Nutritional risk and survival in elderly veterans: a five-year follow-up

J Community Health. 1993 Dec;18(6):327-34. doi: 10.1007/BF01323964.

Abstract

The purpose of this study was to identify predictors of survival over five years in a sample of 377 elderly (age 55+) chronically ill men. Subjects were selected at baseline from consecutive appointments at a geriatric clinic and given extensive medical and psychosocial assessment. Five years later, subjects who could be located were interviewed by telephone. Interviews were completed with 194 (51.4%) subjects, 90 others were confirmed as deceased, 29 subjects could not be interviewed, and 64 more were not located although VA records did not show that any were deceased. The predictor variable of interest was the Nutritional Risk Index (NRI), a 16 item index which measures nutritional dimensions of health status. Other variables included functional health status [Activities of Daily Living (ADL), Instrumental Activities of Daily Living (IADL)], mental health status [Mini-Mental Status Exam (MMSE), morale], health habits (smoking, alcohol use and exercise), use of health services (physician visits, emergency room (ER) visits, hospital stays), and demographic factors (age, income, marital status and living arrangements). Proportional hazard models showed that the variables with statistically significant direct effects on survival time were younger age, higher functional health status on IADL, non-smoking, moderate alcohol use, and perception of adequate income. Nutritional status was indirectly associated with survival.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Activities of Daily Living
  • Aged
  • Chronic Disease
  • Critical Illness / mortality*
  • Follow-Up Studies
  • Health Status
  • Humans
  • Likelihood Functions
  • Logistic Models
  • Male
  • Middle Aged
  • Nutritional Status*
  • Prevalence
  • Proportional Hazards Models
  • Regression Analysis
  • Smoking / epidemiology
  • Survival Analysis
  • United States / epidemiology
  • Veterans / statistics & numerical data*