[Skip to Content]
[Skip to Content Landing]
Views 1,057
Citations 0
Comment & Response
July 9, 2019

Analysis of Long-term Benefits of Intensive Blood Pressure Control

Author Affiliations
  • 1Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, Massachusetts
  • 2Institute for Aging Research, Hebrew SeniorLife, Boston, Massachusetts
JAMA. 2019;322(2):169-170. doi:10.1001/jama.2019.5840

To the Editor The SPRINT MIND Investigators1 assessed whether intensive blood pressure control would reduce risk of probable dementia compared with standard blood pressure control. There were 149 probable dementia events among 4678 participants in the intensive treatment group and 176 events among 4683 participants in the standard treatment group. The hazard ratio was 0.83 (95% CI, 0.67-1.04; P = .10). Based primarily on the hazard ratio analysis, the authors concluded that intensive blood pressure control did not significantly reduce risk of probable dementia. However, valid hazard ratio analysis requires that the ratio of hazards in the 2 treatment groups is constant over time.2-4 The authors acknowledged that this proportional hazards assumption may not have held, which was also indicated graphically by the initial overlap and subsequent separation of the Kaplan-Meier curves in Figure 2 in the article. The increasing separation of the incidence curves after 4 years suggests a potential long-term benefit, which may be biologically plausible. Numerous dementia cases were observed in the standard treatment group after 6 years of follow-up, whereas few cases were observed in the intensive treatment group. Unfortunately, since the proportional hazards assumption likely did not hold, the hazard ratio analysis might have lacked power to detect a true difference between the treatment groups.