Predictors of institutionalization among home-dwelling older Finnish people: a 22-year follow-up study.

Background: Identification of predictive factors on institutionalization provides the basis for the development and application of preadmission assessment. There is a lack of evidence for predictors of institutionalization for older people. Aims: To examine the effect of predictive factors on institutionalization in home-dwelling 70-year-old people. Methods: The data were collected in 1991 by the clinical examinations, a postal questionnaire, and an interview from the residents of Turku, Finland, born in 1920 ( n = 1032). Institutionalization was defined as entry into a nursing home or sheltered housing at any time during a 22-year follow-up. Results: A rate of institutionalization was 22.0%. In multivariable Cox regression analysis, impaired cognitive function (MMSE 18-26) (hazard ratio 1.71, confidence interval 1.24-2.36) and low BMI (<25 kg/m) (compared to both BMI of 25-29.9 and that of ≥30, respectively, 1.88, 1.32-2.67, and 1.66, 1.05-2.60), having several falls during the previous year (2.50, 1.28-4.90). Conclusions: We conclude that impaired cognitive function, low BMI, and frequent falling predicted institutionalization during the 22-year follow-up. To reduce or postpone institutionalization, interventions should target risk factors, such as frailty, physical limitations, and falling. In addition, community-based services according to the needs and functional ability of the home-dwelling older people should be developed. [ABSTRACT FROM AUTHOR]