Predictive ability of a self-rated fall risk assessment tool in community-dwelling older women

Background: Falls are a substantial health problem among older adults. An accessible and reliable tool for assessing individual fall risk is needed. Aims: The predictive ability of a one-page self-rated fall risk assessment form (KaatumisSeula® [KS]) was evaluated among older women in its current form. Methods: A subsample (n = 384) of community-living older women (aged 72–84 years) participating in the Kuopio Fall Prevention Study (KFPS) completed the KS form. Participants’ falls were prospectively registered for 12 months with SMS messages. Their group status and form-based fall risk category were compared to the verified fall events during the KFPS intervention. Negative binomial regression and multinomial regression analyses were used. Physical performance measurements (single leg stance, leg extension strength and grip strength) were used as covariates. Results: During the follow-up, 43.8% of women fell at least once. Among the fallers, 76.8% had at least one self-determined injurious fall, and 26.2% had falls requiring medical attention. According to KS, 7.6% of the women had low fall risk, 75.0% moderate, 15.4% substantial, and only 2.1% high fall risk. Women in the “moderate fall risk” group had 1.47-fold (95% CI 0.74–2.91; nonsignificant), in “substantial fall risk” 4.00-fold (1.93–8.3; p < 0.001) and in “high fall risk” 3.00-fold (0.97–9.22; nonsignificant) higher risk of falls compared to the “low fall risk” group. Performance in physical tests did not account for future falls. Conclusions: The KS form proved to be a feasible tool for self-administered fall risk assessment with moderate predictive ability. Trial registration: ClinicalTrials.gov identifier: NCT02665169, date of first registration 27/01/2016.