Long-term use of benzodiazepines: Definitions, prevalence and usage patterns - a systematic review of register-based studies

Background: Numerous treatment guidelines recommend that long-term use of benzodiazepines (BZD) should be avoided primarily due to development of tolerance and a risk for BZD dependence. Despite this, long-term BZD use remains a controversial subject in clinical patient care with "for and against" debates. However, there is no explicit understanding of what is meant by long-term BZD use in real world. The aim of this study was to assess different definitions, usage patterns, prevalence and other characteristics of long-term BZD use based on published register-based studies. Synthesis of these characteristics is essential to derive a meaningful definition of long-term BZD.

Methods: Systematic review of register-based studies on long-term BZD use published in 1994-2014.

Results: Fourty-one studies met our predetermined inclusion criteria. The length of BZD use defined as "long-term" varied in these studies ranging from one month to several years. The most common definition was six months or longer during a year. The prevalence of long-term BZD use in the general population was estimated to be about 3%. The relative proportion of long-term BZD users (all definitions) in adult BZD users ranged from 6% to 76% (mean 24%; 95% CL 13-36%). The estimates were higher in studies only on the elderly (47%; 95% CL 31-64%). Long-term use involved typically steady treatment with low BZD doses. However, in elderly patients long-term BZD use and exceeding recommended doses was relatively common. Several characteristics associated with long-term use were found.

Conclusions: Long-term BZD use is common and a clinical reality. Uniform definitions for "long-term", which is in line with population-based evidence, is needed to have more comparable results between studies. Our systematic review suggests that duration of BZD treatment over six months, the most common definition for long-term BZD use in the included studies. As also recommended previously, it is a useful starting point for further analyses on disadvantages but also potential advantages associated with long-term BZD use.