Roorda, Leo D., Roebroeck, Marij E., van Tilburg, Theo, Molenaar, Ivo W., Lankhorst, Gustaaf J., Bouter, Lex M., Boonstra, Anne M., de Laat, Fred A., Caron, Jacob J., Burger, Bart J., Heyligers, Ide C., Nollet, Frans, Stover-van Herk, Irene E., Perez, Roberto S., Meijer, Jan W., Rijken, P. Mieke, Amsterdam Movement Sciences, Rehabilitation medicine, Sociology [until 2010], Sociology and Social Gerontology, EMGO+ - Musculoskeletal Health, The Social Context of Aging (SoCA), Research Institute MOVE, and Rehabilitation Medicine
Roorda LD, Roebroeck ME, van Tilburg T, Molenaar IW, Lankhorst GJ, Bouter LM, and the Measuring Mobility Study Group. Measuring activity limitations in walking: development of a hierarchical scale for patients with lower-extremity disorders who live at home. Objective To develop a hierarchical scale that measures activity limitations in walking in patients with lower-extremity disorders who live at home. Design Cross-sectional study. Setting Orthopedic workshops and outpatient clinics of secondary and tertiary care centers. Participants Patients (N=981; mean age ± standard deviation, 58.6±15.4y; 46% men) living at home, with different lower-extremity disorders: stroke, poliomyelitis, osteoarthritis, amputation, complex regional pain syndrome type I, and diabetic and degenerative foot disorders. Interventions Not applicable. Main Outcome Measures (1) Fit of the monotone homogeneity model, indicating whether items can be used for measuring patients; (2) fit of the double monotonicity model, indicating invariant (hierarchical) item ordering; (3) intratest reliability, indicating repeatability of the sum score; (4) robustness, addressing the clinimetric properties within subgroups of patients; and (5) differential item functioning, addressing the validity of comparisons between subgroups of patients. Results Thirty-five of 41 dichotomous items had (1) good fit of the monotone homogeneity model (coefficient H =.50), (2) good fit of the double monotonicity model (coefficient H T =.33), (3) good intratest reliability (coefficient ρ=.95), (4) satisfactory robustness (within subgroups of patients defined by age, sex, and diagnosis), and (5) some differential item functioning (6 items in amputees compared with nonamputees). Conclusions A hierarchical scale, with excellent scaling characteristics, was developed to measure activity limitations in walking in patients with lower-extremity disorders who live at home. The measurements should be interpreted cautiously when making comparisons between amputees and nonamputees.