Abstract
The objective of this study was to examine relationships between maternal perceptions of Fetal Health Locus of Control (FHLC) and perceived risk/benefit related to prenatal marijuana use with maternal marijuana use behaviors during pregnancy. The sample included women seeking prenatal care at The Ohio State University Wexner Medical Center (OSUWMC, Columbus OH) between 2010 and 2015, who participated in a follow-up study (2019–2020). Logistic regression models were run to estimate associations between maternal perceptions and prenatal marijuana use behaviors. Higher perceived benefit to mother (AOR = 1.53, 95% CI: 1.08–2.17) and lower perceived risk to children (AOR = 0.59, 95% CI: 0.45–0.78) were related to increased prenatal marijuana use, adjusted for confounders. In contrast, FHLC perceptions were not associated with maternal marijuana use; however, FHLC predicted women’s perceptions of risk and benefit which in turn was associated with marijuana use during pregnancy. Understanding how maternal perceptions influence health-related risk behaviors during pregnancy is important because perceptions are feasible intervention targets.
Introduction
The prevention of health-related risk behaviors during pregnancy and associated adverse pregnancy outcomes constitute important public health concerns. Women’s health-related beliefs have a strong impact on their utilization of health services and health-risk behaviors such as tobacco use (Ashford & Rayens, 2015; Haslam et al., 2003). As legislation and social attitudes continue to soften on the criminalization and recreational use of marijuana, the number of women using marijuana during pregnancy has rapidly increased (Volkow et al., 2019). Between the years of 2002 and 2017, the number of pregnant women reporting use in the past month (3.4%–7.0%), use in the first trimester (5.7%–12.1%), use in the second trimester (0.6%–2.5%), use in the third trimester (0.5%–2.5%), and daily use of marijuana (0.9%–3.4%) have all increased (Volkow et al., 2019). Research suggests putative, long-term negative effects of prenatal marijuana exposure on children’s behavior and development (Sharapova et al., 2011). How women’s health-related beliefs relate to their marijuana use during pregnancy remains unknown. This information is important as health-related beliefs regarding marijuana use during pregnancy could be modified by intervention to promote subsequent reductions in marijuana use and, in turn, improve child outcomes.
Fetal Health Locus of Control (FHLC) refers to mothers’ beliefs and perceptions regarding factors responsible for their newborn child’s health and well-being (Labs & Wurtele, 1986). FHLC is comprised of three distinct domains: (1) internal control; (2) powerful others control; and (3) chance control. Internal control refers to a woman’s perception that her actions and behaviors during pregnancy are responsible for the health and well-being of her unborn child. Women’s perception of the degree to which health professionals have impact on their unborn child’s wellbeing are captured by the powerful others domain. Lastly, chance control refers to the level at which women perceive chance factors to influence their child’s wellbeing.
FHLC beliefs are associated with a variety of health behaviors and utilization of health services (Ashford & Rayens, 2015; Haslam et al., 2003; Kordi et al., 2017; Stewart & Streiner, 1994, 1995). Women with high levels of internal FHLC beliefs (perception their behavior impacts their child) are more likely to modify risk behaviors during their pregnancy resulting in less maternal risk behaviors. Higher levels of internal FHLC are linked to reductions in substance use (tobacco and alcohol), increased physical activity, and increased healthcare seeking behavior, as well as better adherence to healthcare recommendations among pregnant women (Ashford & Rayens, 2015; Haslam & Lawrence, 2004; Stewart & Streiner, 1994). In contrast, women who continue to engage in health-risk behaviors during pregnancy, such as smoking tobacco, report higher chance-related FHLC compared to their counterparts (Ashford & Rayens, 2015). Powerful others FHLC is often unrelated to health-risk behaviors during pregnancy compared to other FHLC domains (Ashford & Rayens, 2015; Kordi et al., 2017). Interventions that target increasing women’s internal FHLC and reducing chance-related FHLC have shown success (Mirghafourvand et al., 2020). Taken together, women’s FHLC beliefs influence their decision-making regarding a variety of health-risk behaviors during pregnancy and can be readily targeted and modified by intervention. However, the relationships between these FHLC beliefs and maternal marijuana use during pregnancy remains unknown.
The current study assesses the relationships between women’s FHLC, perceptions of risk/benefit related to marijuana use while pregnant, and marijuana use behaviors during pregnancy among a cohort of women with young children, who participated in a perinatal research repository study. Specifically, we seek to estimate associations between maternal marijuana use behaviors during pregnancy and two relevant constructs: (1) FHLC and (2) perceptions of risk and benefit of marijuana use during pregnancy. Interventions that target FHLC beliefs and maternal perceptions of risk represent a feasible avenue for providing pregnant women with accurate information related to their ability to positively impact their child’s well-being through modified health-risk behaviors, such as abstaining from marijuana use, in an effort to support informed decision-making over the course of pregnancy. However, the effectiveness of these strategies is contingent on understanding maternal perceptions that relate to continued marijuana use through pregnancy.
Methods
Participants
This study utilized data from the Ohio Perinatal Research Network’s Perinatal Research Repository (PRR), which has been described previously (Klebanoff et al., 2021). In brief, the repository recruited women, between 2010 and 2015, from among those who attended antenatal clinics at The Ohio State Wexner Medical Center (OSUWMC, Columbus OH). Women were eligible if they were between the ages of 16 to 50 years, could communicate in English, and planned to deliver at the OSUWMC. Women were enrolled as early as possible in pregnancy, which was dependent on when they sought care. At enrollment, women completed various questionnaires, which covered a variety of medical, demographic, socio-economic, and substance use domains. At each trimester, women followed-up and provided urine and blood samples, which were stored at −80°C, and then assayed as part of a subsequent study (see below). Women who enrolled in the PRR provided written informed consent, which included an option to allow re-contact for future IRB-approved research. Roughly 75% of repository-enrolled women consented to be contacted for future research.
As part of a subsequent study, Lifestyle and Early Achievement in Families study (LEAF; Klebanoff et al., 2020), women who consented to be re-contacted were contacted, between 3 and 7 years post-delivery. As part of this subsequent study, women completed a survey assessing their perceptions regarding a hypothetical pregnancy including: perceptions of fetal health locus of control (FHLC) and the risks and benefits of marijuana use while pregnant. These surveys were completed during the years of 2019 and 2020, when women’s children were between the ages of 3.5 and 7 years of age. The current study utilized data from this follow-up survey as well as information from the research repository regarding maternal demographics during prenatal care and substance use during pregnancy.
The PRR and subsequent LEAF study were approved by the IRB at Nationwide Children’s Hospital (NCH) and a reliance agreement was executed with the IRB at the OSUWMC.
Measures
Perceptions of risk and benefit
As part of the LEAF study, women were asked to rate their perception of how risky marijuana use was during pregnancy in relation to a hypothetical pregnancy. This item included five options: (1) Not at all risky to baby; (2) Hardly risky to baby; (3) Somewhat risky to baby; (4) Risky to baby; and (5) Extremely risky to baby. Women were also queried about their perceptions of benefit to mothers who engaged in marijuana use while pregnant, via a 5-item scale: (1) Not at all good for herself; (2) hardly good for herself; (3) somewhat good for herself; (4) good for herself; and (5) extremely good for herself.
Fetal health locus of control
The Fetal Health Locus of Control Scale (FHLCS; Labs & Wurtele, 1986) was administered to assess FHLC as part of the LEAF assessment. The FHLCS includes three subscale scores representing the three dimensions of FHLC (internal control, powerful others, and chance). Each subscale included six items, with each item scored on a 6-point likert scale, with “1” representing “strongly disagree” and “6” representing “strongly agree” (Baldwin, 2006; Labs & Wurtele, 1986). The total item score is summed for each subscale resulting in three subscale scores ranging from 6 to 36.
Marijuana use
Maternal marijuana use during pregnancy was assessed using three measurement tools. First, urine samples from the research repository 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 (Foltz et al., 1983; Huang et al., 2001; Quintela et al., 2007; Wilkins et al., 1995). Secondly, women’s obstetrical record was abstracted by a trained research nurse. Last, women completed a self-report assessment of their substance use during the PRR intake. Women were determined to have used marijuana while pregnant, if marijuana use was noted during pregnancy on the intake questionnaire, record abstraction, or if any urine screen had a Δ9-THC-COOH concentration of >15 ng/ml, the concentration considered to represent active use when employing mass spectrometry (SAMSHA, 2012).
Demographics
As part of the research repository, women completed intake questionnaires that documented demographic information at the time of the prenatal visit including: maternal age, race, marital status, and education. Women also provided information regarding tobacco use during their pregnancy.
Statistical Analysis
Descriptive analyses were run to categorize the present sample. Pearson’s correlation coefficient and Point Biserial correlation coefficient tests were run to explore the presence of relationships between FHLC perceptions, perceptions of risk/benefit associated with marijuana use during pregnancy, and marijuana use behaviors during pregnancy. Chi-square and independent samples t-tests were run to explore group differences in demographic factors (race, age, education, marital status, etc.) between women who did and did not use marijuana during their pregnancy. Logistic regression models were run to estimate associations between aforementioned maternal perceptions (FHLC and risk/benefit related to marijuana use during pregnancy) and marijuana use during pregnancy. Adjusted models included theorized confounds identified a priori including: maternal race, age, education, marital status, and tobacco use during pregnancy. Confounds controlled for in the models were identified from prior literature and prior findings from the LEAF cohort that reflect significant relationships between these factors and maternal marijuana use as well as maternal FHLC perceptions (Ashford & Rayens, 2015; Bolhuis et al., 2018; Murnan et al., 2021; Sharapova et al., 2011; Sonon et al., 2015). Linear regression analyses were also run to examine relationships between FHLC and perceptions of risk/benefit related to marijuana use during pregnancy. The current study represents a secondary analysis; therefore, the study was not specifically powered to address the current research questions in particular.
Results
The current sample included 183 mothers’ survey assessments coupled with PRR data from during their pregnancy. Of the sample, 120 mothers self-identified as African-American, 58 as White, and 5 as “Other.” Women’s ages at the time of prenatal visit/repository assessment ranged from 16 to 47 years of age (µ = 27.3, SD = 5.7). Fifty-five (30.2%) mothers in the sample were identified as using marijuana during their pregnancy per their repository data. Women who used marijuana were more likely to be Black (χ2 = 5.5, p = .02), more likely to use tobacco during their pregnancy (p < .001), and less likely to be married (p < .001). For additional details pertaining to the current sample and group comparisons between women in the sample who did and did not use marijuana during their pregnancy, see Table 1.
Comparison Across Sample Characteristics Between Participant Women Who Did and Did Not Use Marijuana During Their Pregnancy (n = 183).
Note. HS = high school; GED = graduate equivalence degree.
p-values correspond with chi-square tests (or independent samples t-tests) run to explore group differences between women who did and did not use marijuana during their pregnancy.
Missing data: maternal age (0); marital status (20); maternal education (14); health insurance (17); tobacco use during pregnancy (1).
Generally, women perceived marijuana use during pregnancy to be risky to their child (µ = 4.2; SD = 1.4, Median = 5.0) and low benefit to themselves (µ = 1.5; SD = 1.0, Median = 1.0). Data for perceived risk to child was positively skewed (Skewness = −1.6) and Platykurtic (Kurtosis = 0.9). When asked about perceived risk to child related to marijuana use during pregnancy, 127 (69.4%) participants described it as “Extremely Risk to Baby”; 18 (9.8%) described it as “Risky to Baby”; 8 (4.4%) described it as “Somewhat Risky to Baby”; 10 (5.5) described it as “Hardly Risky to Baby”; and 19 (10.4%) described it as “Not at All Risky to Baby.” In contrast, data for perceived benefit to mother was negatively skewed (Skewness = 2.2) and Leptokurtic (Kurtosis = 4.1). When asked about perceived benefit to mother related to marijuana use while pregnant, 142 (77.6%) participants described it as “Not At All Good for Herself”; 15 (8.2%) described it as “Hardly Good For Herself”; 11 (6.0%) women described it as “Somewhat Good for Herself”; 8 (4.4%) women described it as “Good for Herself”; 6 (3.3%) described it as “Extremely Good for Herself.” Women’s perception of risk for child associated with maternal marijuana use during pregnancy was negatively related to their perceptions that prenatal marijuana use was beneficial to mothers while pregnant (r(182) = −.23, p = .002) and were positively associated with women’s internal fetal health locus of control (r(182) = .16, p = .03).
Contrary to study hypotheses, neither unadjusted or adjusted logistic regression models indicated FHLC perceptions (internal, chance, powerful others) were related to maternal marijuana use during pregnancy. Logistic regression models estimating associations between maternal perceptions and marijuana use during pregnancy indicate that women who perceived marijuana to be risky was associated with reduced maternal marijuana use during pregnancy (adjusted OR = 0.59, 95% CI: 0.45–0.78). That is to say, for each point increase in perceived risk to child there is a .59 times lower odds of mother using marijuana while pregnant while holding all other variables constant. Additionally, increases in maternal perceptions that marijuana use is beneficial to women while pregnant related to increased maternal marijuana use during pregnancy (adjusted OR = 1.53, 95% CI: 1.08–2.17). That is to say, for each point increase in perceived benefit to mother there is a 1.53 times higher odds of mother using marijuana while pregnant while holding all other variables constant. The adjusted odds ratio co-efficient corresponds with the change in odds of maternal marijuana use for each point increase in maternal risk perception. Findings from these models are reported in Table 2
Associations Between Maternal Perceptions and Marijuana Use During Pregnancy, Logistic Regression Models (n = 183).
Adjusted Models include maternal race, age, education, marital status, and tobacco use during pregnancy as co-variates.
Odds ratios co-efficient corresponds with the change in odds of maternal marijuana use for each point increase in maternal risk perception, while holding all other variables constant.
Linear regression models were also run to explore potential pathways that may connect FHLC perceptions to maternal marijuana use during pregnancy via indirect effects (maternal perceptions of risk/benefit of marijuana use). Internal FHLC significantly predicted perceptions of risk to child related to maternal marijuana use during pregnancy (β = .05; 95% CI: 0.01–0.09). Internal FHLC did not predict perceptions of benefit to mothers related to maternal marijuana use during pregnancy (β = −.014; 95% CI: −0.05, 0.02). No significant associations were observed between chance-related FHLC and maternal perceptions of risk to child (β = .002; 95% CI: −0.03, 0.04) or benefit to mother (β = .02; 95% CI: 0.00, 0.04) related to marijuana use during pregnancy. Similarly, no significant associations were observed between powerful others FHLC and maternal perceptions of risk to child (β = .006; 95% CI: −0.03, 0.04) or benefit to mother (β = −.004; 95% CI: −0.02, 0.03) related to marijuana use during pregnancy.
Discussion
The current study reports on relationships between maternal FHLC, perceptions of risk/benefit related to prenatal marijuana use, and maternal marijuana use during pregnancy. In line with prior research, women who used marijuana within the sample were more likely to be Black, unmarried, and to report tobacco use during pregnancy (Bolhuis et al., 2018; Goldschmidt et al., 2000, 2008; Gray et al., 2005; Richardson et al., 2002; Sonon et al., 2015). Maternal perceptions that marijuana use while pregnant had lower risk to child and greater benefit to mother related to increased probability of marijuana use during pregnancy. Contrary to study hypotheses, FHLC did not relate to maternal marijuana use during pregnancy.
Extensive evidence across areas of research (tobacco use, cybersecurity, cardiovascular disease, sexual behaviors, bike helmet safety, diet, etc.) reflect that perceptions of risk related to a behavior/activity strongly influence individuals’ engagement in that behavior/activity (Bajos, 1997; Boss et al., 2015; Hasan et al., 2017; Huang et al., 2011; Slovic, 2000, 2010; Weinstein et al., 2005). Present findings were in line with these past works as maternal perceptions of risk for children and benefit for mothers associated with prenatal marijuana use related to maternal marijuana use behaviors during pregnancy. These findings reflect women’s perceptions of risk and benefits influence and inform their decision-making and likelihood their use of marijuana during pregnancy. The provision of information related to risks of marijuana use during pregnancy by healthcare providers may stand to influence women’s perceptions and subsequent decision-making related to marijuana use, disclosure to providers, and seeking treatment or alternative options. It is important to note that substance use does not occur in a vacuum. For some women, providers may also need to help women address underlying contributors to their substance use in order to affect marijuana use behavior change.
Findings did not reflect differences in maternal FHLC perceptions between women who did and did not use marijuana during their pregnancy. Null findings are contrary to study hypotheses and documented relationships between FHLC and other risk behaviors during pregnancy. In prior studies focused on a variety of other health-related risk behaviors, women with high internal FHLC were more likely to reduce risk behaviors (such as tobacco use) and more likely to engage in positive health behaviors during pregnancy (exercise, healthy diet, etc.; Ashford & Rayens, 2015; Haslam & Lawrence, 2004; Stewart & Streiner, 1995). Previous studies also repeatedly demonstrate women engaged in health risk behaviors report higher chance-related FHLC (Ashford & Rayens, 2015). In the present study, internal FHLC related to increased perceptions of risk, which in turn related with reduced maternal marijuana use during pregnancy. This may suggest indirect pathways in which internal FHLC indirectly is associated with reduced marijuana use behaviors during pregnancy. Reduced clarity regarding harms of prenatal marijuana exposure may make these relationships less straight forward than with tobacco use, where the body of evidence is larger and treatment providers recommendations are more consistent and uniform. Future studies should seek to explore this mechanism as a pathway to understanding how FHLC perceptions may influence maternal marijuana use during their pregnancy.
Limitations
One limitation is that the assessment of maternal perceptions including: FHLC and perceived risk/benefits of marijuana use was collected a few years after women gave birth. In contrast, maternal marijuana use was collected during their pregnancy using multiple data sources. While we would expect observed relationships between perceived risk/benefit associated with marijuana use and actual maternal marijuana use behaviors to persist, it may be the case that the collection of maternal perception and prenatal marijuana use data at different times may have contributed to null findings related to direct relationships between maternal FHLC perceptions and prenatal marijuana use behaviors. It may be the case that women’s experiences related to their prenatal marijuana use may influence their FHLC and risk/benefit perceptions over time. This limitation should be considered when interpreting current findings and future research should seek to explore these relationships prospectively by following a prenatal cohort and measuring maternal perceptions and prenatal marijuana use repeatedly.
Conclusions
The present study provides insight into the relationships between maternal perceptions of FHLC as well as risks and benefits related to marijuana use during pregnancy with maternal marijuana use behaviors. Findings did not provide support for associations between maternal FHLC and maternal marijuana use during pregnancy; however, FHLC did predict women’s perceptions of risk to child from marijuana use during pregnancy. In turn, perceptions of risk related to marijuana use influenced maternal marijuana use behaviors while pregnant. Women who perceived less risk to child also perceived greater benefit to expectant mothers. Additional research in this area is needed to disentangle and explore the nuanced interplay between FHLC, maternal perception of risk and benefit related to marijuana use while pregnant, and marijuana use during pregnancy. Understanding how maternal perceptions intersect and influence health-related risk behaviors during pregnancy is important because perceptions are feasible intervention targets.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The current study was funded by the National Institute on Drug Abuse (NIDA; R01DA042948-02S1), the March of Dimes Foundation (grant #6-FY16-160), and the National Center for Advancing Translational Sciences/ National Institutes of Health (UL1TR001070). The funding sources had no 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.
