Abstract
Objective:
Marijuana use among pregnant women is on the rise in part due to the perception that marijuana may improve problems related to pregnancy such as poor sleep. This study’s objective was to examine associations between marijuana use and sleep quality among a sample of women during pregnancy.
Material and methods:
The sample included women seeking prenatal care at The Ohio State University Wexner Medical Center (2010–2015). Intake assessments included medical, demographic, and socioeconomic domains, as well as the Pittsburgh Sleep Quality Index. Marijuana use during pregnancy was determined using urine screens, chart abstraction, and self-report. Women completed standardized questionnaires regarding sleep quality, depressive symptoms, anxiety, stress, and discrimination at enrollment and each subsequent trimester. A linear mixed-effect model was used to assess the relationship between sleep variables and marijuana use adjusted for maternal race, education, household income, age, marital status, depressive symptoms, anxiety, stress, discrimination, and use of tobacco and other substances during pregnancy. Women completed the sleep quality assessments for a total of 294 pregnancies, which comprise the study population.
Results:
Among the study sample (n = 294), 93 women used marijuana and 201 women did not use marijuana during their pregnancies. Women who used marijuana (n = 93) were more likely to identify as African-American (73% vs 58%; p = .01), report government health insurance (98% vs 89%; p = .001), use tobacco during pregnancy (66% vs 33%; p < .001), report less household income (70% vs 43% < 10,000 annual household income; p < .001), and be unmarried (69% vs 49%; p < .001) compared to women who did not. Mean sleep quality was similar among women who did (μ = 7.6; SD = 4.0) and did not use marijuana during pregnancy (μ = 7.7; SD = 4.0), and both groups had a mean score worse than the conventional cutoff for poor sleep quality (>5). In fact, both groups reported worse sleep than is typically observed among cohorts reporting poor sleep, which have ranged from 5.3 to 6.3.
Conclusions:
Current findings did not suggest differences in sleep quality between women who used and did not use marijuana during pregnancy. Findings are contrary to the perception that marijuana use alleviates sleep-related problems during pregnancy. Given well-documented adverse outcomes associated with prenatal marijuana exposure for children and the increase in women using marijuana during pregnancy, providers should be prepared to discuss possible harms associated with marijuana use during pregnancy as well as provide psychoeducational information and service referrals to those interested. Future studies could improve upon this design by assessing objective measures of sleep, such as actigraphy, as well as marijuana use repeatedly throughout pregnancy, which may be a more optimal strategy for illuminating potential relationships between marijuana use and sleep during pregnancy.
Keywords: Sleep quality, marijuana use, cannabis use, pregnancy, obstetrics
Introduction
Pregnant women commonly report poor sleep quality during pregnancy. Prior findings indicate pregnancy is strongly related to a variety of sleep-related problems including insomnia, sleep disturbances, and poor sleep quality [1]. These sleep-related issues are problematic given their association with poor outcomes for women, such as poor mental health, gestational diabetes, and hypertensive disorders [2,3]; as well as negative short- and long-term outcomes for their offspring including: pre-term birth, reduced gestational length, and increased behavioral problems during childhood [4–6]. Due to the associated consequences of sleep-related problems during pregnancy, many pregnant women seek safe strategies to improve their sleep.
The prevalence of marijuana use among pregnant women is increasing [7]. Shifts in legislation, mixed research findings, and softening social attitudes related to consequences and benefits as well as mixed messaging from healthcare providers regarding the safety of marijuana use while pregnant [8,9], may influence pregnant women’s decision-making regarding marijuana use. Similar to the general population, research suggests pregnant women may use marijuana because of the perception that it is a low-risk way to improve pregnancy-related symptoms, such as poor sleep [10,11]. This is problematic as marijuana use during pregnancy is associated with adverse child and maternal health outcomes [12]. Among non-pregnant populations, marijuana use is associated with decreased sleep onset latency and increased slow wave sleep; however, use also has a negative impact on wake onset latency, REM sleep, and overall quality of sleep [13]. Our understanding is limited and additional research is needed regarding potential hazards, motivations, and contributing factors of marijuana use while pregnant as well as how marijuana use and sleep intersect among this important population. Marijuana use is strongly associated with poor mental health [14]. Given the link between marijuana use and mental health as well as well-documented associations between poor sleep and poor mental health, mental health constitutes a critical factor to account for in the study of marijuana use and sleep problems. To the best of our knowledge, no research to date has examined the association between marijuana use during pregnancy and sleep quality.
Objectives
This study represents an initial exploratory step toward addressing this gap by evaluating the association between marijuana use and maternal sleep during pregnancy. Specifically, this study reports on associations between marijuana use and several dimensions of sleep quality (including, duration, disturbance, latency, daytime dysfunction, efficiency, medication use, and overall quality), while controlling for several mental health factors, among a sample of pregnant women. We hypothesized that pregnant women who used marijuana would report poorer sleep quality than those who did not use marijuana.
Materials and methods
Study population
This study included women who were enrolled during pregnancy in the Ohio Perinatal Research Network Perinatal Research Repository (PRR, Columbus, Ohio, USA), which has been described previously [15]. The PRR recruited women from 2010 to 2015 who were attending several antenatal clinics at The Ohio State University Wexner Medical Center (OSUWMC). Women were eligible if they were 16–50 years of age, could communicate in English, and intended to deliver at OSUWMC. Women were enrolled as early as possible in pregnancy. Women enrolled in the PRR provided written informed consent, which included an option to allow re-contact for participation in future IRB-approved research, for which approximately 75% of PRR enrolled women agreed. At enrollment, women completed an intake questionnaire covering a variety of medical, demographic, and socioeconomic domains, including items on household income and use of tobacco, marijuana, and substances associated with maintenance of opioid use disorders (such as methadone, buprenorphine, and suboxone) during the pregnancy. At enrollment and once each subsequent trimester, women also completed standard questionnaires regarding sleep quality [16], depressive symptoms [17], trait anxiety [18], perceived stress [19], and experienced discrimination [20]. At these same time points, women provided urine and blood samples for unspecified future use, which were stored at −80°C.At the conclusion of the pregnancy, the obstetric record was abstracted by trained research personnel onto a pre-coded form with closed-ended items regarding clinically recognized substance use, including marijuana, during pregnancy.
As part of a subsequent study, the Lifestyle and Early Achievement in Family study [LEAF; 21], the archived urine samples were assayed for all PRR participants who agreed to be contacted for future research and whose children were age-eligible to participate. The current study used data, including urine assay results, from these participants, which comprise the current sample.
The PRR and LEAF studies were approved by the Institutional Reviewer Board (IRB) at Nationwide Children’s Hospital and a reliance agreement was executed with the IRB at OSUWMC.
Exposure
Urine samples were assayed for 11-nor-carboxy-Δ9-tetrahydrocannabinol (Δ9-THC-COOH), the primary pharmacologically active component of marijuana, by gas chromatography-tandem mass spectrometry following hydrolysis [15,21]. The samples were also analyzed for several other substances including substances commonly associated with medically-assisted treatment for opioid use disorders (methadone, buprenorphine, suboxone). Women were considered to have used marijuana if they self-disclosed marijuana use on their intake questionnaire, if use was noted on the obstetrical record abstraction, or if any urine specimen had a Δ9-THC-COOH concentration of >15 ng/ml, the concentration considered to represent active use when employing mass spectrometry [22].
Outcome
Sleep quality was evaluated with the Pittsburgh Sleep Quality Inventory (PSQI), a 19-item, self-rated questionnaire designed to measure sleep quality and disturbances over the past month [14], which has been validated in pregnant women [23]. The 19 items are combined into 7 clinically-derived component scores including 1) sleep duration, 2) sleep disturbance, 3) sleep latency, 4) daytime dysfunction due to sleepiness, 5) sleep efficiency, 6) sleep quality, and 7) sleep medication use. Each sleep component yielded an ordinal score ranging from 0 to 3, with 3 indicating the greatest dysfunction. Sleep component scores were summed to yield a total global score with the higher total score indicating worse sleep quality; a global score >5 distinguishes good from poor sleepers. We also created a priori a binary definition of poor sleep quality for each component, which included moderate sleep disturbances (a scale score of 2) or severe sleep disturbances (a sleep score of 3) [16].
Covariates
The Perceived Stress Scale (PSS) is a psychometrically strong measure of perceived stress [19] and was administered to measure maternal perceived stress. The Center for Epidemiological Studies-Depression (CES-D), a reliable and valid tool with diverse populations, assessed maternal depression [17]. The Spielberger State-Trait Anxiety Inventory (STAI) assessed individual differences in anxiety as a situation-specific personality trait, which has demonstrated psychometric strengths among many populations including pregnant women [18,24]. Lastly, The Everyday Discrimination Scale (EDS), a 9-item measure of subjective experiences of discrimination was administered [19]. All the above covariate measures were administered at enrollment and each subsequent trimester. Demographic factors from the intake including: maternal race, education, age, household income, marital status, and use of tobacco and other substances during pregnancy (including methadone, buprenorphine, and suboxone) were also included as covariates. The identification of use of other substances (methadone, buprenorphine, suboxone) was based on self-report from maternal intake questionnaire as well as urine screens conducted throughout the pregnancy, a positive on either resulted in women’s classification as having used that substance. The inclusion of this particular set of confounders was based on our a priori hypothesized associations we would expect to observe between each confounder and marijuana use and sleep.
Statistical analysis
Categorical variables are reported as numbers and percentages, and differences in participant characteristics between women who did and did not use marijuana were compared with the chi-square test. Continuous variables (maternal age, PSQI score) are summarized as means and standard deviations (SD), and tested by the two sample t-test. All women completed assessment questionnaires at least once; however, not all women completed these questionnaires in all trimesters because they enrolled late, they delivered early, or they lacked time during the clinic visit. Thus, the mean of the values from the available questionnaires from each woman was calculated and adjusted in the analysis. A linear mixed-effect model with repeated measures (for multiple trimesters in a pregnancy) and clustering (for more than 1 pregnancy to a woman) was applied to examine the associations between the numeric PSQI and its 7 sub-components and marijuana use [25]. Interaction effects between marijuana use and trimester were examined as well. p values <.05 were considered statistically significant. Adjusted models included CESD, PSS, and STAI scores (all included in the model as the mean value of each trimester scores); maternal race; household income; education; age; marital status; and use of tobacco and others substances (methadone, buprenorphine, suboxone) during pregnancy.
Among women in the current sample, less than 7% were missing data at random for at least one con-founding variable. Multiple imputation by fully conditional specification (FCS MI) was performed by SAS PROC MI to impute missing values for both continuous and categorical variables with arbitrary missing patterns, which allows convergence on a prediction even when their exact distribution is unknown [26]. FCS regression, logistic regression, and discriminant function were used to impute continuous, binary, and nominal variables, separately. Twenty-five imputed datasets were created, and results were combined using SAS PROC MIANALYZE [27]. Only missing covariates were imputed; no outcome variables were imputed. All analyses were performed using SAS version 9.4 [28].
Results
Of the 497 pregnancies enrolled in the PRR that were age eligible for LEAF, 389 agreed to be contacted for future research; 1 mother requested her data be purged from the repository and 11 pregnancies involved twins, which yielded 11 duplicate pregnancies. The PSQI was added to the repository protocol after the first year; therefore, the first 83 women did not have sleep data. As a result, of the remaining 377 pregnancies in the PRR, 294 pregnancies included at least one completed maternal sleep assessment questionnaire (details below), which represents the study sample. Among the study sample, 93 (31.6%) women used marijuana during pregnancy whereas the remaining 201 pregnancies (68.4%) did not include any indication of marijuana use during pregnancy across all assessment measures.
Among the 294 pregnancies in the current sample, 62.9% identified as African-American, 33% identified as White, and 4.1% identified as an alternative racial identity by selecting “Other.” On average women were 26.3 years of age (SD = 5.2). A majority of participating women were not married (75.2%) and had a high school diploma or GED (79.3%). Of the 93 pregnancies for which marijuana was used, 30 pregnancies (32.3%) were identified using maternal report; however, among pregnancies for which marijuana use was self-reported, most (90.0%) also reflected marijuana use in the record abstraction or urine screen. In contrast, marijuana was identified based solely on urine screen in 36 pregnancies. Compared to women who did not use marijuana, those who did were more likely to identify as African-American (p = .01), to have government health insurance (p = .01), to report less household income (p < .01), to use tobacco during pregnancy (p < .01), and to be unmarried (p < .01). They did not differ in ethnicity, education, parity, singleton pregnancy, or having experienced homelessness or physical abuse during the previous 12 months. Of note, the mean total PSQI score was similar among women who did (μ = 7.6; SD = 4.0) and did not use marijuana during pregnancy (μ = 7.7; SD = 4.0), and both groups had a mean score worse than the conventional cutoff for poor sleep quality of 5 [16]. For additional details regarding sample characteristics across maternal marijuana use groups see Table 1.
Table 1.
Characteristics of the study population, 294 pregnancies.
| N of Pregnancies (%) / Mean (SD) | ||
|---|---|---|
| Characteristics | Marijuana non-users (n = 201) | Marijuana users (n = 93) |
| Maternal age (Mean, SD)a | 26.52 (5.22) | 25.85 (5.06) |
| Race | ||
| Black/African American | 117 (58.2) | 68 (73.1) |
| White/Caucasian or Other | 84 (41.8) | 25 (26.9) |
| Ethnicity | ||
| Hispanic | 8 (3.9) | 3 (3.3) |
| Non-Hispanic | 193 (96.1) | 90 (96.7) |
| Health insurance | ||
| Government-assisted insurance | 179 (89.1) | 91 (97.9) |
| Private insurance | 16 (7.9) | 4 (2.2) |
| Uninsured or self-pay | 6 (3.0) | 0 (0.00) |
| Annual household income | ||
| Less than $10,000 (less than $833 per month) | 87 (43.1) | 65 (70.3) |
| $10,000–$29,999 ($833–$2,499 per month) | 83 (41.3) | 22 (23.4) |
| $30,000+ ($2,500+ per month) | 31 (15.6) | 6 (6.3) |
| Tobacco use during pregnancy | ||
| Yes | 67 (33.3) | 61 (65.6) |
| No | 134 (66.7) | 32 (34.4) |
| Education | ||
| Less than high school | 44 (21.9) | 21 (22.7) |
| High school/GED | 70 (35.0) | 36 (38.3) |
| Some college/Associate’s degree | 64 (31.9) | 32 (34.6) |
| Bachelor’s degree or higher | 22 (11.1) | 4 (4.4) |
| Homeless | ||
| Yes | 21 (10.6) | 17 (18.5) |
| No | 180 (89.4) | 76 (81.6) |
| Parity | ||
| Primiparous | 25 (12.4) | 12 (12.9) |
| Multiparous | 176 (87.6) | 81 (87.1) |
| Physical abuse in past year | ||
| Yes | 20 (10.1) | 13 (13.7) |
| No | 181 (89.9) | 80 (86.3) |
| Alcohol use in current pregnancy | ||
| Yes | 39 (19.6) | 26 (28.1) |
| No | 162 (80.4) | 67 (71.9) |
| Marital status | ||
| Never | 99 (49.4) | 64 (68.9) |
| Married | 51 (25.5) | 6 (6.7) |
| Living as married | 25 (12.4) | 16 (17.7) |
| Formerly married | 26 (12.7) | 6 (6.7) |
| Multiple birth | ||
| Yes | 3 (1.5) | 3 (3.2) |
| No | 198 (98.5) | 90 (96.8) |
| Gestational Length | ||
| Preterm (<37 weeks) | 68 (33.7) | 37 (39.8) |
| Term (37+ weeks) | 133 (66.3) | 56 (60.2) |
| Methadone/buprenorphine/suboxone use | ||
| Yes | 15 (7.5) | 4 (4.3) |
| No | 186 (92.5) | 89 (95.7) |
| PSQI Score (Mean, SD)a | 7.60 (4.0)b | 7.69 (4.0)b |
Continuous-variable (Maternal age, PSQI score) shown as mean and standard deviation (SD).
Missing: 39 PSQI Score missing in non-marijuana users, and 22 missing in marijuana users.
The associations between marijuana use and the total PSQI score as well the component scores are presented separately for each trimester, and overall for the entire pregnancy in Table 2, as are the adjusted scores. The number of women who completed at least one subscale on the PSQI at each trimester is also shown. Although global PSQI score increased significantly among women not using marijuana (p = 0.04), in the group as a whole, there was no statistical trend for the global PSQI to change across trimester and the interaction term between marijuana use and trimester was not significant. This implies that the pattern of global PSQI scores across trimesters was not statistically different between women who did and did not use marijuana. There was no consistent pattern of differences between women who did and did not use marijuana in either the global score or the sub-scores across trimesters. For example, women who used marijuana had global scores 1.3 points higher in the first trimester, 0.5 points lower in the second trimester, and nearly identical in the third trimester. Overall, women who used marijuana had mean scores that were approximately 1 point higher than those who did not in both adjusted and unadjusted analysis; none of the differences approached statistical significance. The inclusion of maternal mental health factors (depressive symptoms, anxiety, stress), and other maternal characteristics (maternal race, education, age, marital status, household income, and use of tobacco and other substances during pregnancy) did not appear to influence the relationships between marijuana use and sleep quality nor did the inclusion of these factors significantly improve the models. Poor sleep (global score >5) was extremely common regardless of marijuana use—70.6% and 59.5% of women who did and did not use marijuana, respectively, reported poor sleep in the first trimester; the corresponding percentages for the third trimester were 71.7% and 58.5%.
Table 2.
Associations between Pittsburgh Sleep Quality Index (PSQI) and marijuana use, mixed model (visits = 520).
| PSQI score | 1st trimester mean (SD)a | 2nd trimester mean (SD)a | 3rd trimester mean (SD)a | Unadjusted β (95% Cl) |
Adjustedb β (95% Cl) |
|||
|---|---|---|---|---|---|---|---|---|
| Marijuana Non-Users | Marijuana Users | Marijuana Non-Users | Marijuana Users | Marijuana Non-Users | Marijuana Users | |||
| PSQI global score | 6.62 (2.85) n = 37 | 7.94 (3.55) n = 34 | 7.39 (4.10) n = 119 | 6.82 (3.56) n = 51 | 8.03 (4.19) n = 127 | 8.05 (4.03) n = 60 | 0.91 (−1.03, 2.85) | 1.10 (−0.72, 2.92) |
| Sleep duration | 0.46 (0.64) n = 41 | 0.94 (1.11) n = 35 | 0.89 (1.10) n = 129 | 0.61 (1.04) n = 56 | 0.83 (1.09) n = 132 | 0.91 (1.18) n = 65 | 0.25 (−0.27, 0.77) | 0.29 (−0.25, 0.82) |
| Sleep disturbances | 1.85 (0.85) n = 62 | 2.07 (0.81) n = 45 | 1.64 (0.71) n = 136 | 1.63 (0.68) n = 64 | 1.69 (0.68) n = 139 | 1.85 (0.69) n = 71 | 0.12 (−0.28, 0.52) | 0.07 (−0.33, 0.46) |
| Sleep latency | 1.31 (1.00) n = 42 | 1.50 (1.08) n = 36 | 1.51 (1.02) n = 127 | 1.19 (0.95) n = 54 | 1.39 (1.02) n = 132 | 1.39 (0.96) n = 62 | −0.22 (−0.74, 0.29) | −0.19 (−0.68, 0.30) |
| Daytime dysfunction | 0.88 (0.79) n = 43 | 1.09 (0.74) n = 35 | 0.93 (0.89) n = 126 | 0.83 (0.76) n = 60 | 0.99 (0.86) n = 136 | 0.91 (0.84) n = 69 | 0.18 (−0.26, 0.61) | 0.21 (−0.20, 0.61) |
| Habitual sleep efficiency | 0.95 (1.25) n = 43 | 0.92 (1.23) n = 36 | 0.98 (1.22) n = 128 | 0.98 (1.19) n = 57 | 1.21 (1.24) n = 134 | 1.21 (1.23) n = 66 | −0.05 (−0.78, 0.68) | 0.04 (−0.69, 0.76) |
| sleep quality | 1.18 (0.72) n = 44 | 1.26 (0.82) n = 35 | 1.28 (0.88) n = 132 | 1.16 (0.68) n = 62 | 1.37 (0.81) n = 138 | 1.46 (0.91) n = 70 | −0.03 (−0.46, 0.40) | −0.06 (−0.46, 0.35) |
| Use of sleeping medication | 0.14 (0.46) n = 44 | 0.29 (0.71) n = 35 | 0.28 (0.73) n = 131 | 0.18 (0.67) n = 62 | 0.54 (1.07) n = 138 | 0.36 (0.82) n = 69 | 0.21 (−0.14, 0.55) | 0.19 (−0.16, 0.54) |
Note:
SD: Standard Deviation.
Adjusted models: Adjusted for CESD, EDS, PSS, STAI (mean of three trimesters); race, education, tobacco use during pregnancy, age, household income, use of other drugs (methadone, buprenorphine, suboxone), and marital status.
Discussion
This prospective cohort study found that women who did and did not use marijuana during pregnancy did not report statistically significantly different sleep quality, as determined by a standard self-report instrument. We also found that the global PSQI score did not increase significantly as pregnancy progressed, which is contrary to past findings. However, we also noted that by the conventional definition, poor sleep was the rule rather than the exception within both groups of women and sleep quality was worse than in previously studied samples. The models were not improved by the inclusion of maternal mental health factors, use of tobacco and other substances during pregnancy, and maternal demographic characteristics (age, race, education, and household income). These factors did not appear to influence the relationship between marijuana use and sleep-related issues among our sample of pregnant women.
Current findings spotlight limitations of studying marijuana use during pregnancy strictly utilizing maternal report data. In line with previous research, the current study would have misclassified a large portion of participant’s marijuana use had we relied only on maternal report [29]. Specifically, 64 women (68.8%) of the 93 women who used marijuana would have been misclassified as non-users in the current sample, which would have likely biased results.
We are unaware of previous reports that compared sleep quality during pregnancy among women who used and did not use marijuana. Prior research reflects pregnant women commonly report difficulties sleeping that increase over the course of pregnancy [30,31] and report use of marijuana during pregnancy to address pregnancy-related, such as sleep-related issues [9–11]. Facco, et al. [32] followed women from the first to the third trimester of pregnancy and noted a decrease in total sleep time, and increases in frequent snoring and meeting diagnostic criteria for restless leg syndrome from the first to the third trimester. In that study the mean PSQI score increased from 5.4 to 6.3, and the fraction of women reporting poor sleep, increased from 39.0% to 53.5% from the first to the third trimester, respectively; both increases were highly statistically significant. Facco, et al. [32] studied nulliparous women without preexisting medical conditions, which may account for the somewhat lower occurrence of poor quality sleep in that population. Oken, et al. [33] reported that pregnant women exhibited a mean PSQI score of 5.34 from 10–20 weeks’ gestation, and that poor sleep was more common among women of lower socioeconomic status than among other women. In comparison, the mean PSQI score within the current sample was higher at 7.63, suggesting our sample experienced worse sleep in general.
Our study had numerous strengths. Marijuana use was assessed by 3 modes: self-report, clinical recognition, and urine toxicology. We employed standardized, well-recognized instruments to assess sleep, as well as depression, anxiety, stress, and discrimination. The women were enrolled prospectively into a general cohort, not specifically focused on marijuana or sleep, thereby minimizing the incentive for women who used marijuana and/or had sleep difficulties to enroll preferentially. We utilized state of the art data analytic methods that could incorporate missing data.
However, there were limitations. Most notably, the PSQI is a self-reported instrument; we did not utilize objective measures of sleep such as actigraphy or polysomnography. It was designed to distinguish “good” from “poor” sleepers, but not necessarily to make a clinical diagnosis [16]. The PSQI is well-accepted in both pregnant and non-pregnant women; however, women who use marijuana might perceive their sleep differently than women who do not, even in the absence of objectively measured differences. Secondly, our exposure assessment did not allow determination of the dose and timing of marijuana use. Therefore, this study was unable to differentiate occasional use from habitual use of marijuana. Further, our study design would not readily have distinguished whether women who slept poorly used marijuana as a form of self-medication versus marijuana use being a cause of sleep disruption; however, our null results mitigate this concern somewhat. It is also important to note that findings may not generalize to the general population due to the racial breakdown within the present sample (62.9% African-American, 33% White) being notably different from that within the US population.
Conclusions
In conclusion, we did not find marijuana use to be associated with poorer quality sleep among pregnant women. Future studies should recruit larger numbers of women with both sleep and marijuana use measured prospectively across pregnancy in each trimester, to better distinguish timing of use and timing of sleep difficulties. Objective sleep measures, such as actigraphy, that are readily adaptable to home use should be utilized in future work. Such efforts would provide further insight toward how marijuana use may or may not impact sleep during pregnancy. Nevertheless, the present study offers no evidence that marijuana is helpful or detrimental to sleep during pregnancy. Given the established harms to the developing fetus, marijuana remains contraindicated during pregnancy and advice to abstain should consistently be part of routine prenatal care. Because the prevalence of women using marijuana during pregnancy is increasing and many women perceive it to be minimal to no harm, treatment providers should be cognizant of potential need to offer information regarding potential risks as well as service connection to support motivation for behavioral change and reduced substance use among women who disclose marijuana use during prenatal care.
Funding
This work was supported by the National Institute on Drug Abuse [R01DA042948]; the March of Dimes Foundation [grant# 6-FY16-160]; and the National Center for Advancing Translational Sciences/National Institutes of Health [UL1TR001070]. None of the funding sources had input in the study design; the data collection, interpretation or analysis; the writing of this report; or the decision to submit the article for publication.
Footnotes
Disclosure statement
The authors have no conflicts of interest to disclose.
References
- [1].Sedov I, Cameron E, Madigan S, et al. Sleep quality during pregnancy: a meta-analysis. Sleep Med Rev. 2018;38:168–176. [DOI] [PubMed] [Google Scholar]
- [2].Facco F, Parker C, Reddy U, et al. Association between sleep-disordered breathing and hypertensive disorders of pregnancy and gestational diabetes mellitus. Obstet Gynecol. 2017;129(1):31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Sweet L, Arjyal S, Kuller J, et al. A review of sleep architecture and sleep changes during pregnancy. Obstet Gynecol Surv. 2020;75(4):253–262. [DOI] [PubMed] [Google Scholar]
- [4].Bacaro V, Benz F, Pappaccogli A, et al. Interventions for sleep problems during pregnancy: a systematic review. Sleep Med Rev. 2020;50:101234. [DOI] [PubMed] [Google Scholar]
- [5].Blair L, Porter K, Leblebicioglu B, et al. Poor sleep quality and associated inflammation predict preterm birth: heightened risk among African Americans. Sleep. 2015;38(8):1259–1267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Naghi I, Keypour F, Ahari S, et al. Sleep disturbance in late pregnancy and type and duration of labour. J Obstet Gynaecol. 2011;31(6):489–491. [DOI] [PubMed] [Google Scholar]
- [7].Volkow N, Han B, Compton W, et al. Self-reported medical and nonmedical cannabis use among pregnant women in the United States. JAMA. 2019;322(2): 167–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Choi S, Huang B, Gamaldo C. Therapeutic uses of cannabis on sleep disorders and related conditions. J Clin Neurophysiol. 2020;37(1):39–49. [DOI] [PubMed] [Google Scholar]
- [9].Holland C, Nkumsah M, Morrison P, et al. Anything above marijuana takes priority”: obstetric providers’ attitudes and counseling strategies regarding perinatal marijuana use. Patient Educ Couns. 2016;99(9): 1446–1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Chang J, Tarr J, Holland C, et al. Beliefs and attitudes regarding prenatal marijuana use: perspectives of pregnant women who report use. Drug Alcohol Depend. 2019;196:14–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Young-Wolff K, Gali K, Sarovar V, et al. Women’s questions about perinatal cannabis use and health care providers’ responses. J Women’s Health. 2020;29(7): 919–926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Sharapova SR, Phillips E, Sirocco K, et al. Effects of prenatal marijuana exposure on neuropsychological outcomes in children aged 1–11 years: a systematic review. Paediatr Perinat Epidemiol. 2018;32(6): 512–532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Suraev AS, Marshall NS, Vandrey R, et al. Cannabinoid therapies in the management of sleep disorders: a systematic review of preclinical and clinical studies. Sleep Med Rev. 2020;53:101339. [DOI] [PubMed] [Google Scholar]
- [14].Memedovich KA, Dowsett LE, Spackman E, et al. The adverse health effects and harms related to marijuana use: an overview review. CMAJ Open. 2018;6(3): E339–E346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Klebanoff M, Wilkins D, Keim S. Marijuana use during pregnancy and preterm birth: a prospective cohort study. Am J Perinatol. 2020. DOI: 10.1055/s-0040-1708802 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Buysse D, Reynolds IIC, Monk T, et al. The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2): 193–213. [DOI] [PubMed] [Google Scholar]
- [17].Radloff L The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1(3):385–401. [Google Scholar]
- [18].Spielberger C Test anxiety inventory. In The Corsini encyclopedia of psychology. Vol. 1. Hoboken (NJ): John Wiley & Sons; 2010. [Google Scholar]
- [19].Cohen S, Kamarck T, Mermelstein R. Perceived stress scale. In Measuring stress: a guide for health and social scientists. Vol. 10. Oxford (England): Oxford University Press; 1994. [Google Scholar]
- [20].Williams D, Yu Y, Jackson J, et al. Racial differences in physical and mental health: socio-economic status, stress, and discrimination. J Health Psychol. 1997;2(3): 335–351. [DOI] [PubMed] [Google Scholar]
- [21].Klebanoff M, Fried P, Yeates K, et al. Lifestyle and Early Achievement in Families (LEAF) study: design of an ambidirectional cohort study of prenatal marijuana exposure and child development and behaviour. Paediatr Perinat Epidemiol. 2020;34(6):744–756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Substance Abuse and Mental Health Services Administration. Clinical Drug Testing in Primary Care. Technical Assistance Publication (TAP) 32. HHS Publication No. (SMA) 12–4668. Rockville (MD): Substance Abuse and Mental Health Services Administration; 2012. [Google Scholar]
- [23].Qiu C, Frederick IO, Sorensen T, et al. Sleep disturbances among pregnant women with history of migraines: a cross-sectional study. Cephalalgia. 2015; 35(12):1092–1102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Gunning M, Denison F, Stockley C, et al. Assessing maternal anxiety in pregnancy with the State-Trait Anxiety Inventory (STAI): issues of validity, location and participation. J Reprod Infant Psychol. 2010;28(3): 266–273. [Google Scholar]
- [25].Cnaan A, Laird N, Slasor P. Using the general linear mixed model to analyse unbalanced repeated measures and longitudinal data. Statist Med. 1997;16(20): 2349–2380. [DOI] [PubMed] [Google Scholar]
- [26].Liu Y, De A. Multiple imputation by fully conditional specification for dealing with missing data in a large epidemiologic study. Int J Stats Med Res. 2015;4(3): 287–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Berglund P, Heeringa S. Multiple imputation of missing data using SAS. Cary (NC): SAS Institute; 2014. [Google Scholar]
- [28].SAS [computer program]. Version 9.4. Cary (NC): SAS Institute Inc; 2014. [Google Scholar]
- [29].Young-Wolff K, Tucker L, Alexeff S, et al. Trends in self-reported and biochemically tested marijuana use among pregnant females in California from 2009–2016. JAMA. 2017;318(24):2490–2491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Hedman C, Pohjasvaara T, Tolonen U, et al. Effects of pregnancy on mothers’ sleep. Sleep Med. 2002;3(1): 37–42. [DOI] [PubMed] [Google Scholar]
- [31].Mindell J, Jacobson B. Sleep disturbances during pregnancy. J Obstet Gynecol Neonatal Nurs. 2020; 29(6):590–597. [DOI] [PubMed] [Google Scholar]
- [32].Facco F, Kramer J, Ho K, et al. Sleep disturbances in pregnancy. Obstet Gynecol. 2010;115(1):77–83. [DOI] [PubMed] [Google Scholar]
- [33].Oken B, Yu J, Neuendorf R, et al. Systematic review of mind-body interventions on sleep. J Altern Complement Med. 2014;20(5):A65. [DOI] [PMC free article] [PubMed] [Google Scholar]