Purpose: Benzodiazepines (BZDs) and related drugs (BZRDs) are commonly used to treat diverse psychiatric disorders due to their anxiolytic, hypnotic and sedative properties, despite their known associated side effects, including acute consequences on cognition. Recently, some studies have also suggested that long-term cognitive effects may coexist, as the increased risk of cognitive decline and dementia. This review aims to appraise and summarise published synthesis studies on the risk of dementia development due to BZDs/BZRDs use.

Methods: A comprehensive systematic search was carried out in PubMed, Web of Science, Cochrane Library and Epistemonikos databases. Grey literature and hand search of the studies' reference lists were undertaken. Meta-analysis, systematic and non-systematic reviews were included. Neither language nor date restrictions were applied. Search results other than synthesis studies were excluded. The methodological quality of the included reviews was analysed with AMSTAR-2 and SANRA tools.

Results: Overall, 877 records were initially retrieved and 15 complied with the inclusion criteria. From these, five were systematic reviews with meta-analysis, two were systematic reviews and eight were non-systematic reviews. Most of the primary studies included in the analysed reviews found an association between BZDs/BZRDs use and subsequent dementia, with meta-analysis studies reporting an increased risk for users (ORs ranging from 1.38 to 1.78). However, the considerable clinical and methodological heterogeneity of the primary studies makes it difficult to establish a causal relationship.

Conclusion: Although hampered by the heterogeneity between the studies, the present findings suggest an association between BZDs/BZRDs use and increased risk of dementia in older adults.