Feasibility and Accuracy of Different Methods for Collecting Data on Falls Among Older People With Dementia.

Abstract

This study compared different methods for collecting data on falls among people with dementia to identify which is most feasible and accurate. Eighty-three dyads, comprised of a community-dwelling person with dementia and their informal carer, participated in the TAi ChI for people with demenTia (TACIT) trial. Falls were collected prospectively over 6 months using monthly calendars, weekly and monthly telephone interviews, and 3-monthly telephone interviews with the carer. Unique falls identified across the reporting methods were combined, and this total was compared against each reporting method in isolation and combinations. A higher frequency of falls indicated greater accuracy. Falls data collection was most feasible with weekly telephone interviews (84%), and most accurate with the combination of weekly telephone interviews with monthly calendars (96%). For the greatest completeness and accuracy of falls data with community-dwelling people with dementia, researchers should use both weekly telephone interviews and monthly calendars.