Use of Sedative-Hypnotic Medications and Risk of Dementia: A Systematic Review and Meta-Analysis
Abstract
Abstract
Background: Growing evidence suggests an association between use of sedative-hypnotic
medications and risk of dementia. However, results from population-based studies have been
conflicting, possibly owing to methodological variations, confounding by indication and protopathic
bias, making it difficult to draw conclusive evidence on this association.
Objectives: To examine the possible association between sedative-hypnotic medications use and the
risk of dementia through quantitative synthesis of current evidence using a meta-analysis approach.
Methods: MEDLINE (PubMed) and SCOPUS were systematically searched from the inception of both
databases to May 2020, and EMBASE from May 2010 to May 2020. The search was conducted for
articles that assessed the risk of dementia with sedative-hypnotics use. Extracted records were
evaluated using the Newcastle-Ottawa scale, and an overall odds ratio was pooled using randomeffects model.
Key findings: 35 articles were included in this systematic review and meta-analysis with 4,257,670
participants. Pooled odds ratios (ORs) for dementia were (OR:1.33, 95% CI 1.19-1.49) for unstratified
BDZ use, (OR:1.46, 95% CI 1.23-1.73) for short-acting BDZ use, (OR:1.72, 95% CI 1.48-1.99) for longacting BDZ use, (OR:1.13, 95% CI 0.97-1.32) for BDZ use (after stratification), (OR:1.64, 95% CI 1.13-
2.38) for BDZ and Z-drugs combined use and (OR: 1.43, 95% CI 1.17-1.74) for Z-drugs only use.
However, all these associations did not persist when studies that did not adjust for protopathic bias
were excluded except for short and long-acting benzodiazepines. Furthermore, this analysis failed to
find increased risk of dementia associated with the use of antidepressants (OR:1.14, 95% CI 0.88-
1.46), antipsychotics (OR:0.97, 95% CI 0.68-1.39) or anticonvulsants (OR: 0.98, 95% CI 0.85-1.13).
High levels of heterogeneity were observed between studies, and subgroup analysis showed that
geographic area had the highest impact on the level of heterogeneity.
Conclusions: In this analysis, we found no association between antidepressants, antipsychotics and
anticonvulsants use with increased risk of dementia. On the other hand, benzodiazepine receptor
agonists showed increased risk of dementia, but this association did not persist after exclusion of
studies with potential reverse causation and confounding by indication. Therefore, future research
with long follow-up periods are needed to confirm if there is a genuine association or if it is
influenced by reverse causation and methodological biases