The use of hypnotics and mortality in a population-based retrospective cohort study

Tzuo-Yun Lan, Ya-Fang Zeng, Gau-Jun Tang. National Yang-Ming University, Taipei, Taiwan

Given the fact that sleep disorders, especially chronic insomnia have become one of the major health problems, the use of hypnotics is steadily increasing. Several population-based studies have been conducted to examine the relationship between hypnotic use and mortality, but no consistent results have been established. Furthermore, no studies with large sample size and long-term observation have been conducted to investigate the relationship between hypnotics, especially specific types of hypnotics, and mortality. The purpose of this study was to better understand the long-term effects of different types of hypnotics on mortality.

We conducted a population-based retrospective cohort study using data from the National Health Insurance Research Database in Taiwan. Information of claims data including basic characteristics (age, sex, level of main medical facility, area of residence, and comorbidities), the use of hypnotics, and vital status between 2000 and 2009 for 1,320,322 individuals was included. In addition to Benzodiazepines and non-Benzodiazepines, Zolpidem, one of the generic drug for non-Benzodiazepines, was used in 70% of research subjects taking hypnotics. The use of hypnotics was therefore divided into four major groups, namely,Zolpidem only, other non-Benzodiazepine only (Zaleplon and Zopiclone), Benzodiazepine only (Brotizolam, Estazolam, Flunitrazepam, Flurazepam, Lormetazepam, Midazolam, Nitrazepam, and Triazolam), and mixed user, and analysed by using the defined daily dose and the cumulative length of use. Hazard ratio (HR) calculated from Cox proportional hazards model (1) with two matching techniques (age and sex matching and propensity score matching (2)) was used to better examine the associations. All analyses were performed using SAS statistical software.

Compared to the non-users, both Benzodiazepines (HR=1.81 & 95% CI= 1.78-1.85 for multiple adjustments including Elixhauser comorbidities) and mixed users (HR=1.44 & 95% CI= 1.42-1.47 for multiple adjustments including Elixhauser comorbidities) had higher risk of death, whereas other non- Benzodiazepines users (HR=1.00 & 95% CI= 0.95-1.05 for multiple adjustments including Elixhauser comorbidities) did not have differences and Zolpidem users (HR=0.73 & 95% CI= 0.71-0.75 for multiple adjustments including Elixhauser comorbidities) exhibited lower risk of mortality. This pattern remained similar after considering two matching techniques. The secondary analyses indicated that Zolpidem users show reduced risk of major cause-specific deaths except cancer and the protective effect was dose-response, with over one-year use having the lowest HR (HR=0.57 & 95% CI= 0.52-0.62 for multiple adjustments including Elixhauser comorbidities).

The effects of different types of hypnotics’ usage on mortality are diverse in the long-term large-sample study of observation based on the representative claims data in Taiwan. The use of Zolpidem has no harmful effect as compared with other hypnotic use. Further investigations are needed to clarify our findings.

(1) Cox DR & Oakes D, Chapman & Hall, 1984.

(2) Rosenbaum PR & Rubin DB, Biometrika 70: 41–55, 1983.