J Twiss1, S P McKenna. 1. Galen Research Ltd, Enterprise House, Manchester Science Park, Lloyd Street North, Manchester, M15 6SE, UK, jtwiss@galen-research.com.
Abstract
BACKGROUND: Disease-specific patient-reported outcome (PRO) measures are designed to be highly relevant to one disease. It is widely believed that comparisons of outcomes between patients with different diseases are only possible using generic measures. The present study employs a novel method of using Rasch analysis to co-calibrate scores from different disease-specific PRO measures, allowing scores to be compared across diseases. METHODS: Psoriasis patients (n = 146, mean age = 44.4, males = 50 %) completed the Psoriasis Quality of Life scale (PSORIQoL) and atopic dermatitis patients (n = 146, mean age = 45.5, males = 50 %) the Quality of Life in Atopic Dermatitis scale (QoLIAD). Both measures employ the needs-based model of QoL, and they share five common items-providing a link between assessments. The groups were analysed separately, and then combined to test fit to the Rasch model. RESULTS: Both scales showed good fit to the Rasch model after minor adjustments (PSORIQoL: χ (2) p = 0.25; QoLIAD: χ (2) p = 0.51). For the combined dataset, one common item showing differential item functioning by disease was removed and fit to the Rasch model was achieved (χ (2) p = 0.08). The co-calibrated scale successfully distinguished between perceived severity groups (p < 0.001). CONCLUSIONS: It is possible to co-calibrate scores on the PSORIQoL and QoLIAD. This is one of the first studies in health research to demonstrate how Rasch analysis can be used to make comparisons across diseases using different disease-specific measures. Such an approach maintains the greater relevance and, consequently, accuracy associated with disease-specific measurement.
BACKGROUND: Disease-specific patient-reported outcome (PRO) measures are designed to be highly relevant to one disease. It is widely believed that comparisons of outcomes between patients with different diseases are only possible using generic measures. The present study employs a novel method of using Rasch analysis to co-calibrate scores from different disease-specific PRO measures, allowing scores to be compared across diseases. METHODS:Psoriasispatients (n = 146, mean age = 44.4, males = 50 %) completed the Psoriasis Quality of Life scale (PSORIQoL) and atopic dermatitispatients (n = 146, mean age = 45.5, males = 50 %) the Quality of Life in Atopic Dermatitis scale (QoLIAD). Both measures employ the needs-based model of QoL, and they share five common items-providing a link between assessments. The groups were analysed separately, and then combined to test fit to the Rasch model. RESULTS: Both scales showed good fit to the Rasch model after minor adjustments (PSORIQoL: χ (2) p = 0.25; QoLIAD: χ (2) p = 0.51). For the combined dataset, one common item showing differential item functioning by disease was removed and fit to the Rasch model was achieved (χ (2) p = 0.08). The co-calibrated scale successfully distinguished between perceived severity groups (p < 0.001). CONCLUSIONS: It is possible to co-calibrate scores on the PSORIQoL and QoLIAD. This is one of the first studies in health research to demonstrate how Rasch analysis can be used to make comparisons across diseases using different disease-specific measures. Such an approach maintains the greater relevance and, consequently, accuracy associated with disease-specific measurement.
Authors: Tamar Nijsten; David M Meads; John de Korte; Francesca Sampogna; Joel M Gelfand; Katia Ongenae; Andrea W Evers; Matthias Augustin Journal: J Invest Dermatol Date: 2007-05-10 Impact factor: 8.551
Authors: S P McKenna; S A Cook; D Whalley; L C Doward; H L Richards; C E M Griffiths; D Van Assche Journal: Br J Dermatol Date: 2003-08 Impact factor: 9.302
Authors: Richard Shikiar; Mary Kaye Willian; Martin M Okun; Christine S Thompson; Dennis A Revicki Journal: Health Qual Life Outcomes Date: 2006-09-27 Impact factor: 3.186