TY - JOUR
T1 - Validity and internal consistency of mobility scales for healthy older people in Germany
AU - Brühl, Albert
AU - Avlund, Kirsten
PY - 2012/3
Y1 - 2012/3
N2 - Background/Purpose: This study presents the results of structure validity tests of two measures of mobility for healthy older people. The aim of this article is to analyze whether the results of the validation study by Avlund et al (Scand J Soc Med 1993;21:233-46; Eur J Public Health 1996;6:35-41) can be reproduced in Germany and to explore whether fatigue as measured by one of the scales is predictive of the need for help and nursing costs at 3-year follow-up. Method: Data were collected between 2004 and 2007 from 1197 75+-year-olds in southern Germany. Mobility was measured by questions in relation to fatigue and need of help. Structural validity and internal consistency was tested using the Rasch model for item analysis. The fit of the Rasch model was compared with latent class analysis (LCA) in order to explore whether the used scales produce quantitative or qualitative data. The predictive validity of the scales was analyzed in relation to need of help in mobility and costs to the nursing and health insurance systems in Germany at 3-year follow-up. Results: Two scales on mobility, one in relation to need of help (the Mob-H scale) and one regarding fatigue (the Mob-T scale), were derived. The classes in the latent class solution differed in their sum scores, so that the use of the sum score produced valid differences although the qualitative latent classes showed a better model fit than did the quantitative Rasch model. Furthermore, fatigue as measured by the Mob-T scale turned out to be related to the development of dependency in mobility and to nursing care costs at 3-year follow-up. Further, the strongest predictor of nursing care costs was need of help, measured by the quantitative Mob-H scale. Conclusion: The Mob-T and Mob-H scales developed in Denmark were reproduced in a German population, using the Rasch model for item analysis. Further, fatigue in mobility activities (Mob-T) was predictive of need of help and nursing care costs at 3-year follow-up. These findings are substantial because they suggest that it is highly relevant to put attention to fatigue in mobility in health care programs with the aim of reducing health care and nursing costs.
AB - Background/Purpose: This study presents the results of structure validity tests of two measures of mobility for healthy older people. The aim of this article is to analyze whether the results of the validation study by Avlund et al (Scand J Soc Med 1993;21:233-46; Eur J Public Health 1996;6:35-41) can be reproduced in Germany and to explore whether fatigue as measured by one of the scales is predictive of the need for help and nursing costs at 3-year follow-up. Method: Data were collected between 2004 and 2007 from 1197 75+-year-olds in southern Germany. Mobility was measured by questions in relation to fatigue and need of help. Structural validity and internal consistency was tested using the Rasch model for item analysis. The fit of the Rasch model was compared with latent class analysis (LCA) in order to explore whether the used scales produce quantitative or qualitative data. The predictive validity of the scales was analyzed in relation to need of help in mobility and costs to the nursing and health insurance systems in Germany at 3-year follow-up. Results: Two scales on mobility, one in relation to need of help (the Mob-H scale) and one regarding fatigue (the Mob-T scale), were derived. The classes in the latent class solution differed in their sum scores, so that the use of the sum score produced valid differences although the qualitative latent classes showed a better model fit than did the quantitative Rasch model. Furthermore, fatigue as measured by the Mob-T scale turned out to be related to the development of dependency in mobility and to nursing care costs at 3-year follow-up. Further, the strongest predictor of nursing care costs was need of help, measured by the quantitative Mob-H scale. Conclusion: The Mob-T and Mob-H scales developed in Denmark were reproduced in a German population, using the Rasch model for item analysis. Further, fatigue in mobility activities (Mob-T) was predictive of need of help and nursing care costs at 3-year follow-up. These findings are substantial because they suggest that it is highly relevant to put attention to fatigue in mobility in health care programs with the aim of reducing health care and nursing costs.
U2 - 10.1016/j.jcgg.2011.11.007
DO - 10.1016/j.jcgg.2011.11.007
M3 - Journal article
SN - 2210-8335
VL - 3
SP - 29
EP - 35
JO - Journal of Clinical Gerontology and Geriatrics
JF - Journal of Clinical Gerontology and Geriatrics
ER -
