Middle-aged women prescribed medication for sleep disturbances and insomnia did not report significant long-term improvements compared with unmedicated women with similar issues, according to a new cohort study. "The medications may be very helpful when used short-term or intermittently, but long-term use is not associated with reductions in sleep disturbances," said lead author Dr. Daniel Solomon of Harvard Medical School, in Boston, who specializes in pharmacoepidemiology. The medications considered included benzodiazepines (BZD), like estazolam and lorazepam, as well as selective BZD receptor agonists, like eszopiclone and zolpidem, and drugs with other mechanisms, like doxepin and trazodone. Dr. Solomon and his collaborators found no differences in either sleep quality or duration between participants over a period of one to two years, with both groups reporting similar levels of disturbed sleep - defined by the study as waking too early, difficulty falling asleep, and/or difficulty staying asleep. "Current randomized controlled trial evidence supports the benefits of these medications in the short term (i.e. 1-6 months)," Dr. Solomon told Reuters Health by email, "but 30-40% of adults who start them continue to use them long-term." Published in BMJ Open, the study examined 238 women starting sleep medications and 447 propensity-score-matched non-users also experiencing disturbed sleep. The mean age in both groups was 50 years and the mean BMI, 29 kg/m2. Over half were white (58%), while the remainder was more ethnically diverse with 23% African-American, 4% Hispanic, 7% Chinese and 9% Japanese women. About half the women were either current or past tobacco users, and close to one in five also reported moderate to heavy alcohol use, defined as having one to seven drinks-per-week; 11% to 12% had more than seven drinks-per-week. The two groups were otherwise similar in terms of comorbidities. There were no significant between-group differences in initial sleep-disturbance ratings on a five-point Likert scale ranging from no difficulty on any night (1) to difficulty on five or more nights a week (5). Women starting a medication had a mean score of 2.7 for "difficulty initiating sleep", of 3.8 for "waking frequently" and of 2.8 for "early-morning awakening". Those not taking any sleep medication had scores of 2.6, 3.7 and 2.7, respectively. After a year of treatment, medication users reported a mean of 2.6 for initiating sleep, 3.6 for waking frequently and 2.8 for early-morning awakening. The corresponding rates among non-users were 2.3, 3.5 and 2.5. The two-year findings were similar, with no statistically significant differences between the groups. Dr. Michael Sateia, a psychiatrist and sleep-medicine specialist at Dartmouth's Geisel School of Medicine, in Hanover, New Hampshire, cautioned that the study's methodology left open questions about the specific sleep medications used. He noted that some commonly prescribed drugs, like trazodone, have already been shown to be ineffective for chronic insomnia. "Patients with insomnia are also notoriously inaccurate in reporting the severity of sleep disturbance, especially on a retrospective basis," Dr. Sateia, who was not involved in the study, told Reuters Health by email. "The retrospective reports (two-week retrospective) is not the most reliable methodology for accurately measuring sleep disturbance." "Readers of this report should also be aware that long-term (6-24 months) studies of both eszopiclone and zolpidem, in controlled conditions, have demonstrated continued efficacy of the drugs over time," he added, "without dosage escalation." Despite those caveats, Dr. Sateia said that the findings were consistent with clinical experience, adding that long-term medication has typically not been favored over treating chronic insomnia with cognitive-behavioral therapy, the recommended clinical practice. —Matthew Phelan Source
