Sleep, psychological symptoms, and cannabis use before, during, and after COVID-19 "stay-at-home" orders: a structural equation modeling approach.

Abstract

Background: Given the frequent use of cannabis for sleep and mental health symptoms, we aimed to prospectively examine the reciprocal relationship between sleep, mental health, and cannabis use from before, during, and 1 year after the implementation of COVID-19 stay-at-home orders.

Methods: Five hundred thirty-four young adults (21-34 years old) from Oregon and Washington States who previously completed a survey on their sleep and cannabis use prior to stay-at-home orders (T0), were followed up after initial stay-at-home orders were in place for approximately two months (T1), and one year later (T2), to reassess sleep and cannabis use. Sleep measures included the presence of sleep trouble [yes/no], and weekday and weekend sleep duration. The Cannabis Use Disorders Identification Test Revised (CUDIT-R) was used to assess past-six-month cannabis use. The follow up surveys additionally measured mental health symptoms and other health behaviors. We applied cross-lagged panel models to assess the association between cannabis use and sleep at all three time points. Multivariate parallel latent change score models were used to assess associations between changes in cannabis use, sleep, mental health symptoms, and other behavioral factors from T1 to T2.

Results: For the cross-lagged models, reporting sleep trouble at T0 was associated (β = 0.18, p < 0.05) with higher CUDIT-R scores at T1, however this relationship did not hold from T1 to T2. CUDIT-R scores were not associated with sleep trouble from T0 to T1, however CUDIT-R at T1 had a positive association (β = 0.35, p < 0.05) with sleep trouble at T2. The two-wave latent change score model, indicated that change in sleep trouble between T1 and T2 was positively associated with changes in depression (r = 0.37, p < 0.05) and anxiety (r = 0.40, p < 0.05) across this period. No significant correlation was found between changes in cannabis use and changes in sleep trouble.

Conclusions: Our findings suggest an inconsistent bidirectional link between sleep trouble and cannabis use. Only sleep trouble, and not cannabis use, predicted mental health measures. These associations lend support for a holistic approach to designing and implementing public health measures during a global pandemic.