Predictors of mortality in older patients admitted to a geriatric hospital

Geriatr Gerontol Int. 2019 Jan;19(1):70-75. doi: 10.1111/ggi.13573. Epub 2018 Nov 22.

Abstract

Aim: The identification of older patients at risk of mortality is important to provide properly tailored care and effectively manage healthcare resources. The present study aimed to identify predictors of all-cause mortality related to geriatric patients' clinical, functional and sociodemographic status at admission.

Methods: A retrospective study was carried out of patients admitted to a geriatric hospital from January to May 2013. A total of 208 patients were enrolled in the study. The outcome measure was 4-year mortality.

Results: The mortality rate was 26%. We found that age, red blood cells count and white blood cells count, as well as C-reactive protein level, albumin level and high-density lipoprotein cholesterol level significantly correlated with mortality. Furthermore, the presence of clinical symptoms, such as pressure ulcers and depressed level of consciousness, was predictive of poor outcome. Multidimensional aspects of aging that are assessed in the Comprehensive Geriatric Assessment - activities of daily living, instrumental activities of daily living, Barthel scale, Mini-Mental State Examination and The Clock Drawing Test - appeared to be strong predictors of 4-year mortality. The expression to estimate the probability of mortality based on the examined variables correctly classified nearly 85% of the analyzed cases.

Conclusions: Early detection of high-risk patients is of particular significance to reach a better survival rate among older adults. Clinicians should put more stress on the comprehensive surveillance of geriatric patients, rather than focusing solely on the treatment of chronic diseases. Geriatr Gerontol Int 2019; 19: 70-75.

Keywords: geriatric medicine, mortality; predictors; preventive medicine.

MeSH terms

  • Aged
  • Aged, 80 and over
  • Female
  • Geriatric Assessment
  • Health Services for the Aged*
  • Health Status
  • Hospitalization*
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Retrospective Studies
  • Risk Factors
  • Survival Rate*