Risk of death after first admission for cardiovascular diseases by country of birth in The Netherlands: a nationwide record-linked retrospective cohort study

Objective: To examine differences in short- (28 days) and long-term (5 years) risk of death in patients hospitalised for the first time for various cardiovascular diseases (CVD) by country of birth and/or parental country of birth. Design: A nationwide prospective cohort of CVD patients. Settings: Entire Netherlands. Patients: 118 691 patients hospitalised for the first time for various CVDs were identified through the national hospital discharge, the Dutch population and the cause-of-death registers. Main outcome measures: Differences in short- term and long-term risk of death. Cox proportional hazard models were used to estimate the mortality hazard ratios. Results: After adjusting for age, compared with Dutch patients, Turkish, other non-Western and Western migrants had both a short- and long-term higher risk, while Suriname patients had only a long-term higher risk of total-mortality and combined-CVD mortality. These higher rates were driven mainly by an increased risk of short- term (hazard ratio 3.21; 95% CI 1.03 to 10.03) and long-term (2.29; 1.14 to 4.60) mortality following congestive heart failure (CHF) among Turkish; short- term (1.56; 1.10 to 2.20) and long-term (1.50; 1.11 to 2.01) mortality following cerebrovascular accident (CVA) among the other non-Western migrants; short- term mortality following CVA (1.10; 1.01 to 1.19) and long-term mortality following CVA (1.10; 1.03 to 1.17), and, to a lesser extent, CHF and myocardial infarction among Western migrants; and a long-term mortality following CVA (1.29; 1.05 to 1.57) among Surinamese patients. Conclusion: Higher mortality after a first episode of CVD was found in ethnic minority patients than in Dutch patients. These differences hardly changed after adjusting for possible confounders, suggesting that treatment and secondary prevention strategies may be less effective in these groups. More research is needed to explain the possible causes of these inequalities.