| Summary: | 碩士 === 國立成功大學 === 臨床藥學研究所 === 99 === Background
The prevalence of long-term benzodiazepines (BZD) use is around 2.0-7.4%. Although efficacy of BZD has been well-documented for short-term treatment, the legitimacy of long-term use remains controversial, given the increased risk of traffic accidents, non-accidental falls and cognitive function impairments. By far the most widely reported problems of long-term BZD use, however, lie in the development of tolerance, dependence or even abuse. Hence, the aims of this study were to examine changes in dosages and proportion of high dose users over time among long-term BZD users, and to identify factors associated with escalation to a high dose.
Methods
This was a retrospective cohort study using the original claim data from 2004 to 2007 of one million individuals randomly sampled from the 2005 Registry for Beneficiaries of National Health Insurance (NHI). We identified long-term BZD users who continuously used BZD for at least 720 days from the first prescription in 2005. Then patients aged younger than 18, not starting BZD in therapeutic range, ever hospitalized in 720 days after index date or not enrolled in NHI in 180 days before index date were excluded.
The study subjects were further grouped by (1) user status: new users (received no BZD prescription in 90 days before index date) and continuing users (received at least one); (2) age: aged 18-44, aged 45-64 and aged 65 or order. We converted dosage of all prescribed BZDs to Defined Daily Dose (DDD) and defined high dose as over 1.5 DDD. Finally, conditional logistic regression analyses were carried out to evaluate characteristics associated with escalation to a high dose.
Results
A total of 6,341 subjects were included and most were continuing users (86.0%) and users aged 65 or older (47.9%). Antidepressants and chronic pain accounted for the highest proportion of comedications and comobidities, respectively.
Overall, the median daily dosage remained constant at 0.68 DDD over 720 days of continuous use, and the mean daily dosage increased from 0.70 DDD to 0.85 DDD. The dosage trend was similar between user groups and among age groups, respectively except for the median daily dosage of new users and users aged 18-44 increased slightly in early stage but reach stable state later. The median daily dosage of continuing users was higher than new users, and young users were higher than old users. The proportion of high dose users significantly increased over time (p＜ 0.0001). Patient characteristics significantly associated with escalation to a high dose included displaying comorbid chronic pain, insomnia and anxiety. Prescribing characteristics including having any prescription which was indicated high dose and containing two or more BZD, mainly using hypnotic BZD, short-acting BZD, and high unit dose formulations of BZD (＞1 DDD) also significantly increased the risk.
Conclusions
Overall, dosage among most long-term BZD users remained constant and well within the therapeutic range. The rise of the mean daily dose reflected there may be some extreme high dose users with increment in number or dosage over time.
Factors significantly increased the risk of escalation to a high dose included those suffering from comorbid chronic pain, insomnia and anxiety, having any prescription which was indicated high dose and containing two or more BZD, mainly using hypnotic BZD, short-acting BZD, and high unit dose formulations of BZD (＞1 DDD).
Keywords
Benzodiazepines, BZD, z drugs, sedatives, hypnotics, long-term use, dosage, dose escalation
|
|---|
