Stewart A, Ware J E, Brook R H
Med Care. 1977 Nov;15(11):939-52. doi: 10.1097/00005650-197711000-00008.
Results of psychometric studies of 14 questionnaire items commonly used to define chronic functional limitations due to poor health are reported. Self-administered questionnaires were used to gather data from 1,209 persons 14 years of age and older. Data were used to study: scalability of items; test-retest reliability of alternate forms of scales; validity of scales in relation to 13 health status variables and age; and precision of scales in detecting differences in health. Three scales pertaining to chronic limitations in mobility, physical activity, and social role activity functions satisfied the criteria of scalogram analysis. Four-month test-retest reliability estimates for alternate forms were very high. Strong associations (some curvilinear) were observed among functional limitation scales, and between these scales and survey measures of physical abilities, general health perceptions, health worry/concern, chronic disease conditions, and age. Measures of physical abilities and functional limitations appeared to define opposite ends of a function-dysfunction continuum. Statistical modeling of precision indicates that, due to the skewed distribution of scores, large sample sizes would be required to detect differences in functional limitations in studies using only a posttest.
