early stage.

Keywords: COVID-19; SARS-CoV-2; meta-analysis; mortality; systematic review.