Table 2

Crude and adjusted HRs of haemorrhagic stroke by statin use

Statin use*
OverallEventsNo (reference)Yesp Value
nn (%)HRHR (95% CI)
Haemorrhagic stroke
 Crude67 882421 (0.62)1.001.21 (0.96 to 1.53)0.108
 Age adjusted67 882421 (0.62)1.001.07 (0.85 to 1.35)0.562
 Model 1†60 353371 (0.61)1.000.98 (0.76 to 1.27)0.895
 Model 2‡58 135360 (0.62)1.000.94 (0.70 to 1.28)0.709
Effect modification§
Antiplatelet agents¶0.011
 No48 879287 (0.59)1.000.80 (0.58 to 1.09)
 Yes11 47484 (0.73)1.001.59 (1.03 to 2.47)
Anticoagulants¶0.719
 No60 009361 (0.60)1.000.98 (0.75 to 1.26)
 Yes34410 (2.91)1.001.24 (0.35 to 4.38)
  • *Statin use is modelled as time varying covariate. That is, USE(t)=0 when patient is not taking statin and USE(t)=1 once patient begins taking statin.

  • †Model 1 adjusted for age, income, marital status, enrolment status in the HT, DM and extension study, baseline history of cerebrovascular disease, hypertension, use of antiplatelet agents and anticoagulants.

  • ‡Model 2 adjusted for variables in model 1 plus BMI, waist-to-hip ratio, physical activity, smoking status, baseline history of diabetes, baseline history of high cholesterol requiring pharmacological treatment and baseline history of cardiac disease.

  • §Adjusted for same variables as model 1 and include interactions terms, statin use and antiplatelet use, statin use and anticoagulant use.

  • ¶Antiplatelet agents included aspirin, dipyridamole and ticlopidine; anticoagulants included warfarin and heparin. See text for definitions of each study variable.

  • BMI, body mass index; DM, dietary modification; HT, hormone therapy.