a new study shows.
“While there are good data from [randomized, controlled trials] that these medications improve sleep disturbances in the short term,” few studies have examined whether they provide long-term benefits, stated the authors of the paper, which was published in BMJ Open.
“The current observational study does not support use of sleep medications over the long term, as there were no self-reported differences at 1 or 2 years of follow-up comparing sleep medication users with nonusers,” author Daniel H. Solomon, MD, MPH, from Brigham and Women’s Hospital, Boston, and colleagues wrote.
Women included in the analysis were drawn from the Study of Women’s Health Across the Nation (SWAN), an ongoing multicenter, longitudinal study examining women during the menopausal transition. The average age of the women included in the cohort was 49.5 years and approximately half were White. All women reported a sleep disturbance on at least 3 nights per week during a 2-week interval. At follow up, women were asked to use a Likert scale to rate three aspects of sleep: difficulty initiating sleep, frequent awakening, and waking up early. On the scale, 1 represented having no difficulties on any nights, 3 represented having difficulties 1-2 nights per week, and 5 represented having difficulty 5-7 nights per week.
Women already using prescription sleep medication at their baseline visit were excluded from the study. Medications used included benzodiazepines, selective BZD receptor agonists, and other hypnotics.
Over the 21 years of follow-up in the SWAN study (1995-2016), Dr. Solomon and colleagues identified 238 women using sleep medication and these were compared with a cohort of 447 propensity score–matched non–sleep medication uses. Overall, the 685 women included were similar in characteristics to each other as well as to the other potentially eligible women not included in the analysis.
Sleep disturbance patterns compared
At baseline, sleep disturbance patterns were similar between the two groups. Among medication users, the mean score for difficulty initiating sleep was 2.7 (95% confidence interval, 2.5-2.9), waking frequently 3.8 (95% CI, 3.6-3.9), and waking early 2.9 (95% CI, 2.7-3.1). Among the nonusers, the baseline scores were 2.6 (95% CI, 2.5-2.7), 3.7 (95% CI, 3.6-3.8), and 2.7 (95% CI, 2.5-2.8), respectively. After 1 year, there was no statistically significant difference in scores between the two groups. The average ratings for medication users were 2.6 (95% CI, 2.3-2.8) for difficulty initiating sleep, 3.8 (95% CI, 3.6-4.0) for waking frequently, and 2.8 (95% CI, 2.6-3.0) for waking early.
Average ratings among nonusers were 2.3 (95% CI, 2.2-2.4), 3.5 (95% CI, 3.3-3.6), and 2.5 (95% CI, 2.3-2.6), respectively.
After 2 years, there were still no statistically significant reductions in sleep disturbances among those taking prescription sleep medications, compared with those not taking medication.
The researchers noted that approximately half of the women in this cohort were current or past tobacco users and that 20% were moderate to heavy alcohol users.
More work-up, not more medication, needed
The study authors acknowledged the limitations of an observational study and noted that, since participants only reported medication use and sleep disturbances at annual visits, they did not know whether patients’ medication use was intermittent or of any interim outcomes. Additionally, the authors pointed out that those classified as “nonusers” may have been using over-the-counter medication.
“Investigations should look at detailed-use patterns, on a daily or weekly basis, with frequent outcomes data,” Dr. Solomon said in an interview. “While our data shed new light on chronic use, we only had data collected on an annual basis; daily or weekly data would provide more granular information.”
Regarding clinician prescribing practices, Dr. Solomon said, “short-term, intermittent use can be helpful, but use these agents sparingly” and “educate patients that chronic regular use of medications for sleep is not associated with improvement in sleep disturbances.”
Commenting on the study, Andrea Matsumura, MD, a sleep specialist at the Oregon Clinic in Portland, echoed this sentiment: “When someone says they are having trouble sleeping this is the tip of the iceberg and it warrants an evaluation to determine if someone has a breathing disorder, a circadian disorder, a life situation, or a type of insomnia that is driving the sleeplessness.”
“I think this study supports what we all should know,” Dr. Matsumura concluded. “Sleep aids are not meant to be used long term” and should not be used for longer than 2 weeks without further work-up.
Funding for this study was provided through a grant from the National Institutes of Health. Dr. Solomon has received salary support from research grants to Brigham and Women’s Hospital for unrelated work from AbbVie, Amgen, Corrona, Genentech and Pfizer. The other authors and Dr. Matsumura have reported no relevant financial relationships.