Prognostic factors and prediction model for 1-year mortality after proximal humeral fracture

Bastiaan Van Grootven1, Sigrid Janssens2, Laurence De Keyser2

  • 1Department of Public Health and Primary Care, Academic Centre for Nursing and Midwifery, KU Leuven, Leuven, Belgium.

Insights

Six pre-fracture patient characteristics can predict 1-year mortality after proximal humeral fracture (PHF). This clinical model aids in making informed treatment decisions for PHF patients.

Area of Science:

  • Gerontology
  • Orthopedic Surgery
  • Public Health

Background:

  • Proximal humeral fractures (PHF) are common in older adults and associated with increased mortality risk.
  • Identifying factors predicting post-fracture mortality is crucial for patient management.

Purpose of the Study:

  • To investigate the utility of patient characteristics in predicting 1-year mortality following proximal humeral fracture (PHF).
  • To develop and validate a clinical prediction model for post-PHF mortality.

Main Methods:

  • Retrospective study of 261 patients aged 65+ treated for PHF.
  • Collected baseline data including demographics, residency, and comorbidities.
  • Developed a prediction model using LASSO regression and validated it with split-sample and bootstrapping methods.

Main Results:

  • 10.3% of participants died within 1 year.
  • Key predictors for survival included pre-fracture ambulation, living at home, younger age, higher BMI, female gender, and fewer comorbidities.
  • A 6-predictor model (age, gender, comorbidity score, BMI, cognitive impairment, nursing home residency) showed good discrimination (0.891 training, 0.878 validation).

Conclusions:

  • A clinical prediction model combining 6 pre-fracture characteristics effectively predicts 1-year mortality after PHF.
  • These findings can inform treatment decisions for patients with proximal humeral fractures.

Related Concept Videos