Objectives: Previous literature has suggested that obesity impacts mortality risk differently in bacterial versus viral infections. This study sought to further elucidate this association in pneumonia versus influenza.

Design: Retrospective cohort study.

Setting and participants: Data were collected from the US Nationwide Readmission Database from 2013 to 2014.

Methods: Patients were categorized into three weight groups: normal weight (BMI 18.5-25.0 kg/m2), obese (BMI 30-40.0 kg/m2), and morbidly obese (BMI ≥ 40 kg/m2). To minimize confounding, we excluded patients with a history of smoking, alcoholism, or chronic wasting conditions, as suggested by the Global BMI Mortality Collaboration. To further isolate obesity from baseline differences across cohorts, we performed a three-way propensity matching analysis. The association between body weight and in-hospital all-cause 30-day mortality was assessed using Cox proportional hazard regression analysis.

Results: 132,965 influenza and 34,177 pneumonia hospitalizations were identified. For patients with influenza, obesity (hazard ratio [HR]: 1.51; 95% CI: 1.01-2.26) and morbid obesity (HR: 1.64; 95% CI: 1.10-2.44) were associated with higher in-hospital 30-day mortality compared to normal weight. For pneumonia, obesity (HR, 0.41; 95% CI, 0.20-0.84) and morbid obesity (HR, 0.49; 95% CI, 0.25-0.96) were associated with reduced 30-day mortality compared to normal weight.

Conclusions and implications: Obesity may increase 30-day mortality risk during influenza hospitalization but provide mortality benefit in pneumonia, a divergent effect not adequately explained by lower admission threshold.