Pranata · Diabetes & metabolism 2021 · Systematic review and dose-response meta-analysis · n=12 studies (34,390 participants)

Body mass index and outcome in patients with COVID-19: A dose-response meta-analysis.

Cited 165 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic review and meta-analysis of observational studies

PubMed 32738402 · doi:10.1016/j.diabet.2020.07.005 · record verified 2026-08-29

What was done

A systematic review and dose-response meta-analysis was conducted across PubMed, Europe PMC, ProQuest, and Cochrane Central Database to assess the relationship between body mass index (BMI) and COVID-19 outcomes. Primary outcome was a composite poor outcome (mortality and disease severity); secondary outcomes were mortality and severity individually. Analyses examined categorical obesity, highest versus reference BMI categories, and continuous dose-response increments per 5 kg/m² BMI increase.

What was found

Across 12 studies (34,390 patients), obesity was significantly associated with composite poor outcome (OR 1.73, 95% CI [1.40, 2.14], P<0.001; I² = 55.6%), mortality (OR 1.55, 95% CI [1.16, 2.06], P=0.003; I² = 74.4%), and severity (OR 1.90, 95% CI [1.45, 2.48], P<0.001; I² = 5.2%). Comparing the highest BMI category to reference BMI showed higher odds for composite poor outcome (aOR 3.02, 95% CI [1.82, 5.00], P<0.001; I² = 59.8%), mortality (aOR 2.85, 95% CI [1.17, 6.92], P=0.002; I² = 79.7%), and severity (aOR 3.08, 95% CI [1.78, 5.33], P<0.001; I² = 11.7%). Dose-response modeling showed each 5 kg/m² increase in BMI conferred an aOR of 1.052 (95% CI [1.028, 1.077], P<0.001) for composite poor outcome, with significant non-linearity (P < 0.001) and a steeper risk curve at higher BMI ranges.

Why it matters

This meta-analysis demonstrates that COVID-19 risk is not merely a binary threshold effect of clinical obesity, but increases continuously and non-linearly with higher BMI.

Limits

All included studies were observational, which precludes causal inference. Substantial statistical heterogeneity was present for mortality outcomes (I² = 74.4% to 79.7%). The abstract does not detail specific adjustment covariates across studies, definitions of reference BMI categories, or variations in regional demographics and treatment standards.

Cited by