The Norwegian General Motor Function assessment as an outcome measure for a frail elderly population: A validity study

Abstract

Aim To establish the validity of the Norwegian General Motor Function (NGMF) assessment scale. Method To establish construct and criteria validity, Spearman's rank correlation coefficients were calculated for the NGMF, and age, sex, medical conditions, history of falls and to four functional tests. Content validity was evaluated by asking participating physiotherapists about the usefulness of the items in the scale. Absolute reliability was evaluated by establishing the standard error of measurement and the minimal detectable change at the 95% level of confidence for total scores of the NGMF subscales for dependence, pain and insecurity. Results Construct validity was established to medical status and medication with subscales dependence and insecurity but not to subscale pain. Criterion validity was established between the NGMF subscales dependence, pain and insecurity, and the Barthel Index, the Falls Efficacy Scale to subscales dependence and insecurity, but not with pain, and the Timed Up-and-Go test, to subscale insecurity. Neither the Chair Stand Test nor registered falls were significantly associated with any of the subscales of the NGMF. Content validity of the NGMF was perceived relevant to work in a geriatric setting and as a communication tool for a multidisciplinary team. Minimal detectable change was calculated for dependence (2.76), pain (4.9) and insecurity (6.1), respectively. Conclusion The construct, criteria and content validity of the NGMF was established