Predicting Clinical Outcomes Related to Cardiovascular Disease Prevention in Primary Care Using the TRANSIT Indicators (303392)

Céline Bareil, HEC Montreal 
Fabie Duhamel, University of Montreal 
Marie-Mireille Gagnon, University of Montreal Hospital Research Centre 
Johanne Goudreau, University of Montreal 
Eveline Hudon, University of Montreal 
*Cynthia Khanji, University of Montreal 
Gilles Lalonde, Médi-Centre Chomedey 
Lyne Lalonde, University of Montreal 
Marie-Thérèse Lussier, University of Montreal 
Sylvie Perreault, University of Montreal 
Mireille Schnitzer, University of Montreal 
Alain Turcotte, Centre de Santé et de Services Sociaux du Lac-des-Deux-Montagnes 

Keywords: quality indicator, cardiovascular disease, predictive validity, primary care

In Quebec, the creation of family medicine group (FMG) clinics and legislative changes allowing nurses and pharmacists to expand their scope of practice may improve cardiovascular disease prevention in primary care. Reliable and valid quality indicators are essential to estimate changes in the quality of primary care. In the TRANSIT study, a participatory research program, primary care actors and researchers developed a set of 81 quality indicators.

The objective of this study is to identify which TRANSIT indicator has the capacity to predict intermediate outcomes, such as the achievement of therapeutic targets for hypertension, dyslipidemia, and diabetes at the end of the study. Methods: Eight FMGs including 102 clinicians and 759 patients participated in the TRANSIT study. Indicators were documented by medical chart review during a 14 months’ follow-up period. Patient’s characteristics were described at baseline. The relationship between the TRANSIT indicators and each of the three clinical outcomes was established using univariate and multivariate logistic regression. Crude and adjusted odds ratios (OR) with 95% confidence interval were computed for each indicator. The Least Absolute Shrinkage and Selection Operator (LASSO) was also used to assess the predictive validity of indicators TRANSIT because this method is more efficient than the logistic regression to identify a limited number of predictors.

For the hypertension model, two indicators were selected by the univariate logistic regression, the multivariate logistic regression, and the LASSO. These indicators are related to blood pressure measurements as recommended in guidelines for hypertensive patients and instructions for home blood glucose monitoring.

Evaluating the predictive validity of the TRANSIT indicators may be useful to reduce the number of indicators and therefore increase the feasibility of using them in current practice.