Association between benzodiazepine use and dementia: nested case-control study

K. Richardson1, K. Mattishent2, Y. K. Loke2, I. Maidment3, N. Steel2, F. Matthews4, C. Fox2, A. Arthur1, C. M. Grossi1, K. Bennett5, P. K. Myint6, N. Campbell7, M. Boustani8, C. Brayne9, L. Robinson10, G. Savva1. 1School of Health Sciences, University of East Anglia, Norwich, UNITED KINGDOM, 2Norwich Medical School, University of East Anglia, Norwich, UNITED KINGDOM, 3Aston University, Birmingham, UNITED KINGDOM, 4Institute for Health and Society, Newcastle University, Newcastle Upon Tyne, UNITED KINGDOM, 5Division of Population Health Sciences, Royal College of Surgeons in Ireland, Dublin, IRELAND, 6Institute for Applied Health Sciences, University of Aberdeen, Aberdeen, UNITED KINGDOM, 7College of Pharmacy, Purdue University, Indianapolis, 8Health Information and Translational Sciences, University of Indiana, Indianapolis, 9Department of Public Health and Primary Care, University of Cambridge, Cambridge, UNITED KINGDOM, 10Institute for Ageing, Newcastle University, Newcastle Upon Tyne, UNITED KINGDOM.

Introduction Benzodiazepines can have short term adverse effects on cognition but there are conflicting findings regarding risk of serious long-term harm e.g. dementia. We aimed to evaluate the association between long-term use of benzodiazepines (including benzodiazepine derivatives) and dementia.

Methods We conducted a nested case-control study using the UK Clinical Practice Research Datalink. We selected participants of age 65-99 years (time period 2006-2015) based on new dementia diagnosis or prescription for a dementia drug and subsequent dementia diagnosis, with up to 7 controls matched on gender, age, deprivation index and data history. We considered the relevant exposure period to be 4-20 years prior to incident dementia, and we assessed dose-response using Defined Daily Doses for benzodiazepines and z-drugs. We used conditional logistic regression to estimate adjusted odds ratios (aOR) and 95% Confidence Intervals (CI) for dementia and drug use, adjusted for potential confounders.

Results We selected 40,770 dementia cases and 283,933 controls, median age 74 years, 60% female. Mean drug exposure period was 7.8 years. When comparing new users to non-users (adjusted for baseline covariates), benzodiazepine use was not associated with a significant risk of dementia (aOR 1.03, 95% CI 1.00 -1.07) whereas Z-drugs had a small association with dementia (aOR 1.07, 95% CI 1.02 - 1.12). Residual confounding is a potential limitation and we did not identify a plausible cumulative dose relationship for the link between dementia and defined daily doses of benzodiazepines or Z-drugs.

Conclusions Despite the findings of our nested case-control study, benzodiazepines can cause a wide range of other adverse effects, and clinicians should continue to follow guidelines on avoiding potentially inappropriate prescription of long-term benzodiazepines, even in the absence of a definitive evidence of any association with dementia. Interestingly, the results of this study conflict with recent systematic review findings. This research was supported by funding from Alzheimer’s Society (AS-PG-2013-017).