Computerized adaptive testing of diabetes impact: A feasibility study of Hispanics and non-Hispanics in an active clinic population
- 26 September 2006
- journal article
- Published by Springer Science and Business Media LLC in Quality of Life Research
- Vol. 15 (9), 1503-1518
- https://doi.org/10.1007/s11136-006-0008-x
Abstract
Diabetes is a leading cause of death and disability in the US and is twice as common among Hispanic Americans as non-Hispanics. The societal costs of diabetes provide an impetus for developing tools that can improve patient care and delay or prevent diabetes complications.We implemented a feasibility study of a Computerized Adaptive Test (CAT) to measure diabetes impact using a sample of 103 English- and 97 Spanish-speaking patients (mean age = 56.5, 66.5% female) in a community medical center with a high proportion of minority patients (28% African-American). The 37 items of the Diabetes Impact Survey were translated using forward-backward translation and cognitive debriefing. Participants were randomized to receive either the full-length tool or the Diabetes-CAT first, in the patient's native language.The number of items and the amount of time to complete the survey for the CAT was reduced to one-sixth the amount for the full-length tool in both languages, across disease severity. Confirmatory Factor Analysis confirmed that the Diabetes Impact Survey is unidimensional. The Diabetes-CAT demonstrated acceptable internal consistency reliability, construct validity, and discriminant validity in the overall sample, although subgroup analyses suggested that the English sample data evidenced higher levels of reliability and validity than the Spanish sample and issues with discriminant validity in the Spanish sample. Differential Item Function analysis revealed differences in responses tendencies by language group in 3 of the 37 items. Participant interviews suggested that the Spanish-speaking patients generally preferred the paper survey to the computer-assisted tool, and were twice as likely to experience difficulties understanding the items.While the Diabetes-CAT demonstrated clear advantages in reducing respondent burden as compared to the full-length tool, simplifying the item bank will be necessary for enhancing the feasibility of the Diabetes-CAT for use with low literacy patients.Keywords
This publication has 19 references indexed in Scilit:
- Standards of Medical Care in DiabetesDiabetes Care, 2005
- Measuring Quality of Life in Routine Oncology Practice Improves Communication and Patient Well-Being: A Randomized Controlled TrialJournal of Clinical Oncology, 2004
- A Generalized Partial Credit ModelPublished by Springer Science and Business Media LLC ,1997
- A 12-Item Short-Form Health SurveyMedical Care, 1996
- Assessment of Diabetes-Related DistressDiabetes Care, 1995
- Linking Clinical Variables With Health-Related Quality of LifeJAMA, 1995
- The Effect of Intensive Treatment of Diabetes on the Development and Progression of Long-Term Complications in Insulin-Dependent Diabetes MellitusThe New England Journal of Medicine, 1993
- Effect of Antihypertensive Therapy on the Kidney in Patients with Diabetes: A Meta-Regression AnalysisAnnals of Internal Medicine, 1993
- Appropriateness measurement with polychotomous item response models and standardized indicesBritish Journal of Mathematical and Statistical Psychology, 1985
- Significance tests and goodness of fit in the analysis of covariance structures.Psychological Bulletin, 1980