Affiliations: [a]
Kaiser Permanente Health Research Institute, Seattle, WA, USA
| [b] Departments of Pharmacy, University of Washington, Seattle, WA, USA
| [c] Department of Neurology, Swedish Hospital Medical Center, Seattle, WA, USA
| [d] Department of Medicine, University of Washington, Seattle, WA, USA
| [e] Departments of Psychosocial and Community Health, University of Washington, Seattle, WA, USA
Correspondence:
[*]
Correspondence to: Paul K. Crane, MD MPH, Box 359780, 325 Ninth Avenue, Seattle, WA 98104, USA. Tel.: +1 206 744 1831; Fax: +1 206 897 4688; E-mail: [email protected].
Note: [1] These authors contributed equally to this work and should be considered co-senior authors.
Abstract: Background:Higher glucose levels are associated with dementia risk in people with and without diabetes. However, little is known about how this association might vary by hypertension status and antihypertensive treatment. Most studies on modifiable dementia risk factors consider each factor in isolation. Objective:To test the hypothesis that hypertension and antihypertensive treatments may modify associations between glucose levels and dementia. Methods:Analyses of data generated from a research study and clinical care of participants from a prospective cohort of dementia-free older adults, including glucose measures, diabetes and antihypertensive treatments, and blood pressure data. We defined groups based on blood pressure (hypertensive versus not, ≥140/90 mmHg versus <140/90 mmHg) and antihypertensive treatment intensity (0, 1, or ≥2 classes of antihypertensives). We used Bayesian joint models to jointly model longitudinal exposure and time to event data. Results:A total of 3,056 participants without diabetes treatment and 480 with diabetes treatment were included (mean age at baseline, 75.1 years; mean 7.5 years of follow-up). Higher glucose levels were associated with greater dementia risk among people without and with treated diabetes. Hazard ratios for dementia were similar across all blood pressure/antihypertensive treatment groups (omnibus p = 0.82 for people without and p = 0.59 for people with treated diabetes). Conclusion:Hypertension and antihypertensive treatments do not appear to affect the association between glucose and dementia risk in this population-based longitudinal cohort study of community-dwelling older adults. Future studies are needed to examine this question in midlife and by specific antihypertensive treatments.
