Cardiovascular Outcomes and Antihypertensive Drug Treatment in Older Women—Reply

TLDR: Concerns are raised that the number of incident events was small, the P values for the overall comparison of therapies only reached borderline statistical significance for cardiovascular mortality, and the confidence intervals reported for the hazard ratios have not been adjusted for multiple comparisons, increasing the likelihood of a chance finding.

Abstract: To the Editor: The cohort study of cardiovascular outcomes and antihypertensive drug treatment by Dr Wassertheil-Smoller and colleagues used data from the Women’s Health Initiative Observational Study. We have a number of concerns that we believe limit drawing conclusions from this study. First, relative to the study sample size, the number of incident events was small and therefore the study may not have had adequate power to detect the end points in each subgroup (Tables 4 and 5 in the original article). For example, among women receiving a calcium channel blocker (CCB) plus an angiotensin-converting enzyme (ACE) inhibitor, only 8 cardiovascular deaths were reported. The P values for the overall comparison of therapies only reached borderline statistical significance for cardiovascular mortality, the outcome with the lowest annualized rates. Second, the confidence intervals reported for the hazard ratios in Tables 5 and 7 in the original article have not been adjusted for multiple comparisons, increasing the likelihood of a chance finding. Finally, the authors report the participants’ use of antihypertensive medications at baseline and in year 3 of follow-up, but did not account for changes in antihypertensive medications that likely occurred over time.