Association Between Use of Benzodiazepines and Development of Dementia

Kyle Walburg, Pharm.D., Broadway Family Medicine

Background: Benzodiazepines are widely prescribed in adults and older adults, but may also be considered concerning as they can cause cognitive deficits. Several studies indicate that benzodiazepine use can increase the risk of developing dementia; however, others have shown they may potentially offer a protective benefit against developing the disease. Other studies have also identified that there may be no association between benzodiazepine use and dementia development. Interestingly, results from these studies can be confounding; prodromal symptoms of dementia such as anxiety or decreased sleep quality can appear almost a decade before dementia is clinically diagnosed. Benzodiazepines are often prescribed to treat these symptoms; as such, studies are often plagued with protopathic bias (when a drug is prescribed to treat symptoms of a yet-to-be diagnosed disease).

Purpose: As the data surrounding this topic is controversial, a group of researchers from Brazil investigated if an association exists between benzodiazepine use and dementia.

Study Design: The study was a systematic review and meta-analysis evaluating 12 studies published between 2011 and 2017: one retrospective and four prospective cohort studies and seven case-control studies. Overall, data from over 980,000 patients was evaluated. Included patients were adults or older adults of any gender; there were no restrictions on previous health conditions. Studies were excluded if patients had diagnoses of dementia at the start of the study. Additionally, studies were excluded if treatment or outcomes were not reported with accuracy. Methodological quality of included studies was assessed by the Newcastle-Ottowa scale and found that the majority of included studies were rated as high quality.

Results: Of the 11 studies included in the meta-analysis (one study evaluated overlapping participants and was not included in the analysis), eight studies determined benzodiazepine use as a risk factor for developing dementia, two studies showed no association, and one study found benzodiazepines offered protection from developing dementia. Evaluating these results, the meta-analysis of these studies concluded that users of benzodiazepines are 1.38 times more likely to develop dementia when compared with never users [95% CI 1.07-1.77]. Additionally, benzodiazepine users experience a 28% higher risk of developing dementia than never users [95% CI 1.06-1.55].

When subgroup analyses of included studies were considered, there was no association between benzodiazepine use and development of dementia. When users of short- and long-acting benzodiazepines were directly compared, there was no significant difference in the development of dementia.

Conclusions: Grading of Recommendations Assessment, Development, and Evaluation (GRADE) rating scoring was used to assess the quality of included evidence; the results of this meta-analysis regarding the association between benzodiazepine use and development of dementia is considered very low quality. Despite including carefully selected evidence and using quality methodology, the heterogeneity of the data had a considerable effect on the low confidence of the results. Although many included studies used a lag time to help minimize protopathic bias (i.e. a latency period before follow-up where benzodiazepines could not be used or dementia could not be diagnosed), there were studies that did not. These studies were more likely to show a lack of association between benzodiazepines and dementia, as perhaps some patients who developed dementia within the timeframe of the study already had dementia before study initiation or use of benzodiazepines, but had not yet been diagnosed. However, it should be noted that when a subgroup analysis was done on only studies that controlled for protopathic bias, authors still found no significant association.

Some key considerations were determined by subgroup analysis. Use of long-acting benzodiazepines was associated with the development of dementia, but use of short-acting benzodiazepines was not. It should be noted that this association was not supported as a statistically significant result of the meta-analysis.

Key Point: The authors of the study recommend a cautious interpretation of the low confidence results suggesting an association between the use of benzodiazepines and development of dementia. Additional long-term, prospective studies are needed that are powered to assess for this outcome. An ambulatory care pharmacist could consider that short-acting benzodiazepines were associated with a lower risk of developing dementia; however, the results were not statistically significant.


1. Lucchetta RC, da Mata BPM, Mastroianni PC. Association between development of dementia and use of benzodiazepines: a systematic review and meta-analysis. Pharmacotherapy. 2018 Oct;38(10):1010-1020.