F J Snoek1, F Pouwer, G W Welch, W H Polonsky. 1. Department of Medical Psychology, Vrije Universiteit Medical Centre, Amsterdam, The Netherlands. fj.snoek.psychol@med.vu.nl
Abstract
OBJECTIVE: To examine the cross-cultural validity of the Problem Areas in Diabetes Scale (PAID) in Dutch and U.S. diabetic patients. RESEARCH DESIGN AND METHODS: A total of 1,472 Dutch people with diabetes completed the PAID along with other self-report measures of affect. Statistics covered Cronbach's alpha, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA), Pearson's product-moment correlation, and t tests. Psychometric properties of PAID were compared for Dutch and U.S. diabetic patients. RESULTS: Internal consistency of the Dutch PAID was high and stable across sex and type of diabetes. Test-retest reliability was high. Principal component analyses confirmed 1 general 20-item factor, whereas EFA identified 4 new subdimensions: negative emotions, treatment problems, food-related problems, and lack of social support. These dimensions were confirmed with CFA and were replicated in the U.S. sample. The PAID and its subscales demonstrated moderate to high associations in the expected direction with other measures of affect. Dutch and U.S. subjects reported having the same problem areas, with U.S. patients reporting higher emotional distress levels both in type 1 and type 2 diabetes. CONCLUSIONS: The Dutch and U.S. 20-item PAID appeared to be psychometrically equivalent, which allowed for cross-cultural comparisons.
OBJECTIVE: To examine the cross-cultural validity of the Problem Areas in Diabetes Scale (PAID) in Dutch and U.S. diabeticpatients. RESEARCH DESIGN AND METHODS: A total of 1,472 Dutch people with diabetes completed the PAID along with other self-report measures of affect. Statistics covered Cronbach's alpha, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA), Pearson's product-moment correlation, and t tests. Psychometric properties of PAID were compared for Dutch and U.S. diabeticpatients. RESULTS: Internal consistency of the Dutch PAID was high and stable across sex and type of diabetes. Test-retest reliability was high. Principal component analyses confirmed 1 general 20-item factor, whereas EFA identified 4 new subdimensions: negative emotions, treatment problems, food-related problems, and lack of social support. These dimensions were confirmed with CFA and were replicated in the U.S. sample. The PAID and its subscales demonstrated moderate to high associations in the expected direction with other measures of affect. Dutch and U.S. subjects reported having the same problem areas, with U.S. patients reporting higher emotional distress levels both in type 1 and type 2 diabetes. CONCLUSIONS: The Dutch and U.S. 20-item PAID appeared to be psychometrically equivalent, which allowed for cross-cultural comparisons.
Authors: Melissa M Franks; Rachel C Hemphill; Amber J Seidel; Mary Ann Parris Stephens; Karen S Rook; James K Salem Journal: Aging Ment Health Date: 2012-04-26 Impact factor: 3.658
Authors: Paulien R Wermeling; Maureen van den Donk; Kees J Gorter; G Ardine de Wit; Yolanda van der Graaf; Guy Ehm Rutten Journal: BMC Fam Pract Date: 2010-05-11 Impact factor: 2.497