Psychometric analysis of the TRANSIT quality indicators for cardiovascular disease prevention in primary care

To assess a selection of psychometric properties of the TRANSIT indicators.

Using medical records, indicators were documented retrospectively during the 14 months preceding the end of the TRANSIT study.

Primary care in Quebec, Canada.

Indicators were documented in a random subsample (n = 123 patients) of the TRANSIT study population (n = 759).

For every patient, the mean compliance to all indicators of a category (subscale score) and to the complete set of indicators (overall scale score) were established. To evaluate test-retest and inter-rater reliabilities, indicators were applied twice, two months apart, by the same evaluator and independently by different evaluators, respectively. To evaluate convergent validity, correlations between TRANSIT indicators, Burge et al. indicators and Institut national d'excellence en santé et en services sociaux (INESSS) indicators were examined.

Test-retest reliability, inter-rater reliability, and convergent validity.

Test-retest reliability, as measured by intraclass correlation coefficients (ICCs) was equal to 0.99 (0.99-0.99) for the overall scale score while inter-rater reliability was equal to 0.95 (0.93-0.97) for the overall scale score. Convergent validity, as measured by Pearson's correlation coefficients, was equal to 0.77 (P < 0.001) for the overall scale score when the TRANSIT indicators were compared to Burge et al. indicators and to 0.82 (P < 0.001) for the overall scale score when the TRANSIT indicators were compared to INESSS indicators.

Reliability was excellent except for eleven indicators while convergent validity was strong except for domains related to the management of CVD risk factors.