Malnutrition at Admission Predicts In-Hospital Falls in Hospitalized Older Adults

Nutrients. 2020 Feb 20;12(2):541. doi: 10.3390/nu12020541.

Abstract

Malnutrition leads to poor prognoses, including a predisposition to falls. Few studies have investigated the relationship between malnutrition and falls during hospitalization. This study aimed to determine malnutrition's association with falls during hospitalization. A retrospective observational study was conducted. Patients aged ≥65 years that were admitted to and discharged from a university hospital between April 2018 and March 2019 were examined. Patients with independent basic activities of daily living were included. Diagnosis of malnutrition was based on the European Society for Clinical Nutrition and Metabolism (ESPEN) criteria at admission. Disease information such as the Charlson Comorbidity Index (CCI) and reasons for hospitalization were reviewed. Kaplan-Meier curve and multivariate Cox regression analyses were performed. Data from 6081 patients (mean age: 74.4 ± 6.1 years; males: 58.1%) were analyzed. The mean CCI was 2.3 ± 2.8 points. Malnutrition was detected in 668 (11.0%) and falls occurred in 55 (0.9%) patients. Malnourished patients experienced a higher fall rate than those without malnutrition (2.4% vs. 0.7%, log-rank test p < 0.001). In multivariate analysis, malnutrition had the highest hazard ratio for falls among covariates (hazard ratio 2.78, 95% confidence interval 1.51-5.00, p = 0.001). In conclusion, malnutrition at the time of admission to hospital predicts in-hospital falls.

Keywords: fall; hospitalization; older adult; undernutrition.

Publication types

  • Observational Study

MeSH terms

  • Accidental Falls / statistics & numerical data*
  • Aged
  • Aged, 80 and over
  • Comorbidity
  • Female
  • Hospitals, University
  • Humans
  • Inpatients / statistics & numerical data*
  • Kaplan-Meier Estimate
  • Male
  • Malnutrition / diagnosis
  • Malnutrition / epidemiology*
  • Nutrition Assessment
  • Nutritional Status
  • Patient Admission / statistics & numerical data*
  • Proportional Hazards Models
  • Retrospective Studies