Table 2.

Hazard ratio for selected outcomes by tertile of SBV

OutcomeNumber of EventsAnalysisHazard Ratio by SBV Tertilea
1b23
Death64Unadjusted1.38 (0.66–2.88)2.89 (1.50–5.56)
Adjusted0.77 (0.28–2.16)2.82 (1.14–6.95)
Cardiovascular death22Unadjusted1.23 (0.33–4.59)3.14 (1.02–9.63)
Adjusted1.09 (0.21–5.57)4.91 (1.12–21.50)
Cardiovascular events133Unadjusted1.68 (1.03–2.76)2.81 (1.77–4.45)
Adjusted1.28 (0.71–2.29)1.23 (0.65–2.34)
Renal events255Unadjusted1.08 (0.78–1.48)1.58 (1.17–2.14)
Adjusted1.00 (0.66–1.51)1.05 (0.67–1.62)
  • Adjustment includes age (continuous), sex, history of heart disease (yes/no), smoking status (current, prior, never), body mass index (continuous), mean systolic BP during (from 3 to 12 months, continuous), proteinuria at 12 months (continuous), serum albumin at 12 months (continuous), measured iothalamate GFR at 12 months (continuous), study drug assignment (metoprolol, ramipril, amlodipine), and compliance with study medications over the first study year (proportion of study visits). SBV, systolic BP visit-to-visit variability.

  • a Hazard ratio of event relative to the lowest tertile of SBV. Hazard ratios for death, cardiovascular death, and renal events are calculated using Cox proportional hazards regression, whereas hazard ratios for cardiovascular events are calculated using Anderson–Gill regression.

  • b Reference.