[Risk factors for falls among the elderly living in a Japanese rural community]

The purpose of the present study is to explore risk factors for falls among the elderly in a Japanese rural community. Subjects comprised 658 elderly persons aged 65 and over living in a rural village of Akita Prefecture (response rate; 96.1%). In 1988, the baseline medical health examinations and an interview survey were performed and repeated at the follow-up survey one year later. Results obtained were as follows; 1) The rate of falls at the follow-up survey was 17.3% in men, 15.6% in women. There was no difference in rate of falls between sexes. The rate of falls in the old-old did not significantly differ from that in the young-old in both sexes. 2) Chi square test or Cochran-Armitage test was performed by sex to examine the relationship between the factors investigated at the baseline survey and occurrence of falls at the follow-up survey. Significant factors in men were income, spouse, need for assistance in daily living, history of stroke, fall experience within one year preceding the baseline survey, ECG abnormalities, standing time with eyes open, grip strength, TMIG (Tokyo Metropolitan Institute of Gerontology) index of competence. In women, mobility, need for assistance in daily living, admission to a hospital within one year preceding the baseline survey, fall experience within one year preceding the baseline survey, standing time with eyes open, grip strength, TMIG index of competence were significant. 3) Using the significant variables in univariate analysis, multivariate logistic regression analysis controlling for age was performed by sex. Significant risk factors for falls in men were fall experience within one year preceding the baseline survey and ECG abnormalities, while for women fall experience within one year preceding the baseline survey was significant. The present longitudinal study revealed that falls experience within one year preceding the baseline survey was the highest risk factor for falls in the elderly residents in the community. This indicates that even a simple interview on fall experience can identify high risk elderly.