Admission nutrition risk as an independent predictor of 28-day mortality following hip fracture: A retrospective cohort study

Introduction

Hip fracture carries substantial early mortality risk in older adults, with reported 30-day mortality of 5–10%. Malnutrition is common in this population and biologically plausible as a contributor to adverse outcomes, yet whether it independently predicts short-term mortality — beyond age, sex, anaesthetic risk, and cognitive status — remains uncertain. This study aimed to determine whether admission nutrition risk, assessed using the Malnutrition Universal Screening Tool (MUST), independently predicts 28-day mortality following hip fracture.

Methods

Retrospective cohort study of 709 consecutive patients admitted with hip fracture. Nutrition risk was categorised as normal or any risk (at-risk/malnourished combined) using MUST. Eighteen patients with an ungradable ASA grade were excluded, leaving 691 in the final analysis. Multivariable logistic regression estimated the association between nutrition risk and 28-day all-cause mortality, adjusting for age, sex, ASA grade, and Abbreviated Mental Test Score (AMTS). Secondary analyses examined post-admission factors (delirium severity, mobilisation status) univariately.

Results

Of 691 patients, 128 (18.5%) had any nutrition risk on admission. 28-day mortality was 14.8% in the nutrition-risk group versus 5.2% in those with normal nutrition status (p<0.001). Nutrition risk was independently associated with mortality after full adjustment (adjusted OR 3.24, 95% CI 1.65–6.35, p=0.0006), with minimal change from the unadjusted estimate (OR 3.21, 95% CI 1.74–5.93), indicating the association is not explained by these baseline differences. Male sex (adjusted OR 3.49, 95% CI 1.80–6.77, p=0.0002) and lower AMTS (adjusted OR 0.90 per point, 95% CI 0.83–0.97, p=0.010) were also independently predictive. In sensitivity analysis retaining three MUST categories, at-risk status remained strongly associated with mortality (adjusted OR 3.87, 95% CI 1.87–7.99, p<0.001). Delirium severity (mean 4AT 4.62 in non-survivors vs 2.07 in survivors, p<0.001) and failure to mobilise postoperatively (17.1% vs 4.9% mortality, p<0.001) were strongly associated with mortality in exploratory analyses.

Conclusion

Admission nutrition risk, assessed by MUST, is independently associated with a more than threefold increase in 28-day mortality following hip fracture, largely unexplained by age, sex, anaesthetic risk, or cognitive status. Routine MUST screening at admission may add meaningful value to early risk stratification in this population and warrants consideration within standard hip fracture pathways.