Time-trends in treatment and cardiovascular events in patients with heart failure: a pharmacosurveillance study

Aims We assessed, in patients with a first hospitalization for heart failure (HF), the temporal relationship of the incidence of cardiovascular events, all-cause mortality, and cardiovascular drug treatment. Methods and results Data were obtained from the PHARMO Record Linkage System, a population-based registry of pharmacy records linked with hospital discharge records in The Netherlands. Patients were selected based on a first hospital discharge diagnosis of documented HF. Two time-periods were compared: 1998-2002 and 2003-07. In each time-period, we analysed all prescribed cardiovascular medications, all-cause mortality, and cardiovascular events (rehospitalization for HF and ischaemic events) within the first year after hospitalization, and the occurrence of ischaemic events separately (myocardial infarction and ischaemic stroke). Cox-regression analysis was performed to calculate hazard ratios (HR) with 95% confidence intervals (CI). We identified 8276 patients in 1998-2002 and 9548 patients from 2003-07. There was an increase in almost all cardiovascular medication prescriptions in the second period: in particular, beta-blocker prescriptions rose from 36% in 1998-2002 to 55% in 2003-07. In the first year after hospitalization, there was no difference in all-cause mortality or any cardiovascular event (HR 1.00, 95% CI: 0.95-1.05), as a composite endpoint or when analysed separately. The incidence of ischaemic events decreased from 2.7 to 1.9% in the first and second time-period, respectively (HR 0.74, 95% CI: 0.61-0.90). Conclusion Prescription of cardiovascular medications in patients with a first hospitalization for HF has increased in recent years, particularly for beta-blockers, and the incidence of ischaemic events may have decreased. There was no decrease in all-cause mortality or cardiovascular events.