Effect of benzodiazepine discontinuation on dementia risk.

OBJECTIVES: This study aimed to examine whether benzodiazepine (BZD) discontinuation would decrease the risk of dementia. DESIGN: A population-based nested case-control study of dementia was used. SETTING: All subjects aged 45 years or older and enrolled in the National Health Insurance Research Database in Taiwan between 1997 and 2007 were randomly selected. PARTICIPANTS: A total of 8,434 cases had been identified with dementia at least three times in ambulatory claims or with one record in inpatient claims. They were individually matched with two comparison subjects (N = 16,706) by age, gender, and index date. MEASUREMENTS: The lengths of discontinuation, cumulative BZD dose, and potential confounding factors, including medical and psychiatric disorders, were measured and used for further analysis. RESULTS: Compared with nonusers, current users had an increased risk of dementia (adjusted odds ratio [aOR] = 2.71; 95% confidence interval [CI], 2.46-2.99). The dementia risk for former users was reduced as the duration of discontinuation lengthened (<1 month aOR = 2.40, 95% CI, 1.98-2.92; 1-3 months aOR = 1.93, 95% CI, 1.67-2.23; 3-6 months aOR = 1.49, 95% CI, 1.28-1.74; 6-12 months aOR = 1.43, 95% CI, 1.25-1.64; 1-2 years aOR = 1.23, 95% CI, 1.09-1.40; 2-3 years aOR = 1.22, 95% CI, 1.06-1.40; and >3 years aOR = 1.08, 95% CI, 0.98-1.20). The decreasing trend was significant (p < 0.001). CONCLUSION: The risk of dementia was high for current users and decreased as the duration of BZD discontinuation lengthened. Further investigations are needed to replicate this association and explore the underlying mechanism that links long-term BZD use, BZD discontinuation, and the pathogenesis of neurocognitive dysfunction.