HMB-Containing Oral Nutritional Supplementation and Mortality After Hip Fracture in Malnourished Older Adults: A Formulation-Specific Subanalysis of a Prospective Cohort

Background: Oral nutritional supplementation (ONS) is commonly prescribed in malnourished older adults after hip fracture, but formulations are heterogeneous, and their comparative association with mortality remains unclear. We aimed to evaluate whether HMB-containing ONS was associated with lower mortality than non-HMB ONS and to explore whether supplement formulation combined with treatment persistence was associated with differential mortality patterns. Methods: This was a formulation-specific subanalysis of a previously described prospective cohort of older adults with hip fracture and malnutrition or significant nutritional risk. Only patients with known ONS formulation were included (n = 107): 59 received HMB-containing ONS, and 48 received non-HMB ONS, including standard, diabetes-specific, and renal-oriented formulations. Mortality at 3, 6, and 12 months was analyzed using crude comparisons and multivariable logistic regression adjusted for sex, age, and Charlson comorbidity index. A 6-month adjusted Cox model was used as the main time-to-event analysis. Exploratory analyses assessed mortality according to supplement formulation and treatment persistence. Results: Overall mortality was 14.0% at 3 months, 23.4% at 6 months, and 29.9% at 12 months. At 6 months, mortality was lower among patients receiving HMB-containing ONS than among those receiving non-HMB ONS (13.6% vs. 35.4%; p = 0.011), and the association remained significant after adjustment (OR 0.267; 95% CI 0.091–0.784; p = 0.016). Associations at 3 and 12 months were directionally consistent but not statistically significant. In the adjusted Cox model, prescription of HMB-containing ONS was associated with a lower hazard of death within 6 months (HR 0.358; 95% CI 0.145–0.885; p = 0.026). Exploratory analyses showed a 6-month mortality gradient according to formulation and persistence, ranging from 0.0% in patients receiving HMB-ONS for ≥3 months to 41.2% in those receiving non-HMB ONS for <3 months. Conclusions: In this formulation-specific subanalysis of malnourished older adults with hip fracture, an association between HMB-containing ONS and lower 6-month mortality was observed compared with non-HMB ONS. Exploratory findings suggested a clinically relevant mortality gradient according to both supplement formulation and treatment persistence, although these results should be interpreted cautiously. Larger prospective studies are warranted to confirm these findings.