[Relationship between eating dysfunction and life span and mortality in institutionalized elderly people].

Aim: The aim of this study was to evaluate the relationship between eating disorders and anticipated life span stage and mortality in institutionalized elderly people.

Methods: Ninety-eight elderly residents (mean 86.3 +/- 5.9 years) at a nursing home were selected for investigation of the date and cause of death. Inclusion was limited to those who died within one week from admission to the institution/hospital. Data on 11 potential prognostic factors were analyzed: underlying disease, activity of daily living (ADL), disorder of anticipatory stage, swallowing function, need for feeding assistance, weight decrease rate in the part 6 months, body mass index (BMI), minimum nutritional assessment (MNA), occlusal status, age and sex. These 11 factors were analyzed by Kaplan-Meier survival curve, and either the log-rank or Wilcoxon test was used to analyze significant differences in survival days in subjects with and without the factors listed above. Moreover, the regression analysis was made using Cox proportional-hazards model, a nonlinear multivariate analysis, to extract factors contributing greatly to short survival days.

Results: The survival days showed significant differences in six factors; ADL, disorder of anticipatory stage, swallowing function, need for feeding assistance, BMI and MNA. Three factors were shown to be related to a significantly shorter mortality by the Cox proportional-hazards model: disorder of anticipatory stage (hazard rate = 2.85, 95% CI = 1.04-7.83), swallowing function (hazard rate = 2.90, 95% CI = 1.06-7.91) and BMI (hazard rate = 2.54, 95% CI = 1.00-6.44) were involved in.

Conclusion: This study speculated that an appropriate response with a focus on feeding training may possibly contribute to prolong the mortality of elderly people.