Abstract
Recent advances in the measurement of health status have resulted in a growing number of generic, multidimensional instruments capable of providing comparable profiles across a broad range of illnesses. One widely popular instrument is the Sickness Impact Profile (SIP) which measures the impact of illness on health-related functioning. While it has established a record of psychometric quality, there have also been concerns about its relatively long length, the internal organization of its subsets, and its utility for less severely disabled populations. This study was designed to develop a short form of the SIP and to evaluate the original and short form with both conceptual and empirical criteria. Using data from a sample of 439 older women with heart disease, item analysis and factor analysis were used to assess the subgrouping of the original SIP and to identify items for the revised SIP. A conceptual framework viewing functioning within physical, mental, and social realms guided the reorganization of items into corresponding categories and dimensions. Item analysis of the full SIP resulted in the elimination of 65 of the original 136 items because of extreme skewness. A principal components analysis provided the rationale to eliminate a further 22 items, and for grouping the remaining 49 items into five functional categories. These factors were interpreted as follows: social and recreational functioning (13 items, Cronbach $\alpha$ =.85); cognitive functioning (8 items, $\alpha$ =.80); physical movement (12 items, $\alpha$ =.80); emotional functioning (6 items, $\alpha$ =.75); and physical activity (10 items, $\alpha$ =.81). The alpha for the total instrument was.93, and the shortened SIP (SIP49) correlated.95 with the full SIP. Factor scores were combined into physical, mental, and social dimension scores. The SIP49 exhibited satisfactory construct validity, as assessed with correlational analysis using 6 minute walk distance, CES-D depressive symptom scores, and the MOS social support scale. Evaluation of the full SIP resulted in conclusions of instrument strengths and weaknesses. While consistently strong in terms of reliability and validity, the SIP exhibits a significant ceiling effect that limits its appropriateness to populations with severe disability. It also blends distinct mental and social functioning into one psychosocial dimension. The SIP49 offers advantages of a shortened instrument with satisfactory psychometric and conceptual qualities.
Showing the abstract — retrieve the full paper via the Exa API.