Abstract
Background
The influence of marijuana use on human fertility has not been well studied. We evaluated the association between female and male use of marijuana and fecundability in Pregnancy Study Online, a prospective cohort of North American couples.
Methods
Female participants completed a baseline questionnaire on which they reported lifestyle and behavioural factors, including frequency of marijuana use within the previous 2 months. Male partners completed an optional baseline questionnaire on similar factors, including marijuana use. Women completed follow-up questionnaires every 8 weeks for 12 months or until pregnancy, initiation of fertility treatment or loss to follow-up, whichever came first. The analysis was restricted to 4194 women (1125 couples) with ≤6 cycles of pregnancy attempt time at study enrolment (2013–2017). Fecundability ratios (FR) and 95% CIs were estimated using proportional probabilities regression models, with adjustment for potential confounders.
Results
Men (14.2%) were more likely than women (11.6%) to be marijuana users. FRs for female marijuana use <1 and ≥1 time/week relative to non-use were 0.99 (95% CI 0.85 to 1.16) and 0.98 (95% CI 0.80 to 1.20), respectively. FRs for male marijuana use <1 and ≥1 time/week relative to non-use were 0.87 (95% CI 0.66 to 1.15) and 1.24 (95% CI 0.90 to 1.70), respectively. Associations for frequent marijuana use (≥1 time/week) were attenuated among non-smoking men (FR=1.21, 95% CI 0.84 to 1.74), but stronger among men reporting intercourse ≥4 times/week (FR=1.35, 95% CI 0.72 to 2.53).
Conclusions
In this preconception cohort study, there was little overall association between female or male marijuana use and fecundability.
INTRODUCTION
About 15% of couples experience infertility, which is defined clinically as the inability to conceive within 12 months of unprotected intercourse.1 Infertility is associated with psychological and financial hardship for affected couples, and costs the US healthcare system an estimated $5 billion per year.2 Few modifiable risk factors for infertility have been identified. Thus, the identification of modifiable behavioural and lifestyle risk factors for infertility is important.
Marijuana is one of the most widely used recreational drugs in North America among individuals of reproductive age, and its prevalence in North America is among the highest worldwide.3 Since 1996, twenty-eight states and Washington, DC, have passed legislation permitting use of marijuana for medical or recreational purposes.4 Estimates of past-year marijuana use in the USA during 2012–2013 among those aged ≥18 years ranged from 6.9% among women to 12.3% among men.5 Although recreational use of marijuana is not legal in Canada, it also has a high prevalence of past-year use (7% among women and 14% among men aged ≥15 years in 2013).6
In female animals ranging from rodents to non-human primates, delta-9-tetrahydrocannabinol (Δ9-THC), the active ingredient in marijuana, has been associated with reduced gonadotropin levels (via suppression of luteinising hormone (LH) pulsatile secretion),7–9 disrupted ovulation10 and menstrual irregularities.7–14 However, in one study of primates, normal menses returned after 3–4 months of chronic exposure.13 In women, studies are more conflicting. One study found a 30% decrease in LH levels among marijuana smokers in the luteal phase, but little effect in the follicular phase.15 In another study, exposure in the periovulatory stage was associated with increased LH levels.16 While intensive marijuana smoking was associated with normal menses in one study,17 other studies have shown increased anovulation,18 19 longer follicular phases18 and shorter luteal phases.19 Finally, recent marijuana use has been associated with reduced oocyte retrieval and fertilisation among couples undergoing in vitro fertilisation20 and an increased risk of ovulatory infertility.21
In male animals, studies have shown inhibition of Leydig cell function,22 reductions in testosterone, gonadotropins23–25 and testicular size,26–28 and abnormal sperm morphology29 30 following acute exposure to Δ9-THC.23–25 Dose-dependent decreases in LH also have been observed with chronic exposure to Δ9-THC,31 but effects were smaller,23 likely due to tolerance.32 Exposure to Δ9-THC 4–5 weeks before and during mating had little effect on fertility in mice.33 34 Although some studies report that chronic marijuana use in men is associated with lower testosterone and LH levels35 36 and poor semen quality,35 37–40 many studies have not confirmed these findings37 38 41–44 and reversible effects have been observed 5–6 weeks after initiation.39 A large population-based Danish study reported an increase in testosterone levels among recent marijuana smokers.38 The direct influence of male marijuana use on fertility has not been studied.
We examined the extent to which marijuana use is associated with fecundability—the average per-cycle probability of conception—in a preconception cohort study of North American couples. Fecundability, the average probability of conception in a menstrual cycle of regular unprotected intercourse, provides a direct measure of couple fecundity. Prospective cohort study designs of fecundability have several advantages over retrospective cohort studies of already-pregnant couples; they enrol a wider spectrum of couples (not just fertile couples), they allow ascertainment of exposures before conception (when reporting accuracy is higher) and they limit selection bias, which might arise if participation depends on both marijuana use and fertility.
METHODS
Study population
Pregnancy Study Online (PRESTO) is an ongoing web-based preconception cohort study of pregnancy planners. The study methods have been described in detail elsewhere.45 Briefly, women aged 21–45 years residing in the USA or Canada, who are in a stable relationship with a male partner and who are not using contraception or fertility treatment at study entry are eligible for participation. Female participants complete an online baseline questionnaire with items on demographics, behavioural factors, medical and reproductive histories, and medication use. After completion of the baseline questionnaire, women are given the option to invite their male partners to participate. Men aged ≥21 years are eligible. Male participation involves completion of a baseline questionnaire similar to the female baseline questionnaire.
Ten days after enrolment, participants are invited to complete a baseline questionnaire and the National Cancer Institute’s Dietary Health Questionnaire II,46 a web-based food frequency questionnaire (FFQ). Women completed follow-up questionnaires every 8 weeks for up to 12 months to ascertain pregnancy status and updated information on any factors that may have changed over time. PRESTO was approved by the Institutional Review Board at Boston Medical Center, and online informed consent was obtained from all participants.
From June 2013 to August 2017, up to 5394 eligible women completed the baseline questionnaire. We excluded 141 women whose baseline date of last menstrual period (LMP) was >6 months before study entry, and 78 women with missing/implausible LMP data or who were pregnant at study entry. We then excluded 981 women who had been trying to achieve pregnancy for >6 cycles at enrolment, to reduce potential for changes in marijuana use in response to subfertility. The final study population comprised 4194 women, of whom 55% invited their male partners to participate. A total of 1125 (49%) men successfully enrolled in the study.
Assessment of marijuana use and covariates
On the female and male baseline questionnaires, participants reported their marijuana use during the previous 2 months and frequency of marijuana use during that time period, with response options of ‘every day’, ‘4–6 times per week’, ‘1–3 times per week’ and ‘less than 1 time per week’. Covariate data for both partners were collected on age, geographic region of residence, race/ethnicity, education, use of multivitamins or folate supplements, height, weight, cigarette smoking history, exposure to environmental tobacco smoke, physical activity, alcohol consumption, caffeine intake, soda intake, reproductive history, intercourse frequency, history of sexually transmitted infections (STI), history of physician-diagnosed medical conditions (eg, depression, anxiety), employment status, average work hours per week, average sleep duration in the previous month, the 10-item version of the Perceived Stress Scale (PSS-10)47 and depressive symptoms via the Major Depression Inventory (MDI).48 The PSS-10 and MDI were added to the male baseline questionnaire approximately 2 years after study initiation. Women additionally reported at baseline on their last method of contraception, whether they were doing anything to improve their chances of conception (eg, charting menses, basal body temperature, ovulation predictor kit), and household income. We used FFQ data to calculate the Healthy Eating Index (HEI), a measure of overall diet quality.49
Assessment of time to pregnancy
We estimated time to pregnancy (TTP) using data from the female baseline and follow-up questionnaires. At baseline, women reported their LMP date, usual cycle length (regularly cycling women only) and the number of cycles they had attempted conception. On each follow-up questionnaire, women reported their most recent LMP date and whether they had become pregnant since the previous questionnaire. Among women with irregular cycles, defined as those who reported not being able to ‘predict from one menstrual period to the next about when the next menstrual period would start,’ we estimated cycle length based on date of LMP at baseline and prospectively reported LMP dates during follow-up. TTP was calculated as follows: menstrual cycles of attempt at study entry + [(LMP date from the most recent follow-up questionnaire − date of baseline questionnaire completion)/usual menstrual cycle length] + 1.
Data analysis
Couples who did not conceive within 12 cycles of attempted conception were censored at 12 cycles. Couples contributed at-risk menstrual cycles to the analysis from study entry until a reported pregnancy (57.6%) or a censoring event (initiation of fertility treatment: 7.0%; loss to follow-up: 20.4%; or 12 cycles of attempt time: 15.0%), whichever came first. Participants who were and were not lost to follow-up were similar according to mean age (29.8 vs 30.1 years), alcohol intake (3.4 vs 3.0 drinks/week) and parity (31.0% vs 28.9%), but differed according to body mass index (BMI) (29.0 vs 26.5 kg/m2), race/ethnicity (non-Hispanic White: 79.1% vs 85.5%), education (<16 years: 33.6% vs 19.8%), income (<$50 000: 27.2% vs 16.0%), current cigarette smoking (8.9% vs 5.1%) and current marijuana use (15.4% vs 10.6%).
We used life table methods to compute the proportion of couples who conceived during follow-up, accounting for censoring. To account for variation in attempt times at study entry (range: 0–6 cycles) and to reduce bias due to left truncation, we analysed observed cycles only using the Anderson-Gill data structure.50 We used proportional probabilities regression models51 to estimate fecundability ratios (FR) and 95% CIs for the association between marijuana use and fecundability. The FR represents the ratio of fecundability in each exposure category compared with the reference category. Our models incorporated the baseline decline in fecundability over time and accounted for left truncation.
Frequency of marijuana use in the previous 2 months was categorised as follows: no current use, <1 time per week and ≥1 time/week. Time-varying analyses of marijuana use were performed for female participants only. Potential confounders were selected based on prior literature and assessment of a causal graph. We considered potential determinants of subfertility that were associated with marijuana use. Final models examining the effects of female marijuana use were adjusted for female age (<25, 25–29, 30–34, ≥35 years), race/ethnicity (non-Hispanic White, other race/ethnicity), education (less than college degree, college degree, graduate school), annual household income (<US$50 000, US$50 000–US$99 999, US$100 000–US$149 999, ≥US$150 000), cigarette smoking (never, former, current occasional, current regular), alcohol use (<1, 1–6, 7–13, ≥14 drinks/week), caffeine intake (continuous mg/day), BMI (<25, 25–29, 30–34, ≥35 kg/m2), multivitamin or folate use (yes, no), HEI score (continuous), average sleep duration (<7, 7–8, ≥9 hours/night), employment status (employed, unemployed), intercourse frequency (<1, 1–3, ≥4 times/week), PSS-10 score (continuous), MDI score (continuous), vigorous physical activity (<1, 1–2, 3–4, ≥5 hours/week) and doing something to improve chances of conception (yes, no).
Final models examining the effects of male marijuana use were adjusted for male age (<30, 30–34, ≥35 years), race/ethnicity (non-Hispanic White, other race/ethnicity), education (less than college degree, college degree, graduate school), annual household income (<US$50 000, US$50 000–US$149 999, ≥US$150 000), cigarette smoking (never, former, current occasional, current regular), alcohol use (<1, 1–6, 7–13, ≥14 drinks/week), caffeine intake (continuous mg/day), sugar-sweetened beverage intake (0, 1, 2–6, ≥7 drinks/week), average sleep duration (<7, 7–8, ≥9 hours/night), employment status (employed, unemployed), intercourse frequency (<1, 1–3, ≥4 times/week), PSS-10 score (continuous), MDI score (continuous) and doing something to improve chances of conception (yes, no). Couple-based exposure models were adjusted for the female and male covariates listed above.
In secondary analyses, we stratified by attempt time at study entry (<3 vs 3–6 cycles) to assess the extent to which ‘reverse causation’ explained our results (eg, if subfertility caused a change in marijuana use). We also stratified final models by female age (<30 vs ≥30 years) because ovulatory infertility increases with age and age could be an effect modifier of the association. Given that parity could be a causal intermediate, we stratified models by parity and also fit models with and without control for parity. Finally, because cigarette smoking is a potential confounder and is itself a risk factor for infertility,52 subanalyses were restricted to non-smokers of cigarettes.
We used multiple imputation to handle missing data for exposures, covariates and pregnancy status.53 We created five imputed data sets with SAS PROC MI and combined coefficient and SE estimates from the five data sets using SAS PROC MIANALYZE.54 Missingness ranged from 0.1% (marijuana use, cigarette smoking and physical activity) to 50% (PSS-10 and MDI among men); there were no missing values for age. The pattern of missingness for the male PSS-10 and MDI variables was not appreciably related to other study variables. Analyses were performed using SAS statistical software V.9.4.54
RESULTS
During 2013–2017, up to 4194 women contributed 2413 pregnancies and 16 656 menstrual cycles of attempt time (70.6% conceived during follow-up); among couples with complete data for both partners, 1125 contributed 717 pregnancies and 4663 menstrual cycles of pregnancy attempt time (73.0% conceived during follow-up). Approximately 12% of women and 14% of men reported marijuana use in the 2 months before baseline.
Baseline characteristics of study participants according to current marijuana use are presented in table 1. Among women, frequent marijuana use was positively associated with BMI, intake of alcohol and caffeine, perceived stress, depressive symptoms, history of STIs, intercourse frequency, active and passive smoking, residence in Canada, and having a partner who was a current marijuana user, and inversely associated with education, income and daily multivitamin use. Among men, frequent marijuana use was positively associated with intake of caffeine and sugar-sweetened soda, perceived stress, depressive symptoms, intercourse frequency, active and passive smoking, income, residence in Canada, and having a partner who was a current marijuana user, and inversely associated with education, daily multivitamin use and long weekly work hours.
Table 1.
Baseline characteristics* of 4194 female and 1125 male PRESTO participants according to current marijuana use, 2013–2017
| Characteristic† | Female marijuana use |
Male marijuana use |
||||
|---|---|---|---|---|---|---|
| None | <1 time/week | ≥1 time/week | None | <1 time/week | ≥1 time/week | |
|
| ||||||
| n (%) | 3709 (88.4) | 272 (6.5) | 213 (5.1) | 965 (85.8) | 91 (8.1) | 69 (6.1) |
| Age at baseline (mean, years) | 30.1 | 29.7 | 29.8 | 31.6 | 31.5 | 32.5 |
| Non-Hispanic White (%) | 15.3 | 20.2 | 18.5 | 12.9 | 13.2 | 12.7 |
| Household income <$50 000/year (%) | 17.9 | 15.0 | 28.2 | 16.5 | 13.1 | 22.0 |
| Household income $50 000-$99 999/year (%) | 37.3 | 43.0 | 39.8 | 35.5 | 38.5 | 42.0 |
| Household income ≥$100 000/year (%) | 44.8 | 42.0 | 32.0 | 48.0 | 48.4 | 36.0 |
| Less than college degree (%) | 21.2 | 26.4 | 42.4 | 27.3 | 27.5 | 45.6 |
| BMI (mean, kg/m2) | 26.9 | 27.1 | 28.4 | 27.7 | 27.8 | 27.4 |
| Vigorous physical activity (mean, hours/week) | 3.0 | 3.5 | 2.4 | 3.0 | 3.1 | 2.5 |
| Alcohol (mean, drinks/week) | 3.1 | 5.4 | 5.2 | 5.5 | 10.2 | 10.0 |
| Caffeine (mean, mg/day) | 116.7 | 135.3 | 148.4 | 171.6 | 193.6 | 239.2 |
| Sugar-sweetened soda intake (mean, drinks/week) | 1.2 | 1.1 | 2.1 | 2.3 | 1.7 | 6.5 |
| Daily multivitamin use (%) | 81.2 | 78.8 | 73.0 | 35.2 | 32.8 | 25.3 |
| Healthy Eating Index score‡ (mean) | 66.7 | 66.9 | 63.4 | 62.1 | 63.1 | 60.1 |
| Average sleep duration <7 hours/night (%) | 23.0 | 21.2 | 26.7 | 33.5 | 29.6 | 42.4 |
| Unemployed (%) | 14.5 | 11.2 | 18.2 | 6.2 | 6.6 | 5.4 |
| Work ≥50 hours/week (among employed) (%) | 12.5 | 10.3 | 10.5 | 31.7 | 26.0 | 24.9 |
| PSS-10 score (mean) | 15.6 | 16.1 | 17.5 | 14.5 | 15.1 | 16.3 |
| MDI score (mean) | 9.7 | 11.6 | 13.5 | 9.2 | 9.6 | 10.4 |
| History of physician-diagnosed anxiety (%) | 19.6 | 26.9 | 29.6 | 7.7 | 9.9 | 8.3 |
| Ever pregnant (women)/impregnated woman (men) (%) | 47.7 | 45.5 | 52.0 | 43.8 | 35.4 | 46.2 |
| History of sexually transmitted infections (%) | 11.1 | 21.6 | 25.0 | 4.6 | 6.6 | 7.7 |
| Doing something to improve chances of conception (%) | 75.1 | 75.6 | 74.9 | 76.6 | 69.1 | 78.8 |
| Intercourse frequency <1 time/week (%) | 20.9 | 22.2 | 18.0 | 20.0 | 24.2 | 20.9 |
| Intercourse frequency ≥4 times/week (%) | 15.4 | 14.4 | 26.0 | 14.9 | 14.1 | 21.3 |
| Hormonal last method of contraception (%) | 38.5 | 40.6 | 37.4 | 36.6 | 33.0 | 27.0 |
| Current environmental tobacco smoke exposure (%) | 6.7 | 9.2 | 21.3 | 11.8 | 9.9 | 26.1 |
| Current cigarette smoker (%) | 8.2 | 13.9 | 30.3 | 8.7 | 25.4 | 31.4 |
| Geographic region of residence (%) | ||||||
| Northeastern United States | 29.3 | 25.8 | 23.9 | 31.4 | 30.9 | 30.3 |
| Southern United States | 24.2 | 17.6 | 19.1 | 23.0 | 13.2 | 15.2 |
| Midwestern United States | 17.3 | 14.5 | 15.8 | 18.4 | 15.3 | 13.3 |
| Western United States | 14.4 | 21.9 | 20.2 | 15.5 | 18.7 | 26.1 |
| Canada | 14.8 | 20.3 | 21.0 | 11.6 | 22.0 | 15.0 |
| Partner is current marijuana user (%) | 8.7 | 62.5 | 83.8 | 3.2 | 36.1 | 57.0 |
All characteristics except for age are standardised to baseline age of cohort.
Characteristics are sex specific, except for household income, doing something to improve chances, intercourse frequency and last method of contraception (couple based).
Healthy Eating Index among men is restricted to the 166 men who completed the FFQ.
BMI, body mass index; FFQ, food frequency questionnaire; MDI, Major Depression Inventory; PRESTO, Pregnancy Study Online; PSS-10, 10-item version of Perceived Stress Scale.
Overall, we found little evidence of an association between female marijuana use and fecundability after adjusting for potential confounders (table 2). Comparing the baseline and last follow-up questionnaires completed by women, 99.1% of non-users remained non-users, 86.0% of users remained users, 0.9% of non-users became users and 14.0% of users became non-users. Time-varying exposure analyses were similar to those based on baseline exposure. Results restricted to women with shorter attempt times at study entry (<3 cycles) also indicated little evidence of an association between female marijuana use and fecundability. Results were also comparable across strata of female age (table 2) and parity (data not shown), and with further control for parity (data not shown).
Table 2.
Marijuana use in relation to couple fecundability among 4194 female pregnancy planners, PRESTO
| Number of pregnancies | Number of cycles | Unadjusted FR (95% CI) | Adjusted FR (95% CI)* | |
|---|---|---|---|---|
|
| ||||
| Full sample of women (n=4194) | ||||
| Female baseline marijuana use | ||||
| Not current user | 2168 | 14 813 | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 149 | 1033 | 0.99 (0.85 to 1.16) | 0.99 (0.85 to 1.16) |
| Current user: ≥1 time/week | 96 | 810 | 0.88 (0.72 to 1.06) | 0.98 (0.80 to 1.20) |
| Not current user | 2168 | 14 813 | 1.00 (Reference) | 1.00 (Reference) |
| Current user | 245 | 1843 | 0.95 (0.84 to 1.07) | 0.99 (0.87 to 1.12) |
| Female time-varying marijuana use | ||||
| Not current user | 2184 | 14 978 | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 132 | 866 | 1.01 (0.85 to 1.19) | 1.02 (0.87 to 1.20) |
| Current user: ≥1 time/week | 97 | 812 | 0.88 (0.73 to 1.07) | 0.99 (0.81 to 1.21) |
| Not current user | 2184 | 14 978 | 1.00 (Reference) | 1.00 (Reference) |
| Current user | 229 | 1678 | 0.95 (0.84 to 1.08) | 1.01 (0.89 to 1.15) |
| Female age <30 years (n=1985) | ||||
| Female baseline marijuana use | ||||
| Not current user | 1040 | 6728 | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 66 | 532 | 0.84 (0.66 to 1.07) | 0.85 (0.67 to 1.07) |
| Current user: ≥1 time/week | 47 | 340 | 0.99 (0.75 to 1.31) | 1.13 (0.84 to 1.52) |
| Female age ≥30 years (n=2209) | ||||
| Female baseline marijuana use | ||||
| Not current user | 1128 | 8085 | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 83 | 501 | 1.15 (0.94 to 1.41) | 1.13 (0.92 to 1.39) |
| Current user: ≥1 time/week | 49 | 470 | 0.80 (0.61 to 1.05) | 0.87 (0.66 to 1.16) |
| <3 cycles of attempt time at entry (n=2706) | ||||
| Female baseline marijuana use | ||||
| Not current user | 1537 | 9971 | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 107 | 670 | 1.02 (0.85 to 1.22) | 1.02 (0.85 to 1.23) |
| Current user: ≥1 time/week | 70 | 436 | 1.01 (0.81 to 1.27) | 1.16 (0.92 to 1.47) |
| 3–6 cycles of attempt time at entry (n=1488) | ||||
| Female baseline marijuana use | ||||
| Not current user | 631 | 4842 | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 42 | 363 | 0.91 (0.67 to 1.22) | 0.95 (0.74 to 1.21) |
| Current user: ≥1 time/week | 26 | 374 | 0.62 (0.41 to 0.92) | 0.76 (0.48 to 1.22) |
Adjusted for female age, race/ethnicity, annual household income, education, cigarette smoking history, alcohol intake, caffeine intake, BMI, vigorous physical activity, multivitamin or folic acid use, intercourse frequency, doing something to improve chances of conception, Healthy Eating Index score, average sleep duration, employment status, PSS-10 score and MDI score.
BMI, body mass index; FR, fecundability ratio; MDI, Major Depression Inventory; PRESTO, Pregnancy Study Online; PSS-10, 10-item version of Perceived Stress Scale.
Among men, current use of marijuana showed little association with fecundability overall (FR=1.01, 95% CI 0.81 to 1.27). While marijuana use at levels of ≥1 time/week was associated with a 24% increase in fecundability (95% CI 0.90 to 1.70) relative to non-use, less frequent use of marijuana was associated with a slight decrease in fecundability (FR=0.87, 95% CI 0.66 to 1.15) (table 3). When we examined partner marijuana use jointly, there was no clear pattern of association. The FR for frequent use of marijuana was slightly weaker among men who did not smoke cigarettes (FR=1.21, 95% CI 0.84 to 1.74). When the analysis was confined to couples with <3 cycles of attempt time at study entry, the results for frequent marijuana use among men were similar (FR=1.21, 95% CI 0.83 to 1.77). Stratification by frequency of intercourse (<4 vs ≥4 times/week) and doing something to improve the chances of conception (yes vs no)—a proxy for intercourse timed to the fertile window—generated similar results, except for higher fecundability observed among male frequent marijuana users with more frequent and untimed intercourse (table 4).
Table 3.
Marijuana use in relation to couple fecundability among 1125 female and male pregnancy planners, PRESTO
| Number of pregnancies | Number of cycles | Unadjusted FR (95% CI) | Adjusted FR (95% CI)* | Adjusted FR (95% CI)† | |
|---|---|---|---|---|---|
|
| |||||
| Full sample of couples (n=1125) | |||||
| Male marijuana use | |||||
| Not current user | 617 | 3992 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 55 | 442 | 0.81 (0.63 to 1.05) | 0.84 (0.64 to 1.09) | 0.87 (0.66 to 1.15) |
| Current user: ≥1 time/week | 45 | 229 | 1.22 (0.93 to 1.59) | 1.22 (0.92 to 1.62) | 1.24 (0.90 to 1.70) |
| Not current user | 617 | 3992 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user | 100 | 671 | 0.96 (0.79 to 1.16) | 0.98 (0.79 to 1.20) | 1.01 (0.81 to 1.27) |
| Marijuana use among both partners | |||||
| Neither partner is a current user | 602 | 3843 | 1.00 (Reference) | 1.00 (Reference) | – |
| Male is current user, female is not | 57 | 392 | 0.90 (0.70 to 1.16) | 0.91 (0.69 to 1.20) | – |
| Female is current user, male is not | 15 | 149 | 0.69 (0.42 to 1.13) | 0.66 (0.40 to 1.11) | – |
| Both partners are current users | 43 | 279 | 1.02 (0.77 to 1.35) | 1.02 (0.75 to 1.37) | – |
| Non-smoking men‡ (n=995) | |||||
| Male marijuana use | |||||
| Not current user | 572 | 3634 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 41 | 286 | 0.90 (0.67 to 1.20) | 0.85 (0.63 to 1.15) | 0.89 (0.65 to 1.22) |
| Current user: ≥1 time/week | 32 | 150 | 1.25 (0.91 to 1.71) | 1.21 (0.87 to 1.68) | 1.21 (0.84 to 1.74) |
| Not current user | 572 | 3634 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user | 73 | 436 | 1.03 (0.83 to 1.28) | 0.98 (0.78 to 1.23) | 1.01 (0.79 to 1.30) |
| <3 cycles of attempt time at study entry (n=772) | |||||
| Male marijuana use | |||||
| Not current user | 460 | 2757 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 41 | 322 | 0.78 (0.58 to 1.06) | 0.85 (0.62 to 1.15) | 0.88 (0.63 to 1.21) |
| Current user: ≥1 time/week | 34 | 153 | 1.22 (0.90 to 1.66) | 1.25 (0.91 to 1.72) | 1.21 (0.83 to 1.77) |
| Not current user | 460 | 2757 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user | 75 | 475 | 0.94 (0.75 to 1.17) | 0.99 (0.78 to 1.26) | 1.01 (0.77 to 1.32) |
| Marijuana use among both partners | |||||
| Neither partner is a current user | 451 | 2670 | 1.00 (Reference) | 1.00 (Reference) | – |
| Male is current user, female is not | 41 | 302 | 0.82 (0.61 to 1.10) | 0.86 (0.62 to 1.18) | – |
| Female is current user, male is not | 9 | 87 | 0.62 (0.32 to 1.19) | 0.60 (0.31 to 1.17) | – |
| Both partners are current users | 34 | 173 | 1.11 (0.82 to 1.51) | 1.06 (0.76 to 1.46) | – |
| 3–6 cycles of attempt time at study entry (n=353) | |||||
| Male marijuana use§ | |||||
| Not current user | 157 | 1235 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user: <1 time/week | 14 | 120 | 0.94 (0.58 to 1.53) | 0.80 (0.47 to 1.36) | 0.85 (0.49 to 1.49) |
| Current user: ≥1 time/week | 11 | 76 | 1.17 (0.66 to 2.05) | 1.03 (0.55 to 1.92) | 1.08 (0.57 to 2.04) |
| Not current user | 157 | 1235 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Current user | 25 | 196 | 1.03 (0.70 to 1.52) | 0.88 (0.57 to 1.37) | 0.94 (0.59 to 1.49) |
| Marijuana use among both partners¶ | |||||
| Neither partner is a current user | 151 | 1173 | 1.00 (Reference) | 1.00 (Reference) | – |
| Male is current user, female is not | 16 | 90 | 1.29 (0.82 to 2.03) | 1.01 (0.64 to 1.58) | – |
| Female is current user, male is not | 6 | 62 | 0.83 (0.39 to 1.73) | 0.86 (0.40 to 1.83) | – |
| Both partners are current users | 9 | 106 | 0.76 (0.39 to 1.48) | 0.84 (0.41 to 1.69) | – |
Adjusted for male age, race/ethnicity, education, annual household income, cigarette smoking history, alcohol intake, caffeine intake, intercourse frequency, doing something to improve chances of conception, PSS-10 score, MDI score, sugar-sweetened soda intake, average sleep duration and employment status. Couple-based models are adjusted for both male and female covariates.
Additionally adjusted for current marijuana habits of female partner.
Men who do not currently smoke cigarettes (regularly or occasionally).
Adjusted for male age, race/ethnicity, education, average sleep duration, alcohol intake, cigarette smoking history, intercourse frequency, sugar-sweetened beverage intake and annual household income.
Adjusted for female age, female race/ethnicity, female education, female cigarette smoking history, female BMI, male average sleep duration, male cigarette smoking history and male sugar-sweetened beverage intake.
FR, fecundability ratio; MDI, Major Depression Inventory; PRESTO, Pregnancy Study Online; PSS-10, 10-item version of Perceived Stress Scale.
Table 4.
Relationship between marijuana use and fecundability, stratified by frequency and timing of sexual intercourse
| Intercourse frequency |
||||||
|---|---|---|---|---|---|---|
| <4 times/week |
≥4 times/week |
|||||
| Number of pregnancies | Number of cycles | Adjusted FR (95% CI) | Number of pregnancies | Number of cycles | Adjusted FR (95% CI) | |
|
| ||||||
| Female baseline marijuana use* | ||||||
| Not current user | 1851 | 12 635 | 1.00 (Ref) | 317 | 2178 | 1.00 (Ref) |
| Current user: <1 time/week | 129 | 899 | 0.99 (0.83 to 1.17) | 20 | 134 | 0.96 (0.58 to 1.61) |
| Current user: ≥1 time/week | 71 | 605 | 0.94 (0.74 to 1.20) | 25 | 205 | 1.08 (0.71 to 1.65) |
| Male baseline marijuana use† | ||||||
| Not current user | 536 | 3386 | 1.00 (Ref) | 81 | 606 | 1.00 (Ref) |
| Current user: <1 time/week | 46 | 375 | 0.80 (0.60 to 1.07) | 9 | 67 | 1.03 (0.56 to 1.88) |
| Current user: ≥1 time/week | 33 | 200 | 1.04 (0.75 to 1.46) | 12 | 29 | 1.35 (0.72 to 2.53) |
|
| ||||||
| Doing something to improve chances of conception |
||||||
| Yes |
No |
|||||
| Number of pregnancies | Number of cycles | Adjusted FR (95% CI) | Number of pregnancies | Number of cycles | Adjusted FR (95% CI) | |
|
| ||||||
| Female baseline marijuana use* | ||||||
| Not current user | 1661 | 10 715 | 1.00 (Ref) | 507 | 4098 | 1.00 (Ref) |
| Current user: <1 time/week | 117 | 757 | 0.99 (0.83 to 1.18) | 32 | 276 | 0.95 (0.67 to 1.35) |
| Current user: ≥1 time/week | 70 | 588 | 0.99 (0.78 to 1.26) | 26 | 222 | 0.90 (0.60 to 1.36) |
| Male baseline marijuana use† | ||||||
| Not current user | 472 | 2977 | 1.00 (Ref) | 145 | 1015 | 1.00 (Ref) |
| Current user: <1 time/week | 40 | 276 | 0.87 (0.64 to 1.18) | 15 | 166 | 0.78 (0.45 to 1.38) |
| Current user: ≥1 time/week | 35 | 191 | 1.05 (0.76 to 1.45) | 10 | 38 | 2.14 (0.81 to 5.69) |
Adjusted for female age, race/ethnicity, annual household income, education, cigarette smoking history, alcohol intake, caffeine intake, BMI, vigorous physical activity, multivitamin or folic acid use, Healthy Eating Index score, average sleep duration, employment status, PSS-10 score and MDI score. Further controlled for intercourse frequency and doing something to improve chances of conception (when applicable).
Adjusted for male age, race/ethnicity, education, annual household income, cigarette smoking history, alcohol intake, caffeine intake, intercourse frequency, doing something to improve chances of conception, PSS-10 score, MDI score, sugar-sweetened soda intake, average sleep duration and employment status. Couple-based models are adjusted for both male and female covariates.
BMI, body mass index; FR, fecundability ratio; MDI, Major Depression Inventory; PSS-10, 10-item version of Perceived Stress Scale.
DISCUSSION
Among these North American couples, female and male marijuana use was neither appreciably nor consistently associated with fecundability. Among women, updating marijuana exposure over the follow-up period to account for behavioural changes with increasing TTP made little difference in the associations. Associations were also similar in analyses stratified by female age.
Comparability of our study with previous epidemiological studies is difficult because studies of marijuana use and fertility have employed different methods, such as daily exposure diaries, ovulation tests and urinary human chorionic gonadotropin concentrations. To our knowledge, ours is the first study to evaluate fecundability as an outcome measure. Although the prevalence of self-reported current marijuana use in PRESTO was slightly higher than the national average for women and men,5 under-reporting of marijuana use was likely. Evaluation of time-varying effects among men was not possible because they completed a single baseline questionnaire. In addition, our measure of marijuana captured use only within the previous 2 months, and we did not elicit data on duration or recency of use, general patterns of use (eg, daily use vs weekends; time of day typically used) or method of ingestion (inhalation (vaporising vs smoking), oral and topical). Thus, our study could not determine whether chronic use of marijuana is more important than acute use.
Pregnancy attempt times at enrolment ranged from zero to six cycles in the present analysis. For the majority of participants who enrolled immediately after discontinuing contraception (<3 cycles), marijuana use was reported before the occurrence of subfertility; thus, any misclassification was likely to be non-differential. Less fertile couples who entered the study after having tried to conceive for longer periods of time (eg, five to six cycles) may have been more likely to change their marijuana use as a result of subfertility. When we stratified by attempt time at enrolment (<3 vs ≥3 cycles), little difference was found among men. However, associations between female marijuana use and fecundability were more inverse among couples with longer attempt times at enrolment, indicating that the overall results may have been biased downward due to reverse causation.
We cannot rule out potential for residual confounding. The lack of information on indication for marijuana use (eg, to treat a medical condition or reduce stress) was an important limitation. Although loss to follow-up (20%) in PRESTO was typical for a preconception cohort study,45 55 56 marijuana users were more likely to be lost to follow-up than non-users. Bias was likely if loss to follow-up was also related to fecundability. Given that half of pregnancies in the USA are unintended, our results may not apply to couples not actively trying to conceive. Nevertheless, marijuana use did not differ markedly by intensity of trying to conceive, as indicated by the similar proportions of couples doing something to improve their chances of conception.
Internet-based recruitment has been criticised because those with and without internet access differ and, among those with internet access, those who volunteer for research studies differ from those who do not. However, internet-based recruitment should not affect the validity of the study results unless the relation between marijuana use and fecundity differed substantially between internet users and non-users, which is unlikely. Furthermore, as we and other researchers have shown, even when participation at cohort enrolment is associated with characteristics such as age, parity or smoking, measures of association are not necessarily biased due to self-selection.57 In the present analysis, selection bias was minimised by restricting to couples with shorter pregnancy attempt times at enrolment.
Strengths of the present study include enrolment during the preconception period, with >65% of couples enrolled during the first three cycles of attempted pregnancy. The study included couples residing in all US states and Canadian provinces. Finally, data were collected on a wide range of confounders, including exercise, anthropometrics and socioeconomic position (eg, education, income).
In conclusion, our study showed no clear association between female or male marijuana use and fecundability among North American pregnancy planners. The extent to which our results were attenuated by misclassification of marijuana use is unclear. Future studies with day-specific data on marijuana use might better be able to distinguish acute from chronic effects of marijuana use, and evaluate whether effects depend on other factors (eg, recency of use, phase of menstrual cycle).
What is already known on this subject
The influence of male or female marijuana use on fertility has not been well studied, and existing literature is conflicting. In women, marijuana use has been associated with alterations in reproductive hormones and menstrual cycle disturbances. Recent marijuana use has been associated with reduced oocyte retrieval and fertilisation among couples undergoing in vitro fertilisation, and an increased risk of ovulatory infertility. In men, chronic marijuana use has been associated with adverse reproductive hormone profiles and poorer semen quality, but recent studies have not confirmed these findings, and in some studies, reversible effects have been observed 5–6 weeks after initiation.
What this study adds
This large population-based cohort study contributes epidemiological data on the association between male and female marijuana use and fecundability, the average per-cycle probability of conception in a given menstrual cycle of regular unprotected intercourse. The present study found little evidence that female or male marijuana use affected fecundability.
Acknowledgements
We are grateful for the contributions of PRESTO participants and staff. We thank Michael Bairos of the Slone Epidemiology Center for his technical support in developing the web-based infrastructure of PRESTO.
Funding
Funding was provided by NICHD (R21-HD072326; R01-HD086742; R03 HD090315); the Boston University Reproductive, Perinatal and Pediatric Epidemiology Training Grant (T32-HD052458); and the Reproductive Scientist Development Program (5K12HD000849–27).
Footnotes
Competing interests None declared.
Ethics approval Study approval was obtained from Boston Medical Center Institutional Review Board.
REFERENCES
- 1.Thoma ME, McLain AC, Louis JF, et al. Prevalence of infertility in the United States as estimated by the current duration approach and a traditional constructed approach. Fertil Steril 2013;99:1324–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Macaluso M, Wright-Schnapp TJ, Chandra A, et al. A public health focus on infertility prevention, detection, and management. Fertil Steril 2010;93:16.e1–10. [DOI] [PubMed] [Google Scholar]
- 3.United Nations Office on Drugs and Crime (UNODC). World Drug Report 2015. New York, USA, 2015. [Google Scholar]
- 4.NORML. Election 2016: marijuana ballot results. 2016. http://norml.org/election-2016 (accessed 16 Mar 2017). [Google Scholar]
- 5.Hasin DS, Saha TD, Kerridge BT, et al. Prevalence of Marijuana Use Disorders in the United States Between 2001–2002 and 2012–2013. JAMA Psychiatry 2015;72:1235–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Health Canada. Canadian Alcohol and Drug Use Monitoring Survey (CADUMS). 2013. https://www.canada.ca/en/health-canada/services/canadian-tobacco-alcohol-drugs-survey/2013-summary.html (accessed 16 May 2017). [Google Scholar]
- 7.Chakravarty I, Sheth AR, Ghosh JJ. Effect of acute delta9-tetrahydrocannabinol treatment on serum luteinizing hormone and prolactin levels in adult female rats. Fertil Steril 1975;26:947–8. [DOI] [PubMed] [Google Scholar]
- 8.Dalterio SL, Mayfield DL, Bartke A. Effects of delta 9-THC on plasma hormone levels in female mice. Subst Alcohol Actions Misuse 1983;4:339–45. [PubMed] [Google Scholar]
- 9.Besch NF, Smith CG, Besch PK, et al. The effect of marihuana (delta-9-tetrahydrocannabinol) on the secretion of luteinizing hormone in the ovariectomized rhesus monkey. Am J Obstet Gynecol 1977;128:635–42. [DOI] [PubMed] [Google Scholar]
- 10.Asch RH, Smith CG, Siler-Khodr TM, et al. Effects of delta 9-tetrahydrocannabinol during the follicular phase of the rhesus monkey (Macaca mulatta). J Clin Endocrinol Metab 1981;52:50–5. [DOI] [PubMed] [Google Scholar]
- 11.Murphy LL, Tyrey L. Induction of luteinizing hormone release by electrochemical stimulation of the medial preoptic area in delta 9-tetrahydrocannabinol-blocked proestrous rats. Neuroendocrinology 1986;43:471–5. [DOI] [PubMed] [Google Scholar]
- 12.Taylor AH, Ang C, Bell SC, et al. The role of the endocannabinoid system in gametogenesis, implantation and early pregnancy. Hum Reprod Update 2007;13:501–13. [DOI] [PubMed] [Google Scholar]
- 13.Smith CG, Almirez RG, Berenberg J, et al. Tolerance develops to the disruptive effects of delta 9-tetrahydrocannabinol on primate menstrual cycle. Science 1983;219:1453–5. [DOI] [PubMed] [Google Scholar]
- 14.Brown TT, Dobs AS. Endocrine effects of marijuana. J Clin Pharmacol 2002;42:90S–6. [DOI] [PubMed] [Google Scholar]
- 15.Mendelson JH, Mello NK, Ellingboe J, et al. Marihuana smoking suppresses luteinizing hormone in women. J Pharmacol Exp Ther 1986;237:862–6. [PubMed] [Google Scholar]
- 16.Mendelson JH, Mello NK. Effects of marijuana on neuroendocrine hormones in human males and females. NIDA Res Monogr 1984;44:97–114. [PubMed] [Google Scholar]
- 17.Mendelson JH, Mello NK, Cristofaro P, et al. Acute effects of marijuana on pituitary and gonadal hormones during the periovulatory phase of the menstrual cycle. NIDA Res Monogr 1984;55:24–31. [PubMed] [Google Scholar]
- 18.Jukic AM, Weinberg CR, Baird DD, et al. Lifestyle and reproductive factors associated with follicular phase length. J Womens Health 2007;16:1340–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bauman J. Marijauna and the female reproductive system. Washington, DC: U.S. Government Printing Office, 1980. [Google Scholar]
- 20.Klonoff-Cohen HS, Natarajan L, Chen RV. A prospective study of the effects of female and male marijuana use on in vitro fertilization (IVF) and gamete intrafallopian transfer (GIFT) outcomes. Am J Obstet Gynecol 2006;194:369–76. [DOI] [PubMed] [Google Scholar]
- 21.Mueller BA, Daling JR, Weiss NS, et al. Recreational drug use and the risk of primary infertility. Epidemiology 1990;1:195–200. [DOI] [PubMed] [Google Scholar]
- 22.Dalterio S, Bartke A, Burstein S. Cannabinoids inhibit testosterone secretion by mouse testes in vitro. Science 1977;196:1472–3. [DOI] [PubMed] [Google Scholar]
- 23.Symons AM, Teale JD, Marks V. Proceedings: Effect of delta9-tetrahydrocannabinol on the hypothalamic-pituitary-gonadal system in the maturing male rat. J Endocrinol 1976;68:43P–4. [PubMed] [Google Scholar]
- 24.Smith CG, Moore CE, Besch NF, et al. The effect of marihuana (delta-9-tetrahydrocannabinol) on the secretion of sex hormones in the mature male rhesus monkey. Clin Chem 1976;22:1184. [Google Scholar]
- 25.Murphy LL, Gher J, Steger RW, et al. Effects of delta 9-tetrahydrocannabinol on copulatory behavior and neuroendocrine responses of male rats to female conspecifics. Pharmacol Biochem Behav 1994;48:1011–7. [DOI] [PubMed] [Google Scholar]
- 26.Dixit VP, Sharma VN, Lohiya NK. The effect of chronically administered cannabis extract on the testicular function of mice. Eur J Pharmacol 1974;26:111–4. [DOI] [PubMed] [Google Scholar]
- 27.Dixit VP, Gupta CL, Agrawal M. Testicular degeneration and necrosis induced by chronic administration of cannabis extract in dogs. Endokrinologie 1977;69:299–305. [PubMed] [Google Scholar]
- 28.Thompson GR, Mason MM, Rosenkrantz H, et al. Chronic oral toxicity of cannabinoids in rats. Toxicol Appl Pharmacol 1973;25:373–90. [DOI] [PubMed] [Google Scholar]
- 29.Huang HFS, Nahas GG, Hembree WC. Effects of marihuana inhalation on spermatogeneiss of the rat. In: Nahas GG, Paton W, eds. Marihuana: biological effects.. New York: Pergamon, 1979:419–27. [Google Scholar]
- 30.Zimmerman AM, Zimerman S, Raj A. Effects of cannabinoids on spermatogeneiss in mice. In: Nahas GG, Paton W, eds. Marihuana: biological effects. New York: Pergamon, 1979. [Google Scholar]
- 31.Collu R, Letarte J, Leboeuf F, et al. Endocrine effects of chronic administration of psychoactive drugs to prepuberal male rats. I: delta9-tetrahydrocannabinol. Life Sci 1975;16:533–42. [PubMed] [Google Scholar]
- 32.Rosenkrantz H, Esber HJ. Cannabinoid-induced hormone changes in monkeys and rats. J Toxicol Environ Health 1980;6:297–313. [DOI] [PubMed] [Google Scholar]
- 33.Legator MS, Weber E, Connor T, et al. Failture to detect mutagenic effects of Delta-9-tetrahydrocannabinol in the dominant lethal test host-mediated assay, blood urine studies and cytogenetic evaluation in mice. In: Braude MC, Szara S, eds. Pharmacology of Marijuana. New Yor: Raven, 1976:699–709.. [Google Scholar]
- 34.Dalterio S, Badr F, Bartke A, et al. Cannabinoids in male mice: effects on fertility and spermatogenesis. Science 1982;216:315–6. [DOI] [PubMed] [Google Scholar]
- 35.Kolodny RC, Masters WH, Kolodner RM, et al. Depression of plasma testosterone levels after chronic intensive marihuana use. N Engl J Med 1974;290:872–4. [DOI] [PubMed] [Google Scholar]
- 36.Kolodny RC, Masters WH, Kolodner RM, et al. Depression of testosterone with acute administration. In: Braude MC, Szara S, eds. Pharmacology of marijuana. New York: Raven, 1976:217–25. [Google Scholar]
- 37.Hembree WC, Zeidenberg P, Nahas GG. Marijuana’s effect on human gonadal function.In: Nahas GG, ed. Marihuana: chemistry, biochemistry, and cellular effects. New York: Springer-Verlag, 1976:52. [Google Scholar]
- 38.Gundersen TD, Jørgensen N, Andersson AM, et al. Association between use of marijuana and male reproductive hormones and semen quality: a study among 1,215 healthy young men. Am J Epidemiol 2015;182:473–81. [DOI] [PubMed] [Google Scholar]
- 39.Hembree WC, Zeidenberg P, Nahas GG. Changes in human spermatozoa associated with high dose marijuana smoking. In: Nahas GG, Paton W, eds. Marihuana: biological effects. New York: Pergamon, 1979:429–39. [DOI] [PubMed] [Google Scholar]
- 40.Issidorides MR. Observations in chronic hashish users: nuclear aberrations in blood and sperm and abnormal acrosomes in spermatozoa. In: Nahas GG, Paton W, eds. Marihuana: biological effects. New York: Pergamon, 1979:377–88. [DOI] [PubMed] [Google Scholar]
- 41.Mendelson JH, Kuehnle J, Ellingboe J, et al. Plasma testosterone levels before, during and after chronic marihuana smoking. N Engl J Med 1974;291:1051–5. [DOI] [PubMed] [Google Scholar]
- 42.Schaefer CF, Gunn CG, Dubowski KM. Letter: Normal plasma testosterone concentrations after marihuana smoking. N Engl J Med 1975;292:867–8. [DOI] [PubMed] [Google Scholar]
- 43.Mendelson JH, Ellingboe J, Kuehnle JC, et al. Effects of chronic marihuana use on integrated plasma testosterone and luteinizing hormone levels. J Pharmacol Exp Ther 1978;207:611–7. [PubMed] [Google Scholar]
- 44.Block RI, Farinpour R, Schlechte JA. Effects of chronic marijuana use on testosterone, luteinizing hormone, follicle stimulating hormone, prolactin and cortisol in men and women. Drug Alcohol Depend 1991;28:121–8. [DOI] [PubMed] [Google Scholar]
- 45.Wise LA, Rothman KJ, Mikkelsen EM, et al. Design and conduct of an internet-based preconception cohort study in north america: pregnancy study online. Paediatr Perinat Epidemiol 2015;29:360–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Subar AF, Thompson FE, Kipnis V, et al. Comparative validation of the Block, Willett, and national cancer institute food frequency questionnaires : the eating at america’s table study. Am J Epidemiol 2001;154:1089–99. [DOI] [PubMed] [Google Scholar]
- 47.Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983;24:385–96. [PubMed] [Google Scholar]
- 48.Olsen LR, Jensen DV, Noerholm V, et al. The internal and external validity of the major depression inventory in measuring severity of depressive states. Psychol Med 2003;33:351–6. [DOI] [PubMed] [Google Scholar]
- 49.Guenther PM, Kirkpatrick SI, Reedy J, et al. The healthy eating index-2010 is a valid and reliable measure of diet quality according to the 2010 dietary guidelines for Americans. J Nutr 2014;144:399–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Therneau TM, Grambsch PM. Modeling survival data: extending the cox model. New York, NY USA: Springer-Verlag, 2000. [Google Scholar]
- 51.Weinberg CR, Wilcox AJ, Baird DD. Reduced fecundability in women with prenatal exposure to cigarette smoking. Am J Epidemiol 1989;129:1072–8. [DOI] [PubMed] [Google Scholar]
- 52.Hull MG, North K, Taylor H, et al. Delayed conception and active and passive smoking. The avon longitudinal study of pregnancy and childhood study team. Fertil Steril 2000;74:725–33. [DOI] [PubMed] [Google Scholar]
- 53.Zhou XH, Eckert GJ, Tierney WM. Multiple imputation in public health research. Stat Med 2001;20:1541–9. [DOI] [PubMed] [Google Scholar]
- 54.SAS. SAS Institute Inc. 2014. SAS/STAT® 9.4 User’s Guide. Cary, NC: SAS Institute, 2014. [Google Scholar]
- 55.Lynch CD, Sundaram R, Buck Louis GM, et al. Are increased levels of self-reported psychosocial stress, anxiety, and depression associated with fecundity? Fertil Steril 2012;98:453–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Buck Louis GM, Schisterman EF, Sweeney AM, et al. Designing prospective cohort studies for assessing reproductive and developmental toxicity during sensitive windows of human reproduction and development-the LIFE Study. Paediatr Perinat Epidemiol 2011;25:413–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Hatch EE, Hahn KA, Wise LA, et al. Evaluation of selection bias in an internet-based study of pregnancy planners. Epidemiology 2016;27:98–104. [DOI] [PMC free article] [PubMed] [Google Scholar]