The effect of an incentive billing code on heart failure management in primary care: a population-based study

Background: To support family physicians (FPs) in managing patients with heart failure (HF), the Ministry of Health in Ontario, Canada, implemented the Q050 billing code in 2008, a pay-for-performance (P4P) incentive for guideline-based HF care. We studied whether the incentive was associated with any change in process-of-care measure, particularly the prescriptions of HF medications. Methods: We identified all patients with HF in Ontario of age>65, who were managed by FPs claiming the Q050 incentive between 2008 and 2021. We counted the number of patients who were prescribed renin-angiotensin system inhibitors (RASi), beta-blockers (BB), mineralocorticoid receptor antagonists (MRA), and diuretics three months before and after the Q050 billing code was claimed for these patients. Where applicable, we classified the agents within each class by whether they are guideline-directed as recommended by the Canadian Cardiovascular Society (CCS). Results: We included 39,425 HF patients in the study. The median age was 80 (IQR 73-85) years; 49% were female. Half were assessed by an internist or cardiologist during the six months before their HF diagnosis. Compared to pre-Q050, there was an increase in RASi prescriptions from 42.5% to 45.8%, BB from 51.9% to 54.4%, MRA from 9.2% to 11.7%, and diuretics from 63.2% to 65.7% after the incentive (p<0.05). There was a decrease in those not on any HF medications from 27.5% to 24.9% (p<0.001). Those with newly diagnosed HF and prompt follow-up with FPs experienced the largest but clinically modest increase in HF medications. Conclusions: To our knowledge, this is the first evaluation of process-of-care measures related to a pay-for-performance program in primary care HF management. The Q050 incentive led to a minimal increase in the prescription of HF medications; there is underutilization of disease-modifying agents. Further research is needed to understand why pay-for-performance programs had no effect on physician prescribing behaviours.