[An examination of two reporting methods of falls among the elderly living in the community].

This study examined the concordance of two reporting methods for falls (3 month and 12 month recall) experienced by elderly living in a rural community. Four interview surveys on the occurrence of falls over a 3 month period were conducted from 1992 to 1993 every three months. In the final interview, subjects were also asked about the occurrence of falls during the past 12 months. A total of 799 subjects (270 men, 529 women) aged 65 and over, living in a village of Niigata Prefecture, responded to all four surveys. The incidence of falls in one year was 19.0% by the 3 month recall, and 19.1% by the 12 month recall. Overall agreement of fall occurrence between the two methods was 98.9%, and a coefficient of kappa as a measure of concordance was 0.96. On the other hand, subjects who sustained injuries during falls reported in the 3 month reporting did not necessarily recall having a fall when interviewed at the end of the study. In any case, this study reveals that for surveying falls by the elderly living in the community a 1 year recall is a reliable method, because of the high agreement of fall occurrence when comparing the two methods.