OBJECTIVE: To assess patterns of restarting antihypertensive drugs after a prolonged period of discontinuation. METHODS: We conducted a retrospective cohort study among new users of blood pressure-lowering medication in the PHARMO database in the Ntherlands, who had a period of at least 180 days without such medication. A multivariable Cox proportional hazard analysis was used to explore the baseline variables associated with reinitiating treatment. Case-crossover analysis was used to evaluate determinants of reinitiating treatment. RESULTS: We identified 35 714 patients as initiating blood pressure-lowering treatment during the period 1 January 1999 to 30 June 2004. Of the 18 357 (51.4%) patients who discontinued blood pressure-lowering treatment, 19.3% restarted treatment within 1 year and 60.7% restarted within 6 years. With every additional year they had been on therapy, patients were more likely to restart [odds ratio (OR) = 1.38; 95% confidence interval (CI) = 1.34-1.42]. The case-crossover analysis revealed that hospitalization for cardiovascular disease (OR = 2.20; 95% CI = 1.84-2.63), as well as refilling of another cardiovascular medication (OR = 1.25; 95% CI = 1.11-1.40), were each independently associated with reinitiating treatment. Refilling non-cardiovascular medications was not associated with reinitiating treatment (OR = 1.03; 95% CI = 0.97-1.10). CONCLUSION: Physicians should be aware that many patients have prolonged periods of discontinuation during the use of blood pressure-lowering medication, and that most of these patients will eventually resume therapy. Ongoing refilling other medications is not associated with reinitiating treatment. This suggests that, for some patients, the decision to discontinue may be drug specific rather than a behavioural characteristic applicable to all chronic treatments.