Poor sleep quality was associated with higher rates of mean cannabis use and lower rates of cessation during the first 6 months following a self-guided attempt to quit, although sleep efficiency/duration was not linked to cannabis use outcomes, a study of U.S. veterans has shown.

Citing previous research associating sleep quality with cannabis quit outcomes, including a study that found 48%-77% of cannabis users reported either relapsing or turning to other substances in order to improve their sleep, Kimberly A. Babson, Ph.D., and her colleagues hypothesized that perceived sleep quality rather than actual sleep efficiency/duration would affect cannabis use following a self-guided quit attempt. The results were published recently in Addictive Behaviors (2013;38:2707-13).

Poor sleep quality was associated with higher rates of mean cannabis use and lower rates of cessation during the first 6 months following a self-guided attempt to quit.

Dr. Babson and her colleagues, all affiliated with the VA Palo Alto Health Care System, Menlo Park, Calif., recruited 102 veterans (95% male) with a mean age of 51 years. All participants met DSM-5 criteria for cannabis dependence, had a self-reported desire to quit of 5 or greater on a scale of 0 to 10 (0 = no interest in quitting, 10 = definite interest), and had a self-reported desire to follow a self-guided cessation program. Candidates were excluded if they already had decreased their cannabis use by 25% or more in the previous month, were pregnant or breast-feeding, or had suicidal ideation.

Anxiety disorders were present in 88% of participants, and 43% had a co-occurring mood disorder. Those with and without an anxiety disorder or mood disorder (analyzed separately) did not differ significantly in terms of perceived sleep quality (self-reported overall quality of sleep), sleep efficiency/duration (self-reported quantity of sleep), or cannabis use over the course of the study, the authors reported.

Dependence on other substances was present in 29% of the study group; 4% had a substance abuse disorder. Just more than half of the entire group reported using a sleep medication at baseline, although this group "did not differ from those who did not use a sleep medication at baseline in terms of perceived sleep quality during the study period." Participants using a sleep medication used less cannabis at baseline than did those not using a sleep medication; however, the two groups did not differ when it came to cannabis use over the course of the study.

Participants were rated on the Clinician-Administered PTSD Scale, minus evaluation for two sleep-related symptoms – nightmares and insomnia. The severity of withdrawal from cannabis symptoms also was assessed at baseline and at each follow-up assessment over the 6-month course, using the Marijuana Withdrawal Checklist-Short Form. Post-traumatic stress disorder (PTSD) symptom severity was used as a covariate; symptom withdrawal was considered a time-varying covariate.

The Pittsburgh Sleep Quality Index was used to determine perceived sleep quality in terms of subjective sleep quality, sleep latency, sleep disturbances, and daytime dysfunction. Sleep efficiency/duration was assessed in terms of the combined amount of time spent in bed, whether or not the person was asleep (efficiency), and the actual amount of time spent sleeping (duration).

To determine links between cannabis use, and pre- and post-quit period sleep disturbances, the investigators created two generalized linear mixed models, applying to each a quadratic function to account for a wider array of variables. The first model was not adjusted for covariates, while the second model was adjusted for baseline age, PTSD, and withdrawal symptom severity as measured at each follow-up over the 6-month period.

The researchers did not find a significant intercept between perceived sleep quality and cannabis use. However, they did find a significant linear and quadratic slope showing that while cannabis use declined sharply at first, it leveled off in the context of perceived sleep quality.

"This indicates that over the course of the study (i.e., aggregated across time points), lower perceived sleep quality was associated with higher mean cannabis use," the investigators wrote. "However, perceived sleep quality did not interact with the linear or quadratic slopes, meaning that the association between perceived sleep quality and mean cannabis use could not be tied to a discrete time during the follow-up period." When the investigators adjusted the results for covariates, they found the that results were consistent.

Dr. Babson and her colleagues concluded that their findings indicated one of two possibilities: Either individuals use cannabis to regulate sleep, or chronic use disrupts sleep. Further prospective studies of mechanisms such as emotional regulation would help explain the associations, they wrote. However, their findings, while potentially limited by the study’s self-report nature and the question of whether it is generalizable to a larger population, could serve to help "inform the timing of sleep interventions in cannabis treatments in order to optimize outcomes."

This study was supported by a Veterans Affairs (VA) Clinical Science Research and Development Career Development Award, and by funds from the VA Health Services Research and Development Service. The authors reported no relevant disclosures.

wmcknight@frontlinemedcom.com

Poor sleep quality associated with poor cannabis cessation outcomes