Comparison of persistence rates with angiotensin-converting enzyme inhibitors used in secondary and primary prevention of cardiovascular disease.

OBJECTIVES: On average, 50% of patients are noncompliant with drugs for chronic health problems, despite their proven efficacy. It is therefore essential to have real-world data to devise suitable methods for improving persistence with these therapies. To measure and compare persistence rates with the angiotensin-converting enzyme inhibitors (ACEIs) in primary and secondary prevention and their determinants.
METHODS: Two cohorts were reconstructed from the Régie de l'assurance maladie du Québec's databases. The subjects had to be newly treated with ACEIs between January 1, 1998 and December 31, 2000. The primary prevention cohort consisted of 4596 hypertensive patients and the secondary prevention cohort of 1620 patients. The cumulative persistence rates were determined by the Kaplan-Meier method. The determinants of nonpersistence were evaluated with a Cox regression model.
RESULTS: The 1-year persistence rates for the nonexclusive use of antihypertensive agents by initial prescribed agent: enalapril, fosinopril, lisinopril, quinapril, and ramipril were 66%, 64%, 69%, 65%, and 72% in the secondary prevention cohort, and of 66%, 72%, 71%, 72%, and 75% in the primary prevention cohort. The adjusted 1.5-year nonpersistence rates in primary prevention were higher for quinapril and enalapril than for ramipril. In secondary prevention all of the ACEIs were equivalent in nonpersistence rate. In secondary prevention cohort, having dyslipidemia, respiratory disease, >or=4 different classes of drugs/month increase the rate of persistence. Among, the primary prevention cohort, the fact of having diabetes, dyslipidemia, respiratory disease, using >or=4 different classes of drugs/month or prior hospitalization increased significantly the rate of persistence. For both cohorts, the fact of having high number of oral doses/day or elevated health-care resource utilization decreased significantly the rate of persistence.
CONCLUSION: The 1.5-year persistence rate was low compared with the threshold of 80% generally accepted. The high-risk patients were less likely to discontinue their treatment. These results can be of help in devising methods for improving the effectiveness of these drugs in routine practice.