Body mass index, age and in-hospital mortality: The NutritionDay multinational survey

Background & aims: Several large and long-term prospective studies have assessed the association of body-mass index (BMI) next to age with the risk of death in the general population, but few have examined the association with in-hospital mortality. We investigated the association between BMI, age and in-hospital mortality. Methods: We used data collected during 9 consecutive one-day/year surveys (NutritionDay in hospital 2006-2014) conducted in non-critically ill adult patients from 2,183 hospitals across 51 nations from 4 continents. We examined the association of BMI and age with the risk of in-hospital (30-day) death using logistic regression analysis adjusted for multiple confounders. Results: Crude mortality rates were 3.6% (95%CI, 3.5-3.7) and 2.1% (95%CI, 2.0-2.3) in the overall cohort (N = 97,344) and in those assessed within 72 hours since admission (N = 32,363), respectively. BMI and age were independently associated with the risk of death (no interaction observed), which decreased with BMI and increased with age. In the overall cohort, compared to normal weight status (BMI 18.5-24.9 kg/m2), death odds ratios for underweight (BMI < 18.5), overweight (BMI 25.0-29.9) and obesity (BMI ≥30) were 1.35 (95%CI, 1.20-1.53), 0.87 (95%CI, 0.77-0.97) and 0.73 (95%CI, 0.62-0.86), respectively. In patients assessed within 72 hours since admission, the associations were comparable: for underweight, 1.48 (95%CI, 1.11-1.96); for overweight, 0.80 (95%CI, 0.65-0.97); for obesity, 0.75 (95%CI, 0.58-0.96). Conclusion: In adult hospitalized patients BMI and age are independent predictors of in-hospital mortality. Low body weight is confirmed being a risk factor for death as in the general population, while overweight and obesity appear protective conditions. In the hospital setting, the use of normal weight status as reference low-risk category could also be challenged.