Introduction: Several studies have highlighted an increased risk of dementia in benzodiazepine users. As dementia incidence and benzodiazepine use are both high, particularly in the elderly, even a moderate increase in this risk would induce a tremendous number of cases and have a major public health impact. Areas covered: The aim of this article was to systematically review published observational studies having assessed the relation between benzodiazepine use and dementia, to assess and rank their quality and to provide a balanced opinion about the plausibility of a causal relationship. Expert opinion: Out of the ten studies retrieved, nine reported an increased risk of dementia in benzodiazepine users. The risk increased with cumulative dose and treatment duration and when long-acting molecules were used. Even if the causal nature of this association remains unproved, the reviewed material provides arguments for evoking a causal link. Further research would be necessary to elucidate the mechanism of this effect, if any, to evaluate the risk of exposure in younger population and the influence of risk factors such as depression. In any case, the body of evidence seems sufficient for avoiding prescriptions or renewals that are not fully justified and indiscriminate long-term use.