Validation of the disease burden morbidity assessment by self-report in a French-speaking population
1 Département de médecine de famille, Université de Sherbrooke, Québec, Canada
2 Centre de santé et de services sociaux de Chicoutimi, Québec, Canada
3 Faculty of medicine, McGill University, Montreal, Canada
4 Centre de santé et de services sociaux de Chicoutimi, 305 St-Vallier, Chicoutimi, Québec G7H 5H6, Canada
BMC Health Services Research 2012, 12:35 doi:10.1186/1472-6963-12-35Published: 14 February 2012
The Disease Burden Morbidity Assessment (DBMA) is a self-report questionnaire used to estimate the disease burden experienced by patients. The aim of this study was to test and to measure the properties of the French translation of the DBMA (DBMA-Fv).
The original version of the DBMA was translated into French (Canadian) and first assessed during cognitive interviews. In the validation study, patients recruited during consecutive consultation periods completed the DBMA-Fv questionnaire while they were in the waiting room of a primary care setting (T1). Participants completed the same questionnaire mailed to their home two weeks later (T2). Concomitant validity of the DBMA-Fv was assessed using the Cumulative Illness Rating Scale (CIRS). Patient medical records were reviewed to verify chronic diseases and past medical history.
Ninety-seven patients were recruited and 85 (88%) returned the mailed questionnaires; 5 (5.9%) were incomplete. DBMA-Fv scores of the 80 participants with a complete questionnaire at T2 ranged from 0 to 30 (median 5.5, mean 7.7, SD = 7.0). Test-retest reliability of the DBMA-Fv was high (ICC: 0.86, 95% CI: 0.79-0.92). The DBMA-Fv and the CIRS correlated moderately at T1 (r = 0.46, 95% CI: 0.26 - 0.62, p < 0.01) and T2 (r = 0.56, 95% CI: 0.38 - 0.70, p < 0.01). The mean (SD) sensitivity of patient reports of a condition in relation to chart review at T2 was 73.9 (8.4) (range 62.5% to 90%). The overall mean (SD) specificity was 92.2 (6.7) (range 77.6% to 98.6%).
The DBMA-Fv's properties are similar to its English counterpart as to its median sensitivity and specificity compared to chart reviews. It correlated moderately with an established index of multimorbidity. A high percentage of patients were able to complete the test correctly as a mail questionnaire and it showed high test-retest reliability.