Psychometric properties of the Revised Illness Perception Questionnaire: Factor structure and reliability among African-origin populations with type 2 diabetes

Abstract

BackgroundThe Revised Illness Perception Questionnaire (IPQ-R) has been used extensively to measure illness perceptions of several patient populations. However, the instrument was developed using participants of mainly European-origin. The reliability and validity of the IPQ-R may therefore need to be established before use among populations of different ethnic and cultural origins. ObjectivesThis study investigated the factor structure and internal consistency reliability of the IPQ-R in African-origin patients with type 2 diabetes. MethodsIn this cross-sectional study, 221 adults of African descent with type 2 diabetes completed the IPQ-R. Participants were recruited from patients attending diabetes and retinal screening clinics in the London boroughs of Brent and Hackney. Confirmatory Factor Analysis based on the covariance matrix was used to determine factorial validity for the Timeline-acute/chronic, Consequences, Personal control, Treatment control, Illness coherence, Timeline-cyclical, Emotional representation and three causal subscales of the IPQ-R. Composite internal consistency reliability for individual subscales was determined using Cronbach's alpha coefficients. ResultsAfter eliminating three items and re-specifying six error covariances associated with large standardised residuals and low factor loadings, the hypothesised model adequately explained the covariance of African and Caribbean patients’ responses to items of the IPQ-R. Also, composite reliability coefficients of all measured subscales were acceptable and inter-correlations between subscales were in line with those reported from other population groups. ConclusionsThe findings in this study suggest that although the IPQ-R may be valid and reliable across cultures, investigators may need to modify (e.g. by rewording) some of its items taking into account any linguistic origins of their populations of study. Further evaluation of the IPQ-R (including the identity subscale) in larger samples of African-origin populations is also recommended.