Are assessment measures for components of the International Classification of Functioning, Disability and Health reproducible for use on pre-frail and frail older adults? A systematic review

Objective: To analyze the internal consistency, reliability and measurement error of assessment measures for body structure and function (postural balance, muscle strength, cardiorespiratory function and physical fitness), activity (walking and mobility) and participation (quality of life and social reintegration) outcomes for use on pre-frail and frail older adults. Methods: Searches were performed in six databases (Medline, Embase, ScienceDirect, Web of Science, The Cochrane Library and Lilacs) and reproducibility studies were included. Cronbach's alpha and the classification proposed by Terwee et al. (2007) were used for the interpretation of internal consistency. The intraclass correlation coefficient (ICC) and the Munro classification were used for the determination of reliability. The standard error of measurement (SEM), minimal detectable change (MDC) as well as percentages of error (SEM%) and change (MDC%) were calculated. Methodological quality of the studies was appraised using Boxes 4, 6 and 7 of the COSMIN Risk of Bias checklist. Results: Six studies were included in the present systematic review, with a total of 835 (82.4 years; 582 women and 253 men) pre-frail and frail older adults analyzed. Seven measures were identified for the assessment of body structure and function, activity and participation outcomes (Hierarchical Assessment of Balance and Mobility, Self-Assessment of Physical Fitness, Sarcopenia Quality of Life, Reintegration to Normal Living Index, Two-Minute Walking test, Six-Minute Walking test and Ten-Meter Walking test). The following measurement properties were identified: internal consistency = 0.84 (mean Cronbach's alpha); reliability = 0.87 (mean ICC); SEM range = 0.06 to 10.10; MDC range = 0.13 to 28.10; SEM% range = 5.2 % to 9.5 %; and MDC% range = 10.7 % to 28.5 %. Conclusion: The present systematic review found adequate internal consistency, high reliability and an acceptable measurement error for assessment measures of body structure and function, activity and participation outcomes for use on pre-frail and frail older adults.