Effects of nonpersistence with medication on outcomes in high-risk patients with cardiovascular disease.

BACKGROUND The impact of nonpersistence on events and of events on persistence is unclear. We studied the effects of nonpersistence on outcomes and events on nonadherence in a randomized placebo controlled trial in 40 countries on 25,620 patients. METHODS In the ONgoing Telmisartan Alone and in combination with Ramipril Global Endpoint Trial (ONTARGET), persistent patients (n = 20,991) were compared with individuals who had permanently stopped study medications (n = 4,629). RESULTS Older age, female gender, less physical activity, less education, and history of stroke/transient ischemic attack, depression, and diabetes were associated with nonpersistence. After adjustment, nonpersistence was associated with the composite end point of cardiovascular death, myocardial infarction, stroke, or hospitalization for heart failure (hazard ratio 1.24, 99% CI 1.09-1.40, P < .0001), cardiovascular death alone (1.87, 1.60-2.19, P < .0001), and heart failure hospitalization alone (1.32, 1.04-1.67, P = .0023). Cardiovascular events increased when medications were stopped, whereas noncardiovascular outcomes did not. Nonpersistence rapidly increased within the first year after nonfatal events such as myocardial infarction (hazard ratio 3.37, 99% CI 2.72-4.16, P < .0001), stroke (3.25, 2.59-4.07, P < .0001), and hospitalization for heart failure (3.67, 2.95-4.57, P < .0001). Persistence was poorer with more frequent and earlier events. Patients stopping medication after an event were at greater risk for subsequent events. CONCLUSIONS Improving medications persistence could interrupt this vicious circle and may improve outcomes.