Abstract
Objective
To evaluate patterns of alcohol and cannabis use in pregnancy and postpartum and identify associated factors in a large, diverse cohort of people with HIV.
Methods
We analyzed data on cannabis and alcohol use from the Surveillance Monitoring for ART Toxicities study, which enrolled pregnant people with HIV at 22 United States sites. We included participants who completed a questionnaire about substance use during pregnancy and postpartum between 2007 and 2019. Postpartum cannabis and alcohol use patterns were categorized as: (1) never (in pregnancy or postpartum), (2) initiation/resumption (no use in pregnancy, any use postpartum), (3) continuation (any use in pregnancy and postpartum), or (4) discontinuation (any use in pregnancy, no use postpartum). Multivariable log‐binomial or Poisson models were fit using generalized estimating equations among pregnancies without use during pregnancy to determine factors associated with cannabis or alcohol initiation/resumption.
Results
Among 1765 pregnancies with alcohol data, 8% reported alcohol use during pregnancy and 44% postpartum. Among 1772 pregnancies with cannabis data, cannabis use during pregnancy was also less frequent (8%) than postpartum (14%). The adjusted relative risk (aRR) of postpartum alcohol initiation/resumption was lower for each additional year of age (aRR, 0.98; 95% confidence interval [CI], 0.97, 0.99), for those with less annual household income (≤$10,000 vs. >$30,000; aRR, 0.77; CI, 0.65, 0.91), and for those with less than high school education (vs. greater; aRR, 0.84; CI, 0.72, 0.99). It was greater for those from a Midwest versus Northeast location (aRR, 1.44; CI, 1.16, 1.80), for participants with a cesarean birth (aRR, 1.14; CI, 1.01, 1.30), and for those who used tobacco during pregnancy (aRR, 1.19; CI, 1.00, 1.40). Factors associated with postpartum cannabis initiation/resumption included younger age (aRR, 0.94; CI, 0.91, 0.97), lower household income (aRR, 1.90; CI, 1.06, 3.40), more recent delivery year (aRR, 1.10 per year; CI, 1.02, 1.18), psychiatric diagnosis during pregnancy (aRR, 1.71; CI, 1.02, 2.86), and tobacco use during pregnancy (aRR, 1.94; CI, 1.29, 2.94).
Conclusions
Risk factors differ for alcohol versus cannabis initiation or resumption after pregnancy among people with HIV. Understanding the clinical and obstetric characteristics associated with postpartum alcohol and/or cannabis use may enhance tailoring of interventions to optimize health among postpartum people with HIV.
Keywords: alcohol, cannabis, HIV, obstetrics, postpartum, substance use, women
1. INTRODUCTION
Alcohol and cannabis are both frequently consumed in the United States, with increasing rates of use among reproductive‐age women over the past decade [1, 2, 3]. These trends extend to pregnant people, among whom both alcohol and cannabis consumption have become more common despite established negative obstetric consequences [1, 2, 4].
Substance use warrants particular attention for pregnant people with HIV (PWH). Substance use can place pregnant PWH at higher risk of suboptimal antiretroviral therapy (ART) adherence and worsening HIV disease control [5, 6, 7, 8]. ART adherence is especially important in pregnancy and postpartum as inadequate HIV viral suppression contributes to adverse maternal and fetal health [9]. Furthermore, alcohol and cannabis consumption can compound mental health conditions, which disproportionately affect PWH due to stigma, trauma, stress, and chronic illness [10, 11, 12, 13].
Cannabis and alcohol use are common among PWH in the United States [14], but few studies focus on patterns of use during pregnancy and postpartum [2, 14, 15, 16, 17, 18, 19]. Further, conflicting results suggest that substance use trends in the nonpregnant population may not translate to the peripartum period [14, 15, 18]. The lack of data in obstetrics is particularly important given the critical opportunity to reduce perinatal HIV transmission through maternal and neonatal medication adherence, especially because the postpartum year may be a period in which birthing people have access to health care that may not exist in other life epochs. The perinatal period is a window of opportunity to intervene to promote maternal and family health.
In this analysis of data collected from participants of the Surveillance Monitoring for ART Toxicities (SMARTT) study, a component of the Pediatric HIV/AIDS Cohort Study (PHACS) network, we aimed (1) to evaluate patterns of alcohol and cannabis use from pregnancy to the postpartum period and (2) to identify factors associated with postpartum alcohol and cannabis initiation or resumption among PWH. We hypothesized that substance use would be higher postpartum compared to use during pregnancy and that antepartum co‐morbidities (e.g., mental health conditions) and pregnancy‐related events (e.g., cesarean birth) would be associated with postpartum initiation or resumption of substances.
2. MATERIALS AND METHODS
The SMARTT study was a multicenter, prospective cohort study, funded by the National Institutes of Health. The SMARTT study investigated the health of children with perinatal HIV exposure who were HIV‐negative and their caregivers through enrollment of mother–newborn pairs during pregnancy through 3 days postpartum in 22 United States sites. Participants attended study visits at enrollment and 1 year postpartum, during which they completed questionnaires and interviews, discussed below. All participants provided informed consent. The study was approved by the institutional review board at the Harvard T.H. Chan School of Public Health and each study site. This manuscript follows the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline.
Eligible for inclusion in this analysis were women with HIV who enrolled in SMARTT between 2007 and 2019 who completed an interview assessing substance use at both the delivery and 1‐year postpartum visits. The primary exposure variable was peripartum period, including pregnancy and postpartum (up to 1 year postpartum). The secondary exposure variables included demographic and clinical risk factors for postpartum initiation or resumption of substance use (defined below).
The primary outcome was self‐reported substance (alcohol or cannabis) use pattern, in which the primary pattern of interest was initiation or resumption of substance use in the postpartum period. Substance use was defined as any self‐reported use of alcohol, cannabis, or both substances. Data on antepartum substance use were collected at the SMARTT delivery visit via confidential interview. Participants were asked about their use of substances in pregnancy overall and about the frequency and trimester of their use. Postpartum substance use was assessed using the Client Diagnostic Questionnaire interview at the 1‐year postpartum visit, with a recall of the last 6 months [20]. Substance use was categorized as (1) never (no use in pregnancy or postpartum), (2) initiation or resumption (no use in pregnancy, any use postpartum), (3) continuation (any use in pregnancy and postpartum), or (4) discontinuation (any use in pregnancy, no use postpartum). Initiation and resumption of substances were considered together (and referred to as “initiation or resumption”) because data on substance use preceding pregnancy were not collected, and, therefore, it was not possible to determine if postpartum substance use was a resumption of use after cessation during pregnancy or new initiation of substance use. Hazardous drinking was defined as > 7 drinks per week or > 4 drinks at one sitting at least weekly [21, 22].
Covariates of interest in the multivariable analysis included several risk factors specified a priori including demographic characteristics (maternal age at delivery, race, ethnicity, annual household income, highest level of education, geographic region), obstetric and clinical events (trimester of earliest prenatal care, clinical chorioamnionitis, cesarean birth, any psychiatric diagnosis in pregnancy, tobacco use in pregnancy, other substance use in pregnancy), and HIV clinical factors (timing of learning HIV diagnosis [receipt of HIV diagnosis during pregnancy vs. pre‐conception] and timing of ART initiation [preconception, first trimester, second trimester, third trimester, none]). For modeling, timing of learning HIV diagnosis was combined with perinatally acquired HIV (PHIV) status as: living with PHIV, regardless of timing of learning HIV diagnosis; living with non‐PHIV (NPHIV) and learned of HIV diagnosis before pregnancy; or living with NPHIV and learned of HIV diagnosis during or after pregnancy.
Descriptive analyses included proportions to evaluate pregnancy and postpartum substance use. We described demographic, clinical, and HIV characteristics of participants by their substance use pattern. Among the subgroup of pregnancies with no reported use during pregnancy of each substance of interest, univariable log‐binomial regression models with generalized estimating equations (GEE) to account for participants with multiple pregnancies were fit to analyze associations between each characteristic of interest and postpartum substance initiation or resumption. Separate models were fit for postpartum initiation or resumption of each substance individually as well as both substances (i.e., models for any alcohol, any cannabis, both substances).
Multivariable models were then fit for each of the outcomes. Risk factors of interest were selected to identify, at the time of the delivery hospitalization (or shortly thereafter), individuals who were at elevated risk of initiation or resumption of the substance(s) of interest in the 1‐year postpartum period; therefore, factors occurring in the postpartum period were not included. Sets of risk factors were evaluated using multivariable log‐binomial models. The final multivariable model, estimated using modified Poisson regression fit with GEE, included demographic characteristics, obstetric and clinical factors, including HIV‐related factors.
A sensitivity analysis restricted to the first‐enrolled SMARTT pregnancy (i.e., excluding repeat pregnancies) was also conducted. Individuals were excluded when invalid data (e.g., illogical answers) were identified in interviews. We conducted all analyses with SAS statistical software, version 9.2 (SAS Institute Inc).
3. RESULTS
A total of 1774 pregnancies of 1429 participants met the inclusion criteria for this analysis; 1765 pregnancies had complete data on alcohol use in both periods, and 1772 pregnancies had complete data on cannabis use in both periods. The median age of women at enrollment in the SMARTT study for each pregnancy was 29 years with participants self‐identifying as Black (67%), Hispanic or Latina (28%), and White (25%). Participants generally had low annual household incomes (67% with annual income < $20,000). Most PWH had acquired HIV in adolescence or adulthood (87% non‐perinatal acquisition) and initiated ART prior to pregnancy (41%) or during the first trimester (21%). The majority (87%) of participants were virally suppressed at the end of pregnancy. Just over half of pregnancies were delivered via cesarean birth. A psychiatric diagnosis during pregnancy was reported for 4% of participants. Use of other non‐prescribed substances (cocaine, crack, sedatives or tranquilizers without medical advice, methamphetamines or other stimulants without medical advice, hallucinogens, heroin, inhalants, or methadone without medical advice) was 3% in pregnancy and 2% postpartum. Tables 1 and 2 summarize demographic and clinical characteristics of participants by pattern of pregnancy‐to‐postpartum alcohol and cannabis use, respectively.
TABLE 1.
Participant characteristics and patterns of alcohol use from pregnancy to 1 year postpartum.
| Characteristic | Total | Never use a | Continuation | Initiation/resumption | Discontinuation |
|---|---|---|---|---|---|
| Total | 1765 (100%) b | 937 (52%) | 104 (6%) | 681 (38%) | 43 (2%) |
| Demographic factors | |||||
| Age at delivery, years (median, IQR) | 29 (24, 33) | 29 (25, 34) | 29 (25, 32) | 28 (24, 32) | 27 (23, 31) |
| Race | |||||
| Asian | 9 (1%) | 5 (1%) | 3 (3%) | 1 (0%) | 0 (0%) |
| Black or African American | 1,180 (67%) | 617 (66%) | 66 (63%) | 476 (70%) | 21 (49%) |
| White | 436 (25%) | 249 (27%) | 21 (20%) | 149 (22%) | 17 (40%) |
| American Indian | 7 (0%) | 3 (0%) | 1 (1%) | 3 (0%) | 0 (0%) |
| More than one race | 25 (1%) | 10 (1%) | 3 (3%) | 12 (2%) | 0 (0%) |
| Unknown | 108 (6%) | 53 (6%) | 10 (10%) | 40 (6%) | 5 (12%) |
| Ethnicity | |||||
| Hispanic or Latina | 502 (28%) | 288 (31%) | 30 (29%) | 165 (24%) | 19 (44%) |
| Non‐Hispanic or Latina | 1,260 (71%) | 648 (69%) | 74 (71%) | 514 (75%) | 24 (56%) |
| Unknown | 3 (0%) | 1 (0%) | 0 (0%) | 2 (0%) | 0 (0%) |
| Annual household income | |||||
| ≤$10,000 | 797 (45%) | 447 (48%) | 39 (38%) | 293 (43%) | 18 (42%) |
| $10,000–$20,000 | 393 (22%) | 211 (23%) | 18 (17%) | 151 (22%) | 13 (30%) |
| $20,000–$30,000 | 205 (12%) | 105 (11%) | 13 (13%) | 86 (13%) | 1 (2%) |
| >$30,000 | 245 (14%) | 106 (11%) | 22 (21%) | 114 (17%) | 3 (7%) |
| Unknown | 125 (7%) | 68 (7%) | 12 (12%) | 37 (5%) | 8 (19%) |
| Highest level of education | |||||
| Did not receive degree diploma | 360 (20%) | 222 (24%) | 17 (16%) | 116 (17%) | 5 (12%) |
| Elementary–8th grade | 167 (9%) | 87 (9%) | 13 (13%) | 59 (9%) | 8 (19%) |
| High school | 459 (26%) | 248 (26%) | 17 (16%) | 186 (27%) | 8 (19%) |
| GED | 159 (9%) | 76 (8%) | 17 (16%) | 58 (9%) | 8 (19%) |
| Greater than high school c | 612 (35%) | 296 (34%) | 40 (39%) | 262 (38%) | 14 (33%) |
| Non‐Professional | 1 (0%) | 1 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Unknown | 7 (0%) | 7 (1%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Geographic region of clinical site | |||||
| Northeast | 401 (23%) | 239 (26%) | 13 (13%) | 141 (21%) | 8 (19%) |
| South | 738 (42%) | 379 (40%) | 40 (38%) | 301 (44%) | 18 (42%) |
| Midwest | 157 (9%) | 52 (6%) | 17 (16%) | 85 (12%) | 3 (7%) |
| West | 311 (18%) | 178 (19%) | 22 (21%) | 106 (16%) | 5 (12%) |
| Puerto Rico | 158 (9%) | 89 (9%) | 12 (12%) | 48 (7%) | 9 (21%) |
| State medical cannabis legalization status during postpartum period | |||||
| Never legal | 420 (24%) | 220 (23%) | 8 (8%) | 187 (27%) | 5 (12%) |
| Before legalization | 821 (47%) | 441 (47%) | 59 (57%) | 294 (43%) | 27 (63%) |
| After legalization | 524 (30%) | 276 (29%) | 37 (36%) | 200 (29%) | 11 (26%) |
| Obstetric and Clinical Factors | |||||
| Trimester of earliest prenatal care | |||||
| <3 months before conception (pre‐conception care) | 3 (0%) | 2 (0%) | 0 (0%) | 1 (0%) | 0 (0%) |
| First trimester | 1,107 (63%) | 578 (62%) | 59 (57%) | 444 (65%) | 26 (60%) |
| Second trimester | 361 (20%) | 186 (20%) | 23 (22%) | 145 (21%) | 7 (16%) |
| Third trimester | 43 (2%) | 32 (3%) | 1 (1%) | 9 (1%) | 1 (2%) |
| No prenatal care | 2 (0%) | 0 (0%) | 1 (1%) | 1 (0%) | 0 (0%) |
| Missing | 249 (14%) | 139 (15%) | 20 (19%) | 81 (12%) | 9 (21%) |
| Clinical chorioamnionitis | |||||
| Yes | 39 (2%) | 20 (2%) | 4 (4%) | 13 (2%) | 2 (5%) |
| No | 1,706 (97%) | 904 (96%) | 99 (95%) | 662 (97%) | 41 (95%) |
| Unknown | 20 (1%) | 13 (1%) | 1 (1%) | 6 (1%) | 0 (0%) |
| Cesarean birth | |||||
| Yes | 1,012 (57%) | 520 (55%) | 61 (59%) | 403 (59%) | 28 (65%) |
| No | 747 (42%) | 413 (44%) | 42 (40%) | 277 (41%) | 15 (35%) |
| Missing | 6 (0%) | 4 (0%) | 1 (1%) | 1 (0%) | 0 (0%) |
| Any psychiatric diagnosis in pregnancy | |||||
| Yes | 70 (4%) | 39 (4%) | 5 (5%) | 22 (3%) | 4 (9%) |
| No | 1,695 (96%) | 898 (96%) | 99 (95%) | 659 (97%) | 39 (91%) |
| Tobacco use | |||||
| Yes | 314 (18%) | 129 (14%) | 46 (44%) | 121 (18%) | 18 (42%) |
| No | 1,451 (82%) | 808 (86%) | 58 (56%) | 560 (82%) | 25 (58%) |
| Unknown | 2 (0%) | – | – | – | – |
| Other substance use d | |||||
| Yes | 53 (3%) | 20 (2%) | 11 (11%) | 13 (2%) | 9 (21%) |
| No | 1,702 (96%) | 911 (97%) | 91 (88%) | 666 (98%) | 34 (79%) |
| Unknown | 10 (1%) | 6 (1%) | 2 (2%) | 2 (0%) | 0 (0%) |
| HIV Factors | |||||
| Timing of HIV acquisition | |||||
| Non‐perinatally acquired HIV | 1,538 (87%) | 838 (89%) | 87 (84%) | 576 (85%) | 37 (86%) |
| Perinatally‐acquired HIV | 188 (11%) | 82 (9%) | 12 (12%) | 90 (13%) | 4 (9%) |
| Unknown/missing | 39 (2%) | 17 (2%) | 5 (5%) | 15 (2%) | 2 (5%) |
| Antiretroviral therapy use during pregnancy | |||||
| Yes | 1,716 (97%) | 915 (98%) | 98 (94%) | 662 (97%) | 41 (95%) |
| No | 22 (1%) | 8 (1%) | 3 (3%) | 9 (1%) | 2 (5%) |
| Missing | 27 (2%) | 14 (1%) | 3 (3%) | 10 (1%) | 0 (0%) |
| Timing of antiretroviral therapy initiation | |||||
| Before/at conception | 731 (41%) | 404 (43%) | 45 (43%) | 267 (39%) | 15 (35%) |
| First trimester | 367 (21%) | 173 (18%) | 21 (20%) | 163 (24%) | 10 (23%) |
| Second trimester | 528 (30%) | 281 (30%) | 25 (24%) | 208 (31%) | 14 (33%) |
| Third trimester | 90 (5%) | 57 (6%) | 7 (7%) | 24 (4%) | 2 (5%) |
| No antiretroviral therapy use | 22 (1%) | 8 (1%) | 3 (3%) | 9 (1%) | 2 (5%) |
| Unknown | 27 (2%) | 14 (1%) | 3 (3%) | 10 (1%) | 0 (0%) |
| Timing of learning HIV diagnosis | |||||
| During pregnancy | 261 (15%) | 148 (16%) | 13 (13%) | 92 (14%) | 8 (19%) |
| <1 year before conception | 122 (7%) | 57 (6%) | 4 (4%) | 58 (9%) | 3 (7%) |
| 1 to < 5 years before conception | 431 (24%) | 229 (24%) | 34 (33%) | 164 (24%) | 4 (9%) |
| 5 to < 10 years before conception | 394 (22%) | 212 (23%) | 19 (18%) | 152 (22%) | 11 (26%) |
| ≥10 years before conception | 460 (26%) | 241 (26%) | 26 (25%) | 181 (27%) | 12 (28%) |
| Missing | 97 (5%) | 50 (5%) | 8 (8%) | 34 (5%) | 5 (12%) |
Substance use was categorized as (1) never (no use in pregnancy or postpartum), (2) continuation (use in pregnancy and postpartum), (3) initiation or resumption (no use in pregnancy, use in postpartum period), or (4) discontinuation (use in pregnancy, no use postpartum).
Nine pregnancies with unknown alcohol use, but known cannabis use data. Values do not sum to 100% in each case due to rounding.
Including trade or technical training, some college or 2‐year degree, college graduate, graduate school or professional school.
Other substance use postpartum collected via the Client Diagnostic Questionnaire (CDQ): cocaine, crack, sedatives or tranquilizers without medical advice, methamphetamines, and other stimulants without medical advice, hallucinogens, heroin, inhalants, or methadone without medical advice.
TABLE 2.
Participant characteristics and patterns of cannabis use from pregnancy to 1 year postpartum.
| Characteristic | Total | Never use a | Continuation |
Initiation/ resumption |
Discontinuation |
|---|---|---|---|---|---|
| Total | 1772 (100%) b | 1471 (83%) | 89 (5%) | 152 (9%) | 60 (3%) |
| Demographic factors | |||||
|
Age at delivery, years (median, IQR) |
29 (24, 33) | 29 (25, 34) | 28 (24, 30) | 26 (23, 30.5) | 27.5 (24, 32.5) |
| Race | |||||
| Asian | 9 (1%) | 8 (1%) | 1 (1%) | 0 (0%) | 0 (0%) |
| Black or African American | 1183 (67%) | 954 (65%) | 67 (75%) | 120 (79%) | 42 (70%) |
| White | 438 (25%) | 384 (26%) | 18 (20%) | 25 (16%) | 11 (18%) |
| American Indian | 7 (0%) | 6 (0%) | 0 (0%) | 0 (0%) | 1 (2%) |
| More than one race | 26 (1%) | 21 (1%) | 0 (0%) | 4 (3%) | 1 (2%) |
| Unknown | 109 (6%) | 98 (7%) | 3 (3%) | 3 (2%) | 5 (8%) |
| Ethnicity | |||||
| Hispanic or Latina | 505 (28%) | 455 (31%) | 13 (15%) | 24 (16%) | 13 (22%) |
| Non‐Hispanic or Latina | 1264 (71%) | 1,013 (69%) | 76 (85%) | 128 (84%) | 47 (78%) |
| Unknown | 3 (0%) | 3 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Annual household income | |||||
| ≤$10,000 | 800 (45%) | 649 (44%) | 37 (42%) | 79 (52%) | 35 (58%) |
| $10,000‐$20,000 | 394 (22%) | 329 (22%) | 19 (21%) | 33 (22%) | 13 (22%) |
| $20,000‐$30,000 | 205 (12%) | 172 (12%) | 12 (13%) | 15 (10%) | 6 (10%) |
| >$30,000 | 247 (14%) | 225 (15%) | 8 (9%) | 13 (9%) | 1 (2%) |
| Unknown | 126 (7%) | 96 (7%) | 13 (15%) | 12 (8%) | 5 (8%) |
| Highest level of education | |||||
| Did not receive degree diploma | 361 (20%) | 292 (20%) | 27 (30%) | 29 (19%) | 13 (22%) |
| Elementary‐8th grade | 168 (9%) | 135 (9%) | 7 (8%) | 14 (9%) | 12 (20%) |
| High school | 462 (26%) | 387 (26%) | 13 (15%) | 48 (32%) | 14 (23%) |
| GED | 160 (9%) | 117 (8%) | 12 (13%) | 24 (16%) | 7 (12%) |
| Greater than high school c | 613 (35%) | 532 (35%) | 30 (34%) | 37 (25%) | 14 (23%) |
| Non‐Professional | 1 (0%) | 1 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Unknown | 7 (0%) | 7 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Geographic region of clinical site | |||||
| Northeast | 403 (23%) | 343 (23%) | 7 (8%) | 35 (23%) | 18 (30%) |
| South | 741 (42%) | 603 (41%) | 45 (51%) | 71 (47%) | 22 (37%) |
| Midwest | 157 (9%) | 114 (8%) | 13 (15%) | 23 (15%) | 7 (12%) |
| West | 313 (18%) | 263 (18%) | 20 (22%) | 21 (14%) | 9 (15%) |
| Puerto Rico | 158 (9%) | 148 (10%) | 4 (4%) | 2 (1%) | 4 (7%) |
| State medical cannabis legalization status during postpartum period | |||||
| Never legal | 423 (24%) | 354 (24%) | 23 (26%) | 38 (25%) | 8 (13%) |
| Before legalization | 822 (46%) | 692 (47%) | 32 (36%) | 59 (39%) | 39 (65%) |
| After legalization | 527 (30%) | 425 (29%) | 34 (38%) | 55 (36%) | 13 (22%) |
| Obstetric and Clinical Factors | |||||
| Trimester of earliest prenatal care | |||||
| <3 months before conception (pre‐conception care) | 4 (0%) | 4 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| First trimester | 1,112 (63%) | 918 (62%) | 57 (64%) | 98 (64%) | 39 (65%) |
| Second trimester | 361 (20%) | 297 (20%) | 20 (22%) | 32 (21%) | 12 (20%) |
| Third trimester | 43 (2%) | 33 (2%) | 4 (4%) | 4 (3%) | 2 (3%) |
| No prenatal care | 2 (0%) | 1 (0%) | 0 (0%) | 1 (1%) | 0 (0%) |
| Missing | 250 (14%) | 218 (15%) | 8 (9%) | 17 (11%) | 7 (12%) |
| Clinical chorioamnionitis | |||||
| Yes | 39 (2%) | 27 (2%) | 4 (4%) | 7 (5%) | 1 (2%) |
| No | 1,713 (97%) | 1,428 (97%) | 84 (94%) | 143 (94%) | 58 (97%) |
| Unknown | 20 (1%) | 16 (1%) | 1 (1%) | 2 (1%) | 1 (2%) |
| Cesarean birth | |||||
| Yes | 1,015 (57%) | 844 (57%) | 47 (53%) | 92 (61%) | 32 (53%) |
| No | 751 (42%) | 622 (42%) | 42 (47%) | 59 (39%) | 28 (47%) |
| Missing | 6 (0%) | 5 (0%) | 0 (0%) | 1 (1%) | 0 (0%) |
| Any psychiatric diagnosis in pregnancy | |||||
| Yes | 70 (4%) | 45 (3%) | 8 (9%) | 13 (9%) | 4 (7%) |
| No | 1,702 (96%) | 1,426 (97%) | 81 (91%) | 139 (91%) | 56 (93%) |
| Tobacco use | |||||
| Yes | 315 (18%) | 192 (13%) | 51 (57%) | 30 (20%) | 42 (70%) |
| No | 1457 (82%) | 1,279 (87%) | 38 (43%) | 122 (80%) | 18 (30%) |
| Unknown | 0 (0%) | – | – | – | – |
| Other substance use d | |||||
| Yes | 53 (3%) | 34 (2%) | 8 (9%) | 1 (1%) | 10 (17%) |
| No | 1,709 (96%) | 1,429 (97%) | 81 (91%) | 150 (99%) | 49 (82%) |
| Unknown | 10 (1%) | 8 (1%) | 0 (0%) | 1 (1%) | 1 (2%) |
| HIV Factors | |||||
| Timing of HIV acquisition | |||||
| Non‐perinatally acquired HIV | 1,543 (87%) | 1,296 (88%) | 79 (89%) | 120 (79%) | 48 (80%) |
| Perinatally‐acquired HIV | 190 (11%) | 142 (10%) | 7 (8%) | 30 (20%) | 11 (18%) |
| Unknown/missing | 39 (2%) | 33 (2%) | 3 (3%) | 2 (1%) | 1 (2%) |
| Antiretroviral therapy use during pregnancy | |||||
| Yes | 1,723 (97%) | 1,431 (97%) | 87 (98%) | 145 (95%) | 60 (100%) |
| No | 22 (1%) | 18 (1%) | 1 (1%) | 3 (2%) | 0 (0%) |
| Missing | 27 (2%) | 22 (1%) | 1 (1%) | 4 (3%) | 0 (0%) |
| Timing of antiretroviral therapy initiation | |||||
| Before/at conception | 735 (41%) | 618 (42%) | 27 (30%) | 66 (43%) | 24 (40%) |
| First trimester | 368 (21%) | 301 (20%) | 25 (28%) | 29 (19%) | 13 (22%) |
| Second trimester | 529 (30%) | 433 (29%) | 27 (30%) | 48 (32%) | 21 (35%) |
| Third trimester | 91 (5%) | 79 (5%) | 8 (9%) | 2 (1%) | 2 (3%) |
| No antiretroviral therapy use | 22 (1%) | 18 (1%) | 1 (1%) | 3 (2%) | 0 (0%) |
| Unknown | 27 (2%) | 22 (1%) | 1 (1%) | 4 (3%) | 0 (0%) |
| Timing of learning HIV diagnosis | |||||
| During pregnancy | 261 (15%) | 217 (15%) | 17 (19%) | 20 (13%) | 7 (12%) |
| <1 year before conception | 122 (7%) | 99 (7%) | 7 (8%) | 10 (7%) | 6 (10%) |
| 1 to < 5 years before conception | 431 (24%) | 363 (25%) | 20 (22%) | 35 (23%) | 13 (22%) |
| 5 to < 10 years before conception | 397 (22%) | 332 (23%) | 20 (22%) | 33 (22%) | 12 (20%) |
| ≥ 10 years before conception | 464 (26%) | 380 (26%) | 20 (22%) | 46 (30%) | 18 (30%) |
| Missing | 97 (5%) | 80 (5%) | 5 (6%) | 8 (5%) | 4 (7%) |
Substance use was categorized as (1) never (no use in pregnancy or postpartum), (2) continuation (use in pregnancy and postpartum), (3) initiation or resumption (no use in pregnancy, use in postpartum period), or (4) discontinuation (use in pregnancy, no use postpartum).
Two pregnancies with unknown cannabis use but known alcohol use data.
Including trade or technical training, some college or 2‐year degree, college graduate, graduate school or professional school.
Other substance use postpartum collected via the Client Diagnostic Questionnaire (CDQ): cocaine, crack, sedatives or tranquilizers without medical advice, methamphetamines and other stimulants without medical advice, hallucinogens, heroin, inhalants, or methadone without medical advice.
3.1. Alcohol
Among 1765 pregnancies (excluding 9/1774 individuals with unknown alcohol use), 8% reported alcohol use during pregnancy and 44% reported alcohol use postpartum (Table 1). From pregnancy to the postpartum period, 6% of PWH reported continued alcohol consumption, 2% reported alcohol discontinuation in pregnancy without postpartum resumption, 38% reported initiation or resumption of alcohol postpartum, and 52% reported no use of alcohol in pregnancy or within the first year postpartum. Of those who consumed alcohol during pregnancy, most reported drinking alcohol in the first trimester (8%), but did not continue into the second (2%) or third (1%; Figure 1). Hazardous drinking was reported among 118 (7%) participants, predominantly during the postpartum period.
FIGURE 1.
Timing and frequency of alcohol and cannabis use during pregnancy among individuals with use in pregnancy.
Individuals with alcohol use during pregnancy used tobacco and other substances more frequently during pregnancy. The presence of a psychiatric diagnosis in pregnancy was more common in pregnancies of participants with alcohol discontinuation. Participants with postpartum alcohol initiation or resumption were younger, more frequently non‐Hispanic or Latina, had higher annual income, with more years of completed education, and located in the Midwest and South as compared to those who never used alcohol (Table 1). Postpartum initiation or resumption of alcohol occurred more frequently for individuals with PHIV.
In univariable models among pregnancies without use of alcohol during pregnancy, non‐Hispanic or Latina ethnicity, higher income, higher education, Midwest relative to Northeast location, and persons with PHIV were associated with postpartum alcohol initiation or resumption (Supporting Information Appendix).
In the final multivariable model, younger age (adjusted relative risk [aRR], 0.98; 95% confidence interval [CI], 0.97, 0.99), higher annual income (>$30,000 referent, ≤$10,000; aRR, 0.77; CI, 0.65, 0.91), higher education (greater than high school education referent, less than high school education; aRR, 0.84; CI 0.72, 0.99), Midwest versus Northeast location (aRR, 1.44; CI, 1.16, 1.80), cesarean birth (aRR, 1.14; CI, 1.01, 1.30), and tobacco use during pregnancy (aRR, 1.19; CI, 1.00, 1.40) were associated with increased risk of postpartum alcohol initiation or resumption (Table 3). No significant associations were observed for other risk factors. Sensitivity analyses demonstrated similar findings.
TABLE 3.
Factors associated with initiation or resumption of alcohol, cannabis, or both during the postpartum period among pregnancies without use during pregnancy a .
| Model outcome | |||
|---|---|---|---|
| Category |
Alcohol initiation or resumption (analysis population N = 1618) Adjusted relative risk (95% confidence interval) |
Cannabis initiation or resumption (analysis population N = 1623) |
Both alcohol and cannabis initiation or resumption (analysis population N = 1721) |
| Demographic Factors | |||
| Age at delivery, years | 0.98 (0.97, 0.99) | 0.94 (0.91, 0.97) | 0.95 (0.92, 0.99) |
| Race | |||
| White | 1.00 | 1.00 | 1.00 |
| Black or African American | 1.07 (0.85, 1.34) | 1.14 (0.61, 2.13) | 0.95 (0.53, 1.71) |
| Other or unknown b | 1.19 (0.78, 1.81) | 0.71 (0.21, 2.38) | 0.51 (0.12, 2.10) |
| Ethnicity | |||
| Hispanic or Latina | 0.97 (0.76, 1.25) | 0.70 (0.25, 1.38) | 0.57 (0.28, 1.19) |
| Annual household income | |||
| >$30,000 | 1.00 | 1.00 | 1.00 |
| $20,000–$30,000 | 0.88 (0.72, 1.08) | 1.34 (0.64, 2.79) | 1.36 (0.72, 2.58) |
| $10,000–$20,000 | 0.85 (0.71, 1.02) | 1.64 (0.89, 3.00) | 1.47 (0.84, 2.58) |
| ≤$10,000 | 0.77 (0.65, 0.91) | 1.90 (1.06, 3.40) | 1.60 (0.94, 2.71) |
| Education at delivery | |||
| Less than high school | 0.84 (0.72, 0.99) | 0.81 (0.55, 1.21) | 0.83 (0.56, 1.23) |
| Geographic region of clinical site | |||
| Northeast | 1.00 | 1.00 | 1.00 |
| Midwest | 1.44 (1.16, 1.80) | 1.54 (0.85, 2.79) | 1.80 (0.96, 3.38) |
| Puerto Rico | 0.83 (0.55, 1.27) | 0.19 (0.02, 1.49) | 0.30 (0.04, 2.34) |
| South | 1.04 (0.82, 1.33) | 1.08 (0.60, 1.95) | 1.51 (0.81, 2.80) |
| West | 0.95 (0.73, 1.25) | 0.66 (0.34, 1.29) | 1.03 (0.52, 2.04) |
| State medical cannabis legalization status during postpartum period d | |||
| Before legalization | 1.00 | 1.00 | 1.00 |
| Never legal | 1.09 (0.89, 1.33) | 0.86 (0.52, 1.43) | 0.75 (0.45, 1.24) |
| After legalization | 1.05 (0.85, 1.31) | 1.46 (0.83, 2.57) | 1.35 (0.76, 2.41) |
| Year of delivery | 1.02 (0.99, 1.04) | 1.10 (1.02, 1.18) | 1.09 (1.02, 1.17) |
| Obstetric Factors | |||
| Trimester of earliest prenatal care | |||
| First trimester | 1.00 | 1.00 | 1.00 |
| Second trimester | 1.04 (0.90, 1.21) | 1.00 (0.67, 1.48) | 0.83 (0.55, 1.25) |
| Third trimester or none | 0.66 (0.41, 1.08) | 1.08 (0.38, 3.08) | 1.05 (0.40, 2.79) |
| Missing c | 0.90 (0.72, 1.14) | 0.69 (0.30, 1.54) | 0.37 (0.14, 0.97) |
| Clinical chorioamnionitis | 0.71 (0.37, 1.38) | 1.94 (0.90, 4.22) | 1.42 (0.67, 2.99) |
| Cesarean birth | 1.14 (1.01, 1.30) | 1.17 (0.84, 1.64) | 1.06 (0.77, 1.46) |
| Any psychiatric diagnosis during pregnancy | 0.83 (0.56, 1.22) | 1.71 (1.02, 2.86) | 1.54 (0.92, 2.59) |
| Tobacco use in pregnancy | 1.19 (1.00, 1.40) | 1.94 (1.29, 2.94) | 2.24 (1.56, 3.23) |
| Other substance use in pregnancy e | 1.07 (0.69, 1.66) | 0.31 (0.06, 1.51) | 0.59 (0.20, 1.74) |
| HIV Factors | |||
| Timing of ART initiation f | |||
| Before/at Conception | 1.00 | 1.00 | 1.00 |
| First Trimester | 1.17 (0.99, 1.38) | 0.86 (0.56, 1.32) | 0.89 (0.60, 1.30) |
| After First Trimester | 1.12 (0.93, 1.34) | 0.88 (0.56, 1.38) | 0.95 (0.61, 1.46) |
| Timing of learning HIV diagnosis g | |||
| NPHIV, learned of diagnosis prior to pregnancy h | 1.00 | 1.00 | 1.00 |
| NPHIV, learned of diagnosis during pregnancy | 0.87 (0.69, 1.08) | 0.90 (0.50, 1.62) | 0.81 (0.46, 1.45) |
| PHIV, regardless of timing of learning HIV diagnosis | 1.03 (0.84, 1.28) | 1.14 (0.73, 1.79) | 1.31 (0.82, 2.08) |
Multivariable models include all variables in the table. Estimates modified Poisson model fit with GEE. Referent is people with never use of the substance of interest.
Other included Asian, American Indian, and more than one race.
Missing treated as valid category.
Before legalization was selected as the referent because it contained the largest total number of participants and represents the baseline policy status prior to legislation to legalize cannabis.
Other substance use postpartum collected via the Client Diagnostic Questionnaire (CDQ): cocaine, crack, sedatives or tranquilizers without medical advice, methamphetamines and other stimulants without medical advice, hallucinogens, heroin, inhalants, or methadone without medical advice.
ART (antiretroviral therapy).
Receipt of HIV diagnosis during pregnancy (vs. pre‐conception). For modeling, this was combined with PHIV status as: living with PHIV, regardless of timing of learning HIV diagnosis; living with NPHIV and learned of HIV diagnosis prior to pregnancy; living with NPHIV and learned of HIV diagnosis during or after pregnancy.
PHIV (perinatally acquired HIV), NPHIV (non‐perinatally acquired HIV). “During pregnancy” also includes pregnancies where the person learned of her HIV diagnosis within 3 days of delivery.
3.2. Cannabis
Cannabis use during pregnancy was less frequent (8%) than postpartum (14%). During pregnancy, cannabis use followed a similar, but less pronounced, trend as alcohol use, with cannabis use more frequently reported early versus later in pregnancy (first trimester 7%; second 4%; third 3%). Eighty‐three percent of participants never used cannabis (Table 2). For cannabis, 5% reported continued cannabis use, 3% reported cannabis discontinuation, and 9% reported cannabis initiation or resumption.
Similar to alcohol initiation or resumption, PWH with postpartum cannabis initiation or resumption were more likely to be younger and non‐Hispanic or Latina ethnicity. In contrast to alcohol use patterns, those with postpartum cannabis initiation or resumption had lower income, fewer completed years of education, and were more commonly in the Northeast and South (Table 2).
Among pregnancies without use of cannabis during pregnancy, in the unadjusted model, younger age, Black race, non‐Hispanic or Latina ethnicity, Midwest relative to Northeast location, later delivery year, chorioamnionitis, a psychiatric diagnosis in pregnancy, postpartum polysubstance use, and PHIV were all associated with postpartum cannabis initiation or resumption in the univariable models (Supporting Information Appendix).
In the final multivariable model (Table 3), younger age (with each additional year of age there was a 6% decrease in the relative risk: aRR, 0.94; CI, 0.91, 0.97), lower household income (>$30,000 referent, ≤$10,000; aRR, 1.90; CI, 1.06, 3.40), later delivery year (aRR, 1.10 per year later; CI, 1.02, 1.18), psychiatric diagnosis in pregnancy (aRR, 1.71; CI, 1.02, 2.86), and tobacco use during pregnancy (aRR, 1.94; CI, 1.29, 2.94) were associated with an increased risk of postpartum cannabis initiation or resumption. In the sensitivity analysis excluding repeat pregnancies, findings were similar.
3.3. Both alcohol and cannabis
Among pregnancies with no use of alcohol and cannabis during pregnancy, initiation or resumption of both alcohol and cannabis was more common among individuals who used tobacco or other substances during pregnancy. Postpartum alcohol and cannabis initiation or resumption was observed more frequently among participants without ART use at conception. Younger age, non‐Hispanic or Latina ethnicity, Midwest and Northeast region, later years of the study period, a psychiatric diagnosis in pregnancy, tobacco use, use of other substances postpartum, chorioamnionitis, and PHIV were also associated with postpartum use of both cannabis and alcohol in univariable models (Supporting Information Appendix).
In the final multivariable model (Table 3), tobacco use during pregnancy (aRR, 2.24; CI, 1.56, 3.23), younger age (aRR, 0.95; CI, 0.92, 0.99), and later delivery year (aRR, 1.09 per year later; CI, 1.02, 1.17) were associated with increased risk of postpartum use of both alcohol and cannabis. Sensitivity analyses excluding repeat pregnancies demonstrated similar findings.
4. DISCUSSION
Alcohol and cannabis use in this cohort of PWH was common during the postpartum period, with alcohol initiation or resumption exceeding that of cannabis in the first postpartum year. While there were some overlapping risk factors, alcohol and cannabis had distinct patterns of consumption and associated risk factors for postpartum initiation or resumption. Risk factors for postpartum alcohol (e.g., cesarean birth, higher education, Midwest location) versus cannabis (e.g., psychiatric diagnosis, later delivery year) initiation or resumption were different. Younger age and tobacco use were associated with postpartum initiation or resumption for both substances. Sensitivity analyses excluding individuals with repeat pregnancies in the study period showed similar findings.
The clinical importance of substance use during pregnancy is well‐established, yet less is known about postpartum substance use for PWH [2, 14, 15, 16, 17, 18, 19]. Few previous studies examined substance use among PWH in any life stage or PWH in the peripartum period in the United States [2, 14, 15, 16, 18, 19, 23]. Compared to nonpregnant PWH, we found lower rates of both cannabis and alcohol use during pregnancy and postpartum [14, 24, 25]. A previous study of PWH in SMARTT from 2007 to 2009 found a higher rate of antepartum alcohol use (10.2%) and a lower rate of cannabis use (7.3%) [19]. In this analysis, the SMARTT cohort had lower rates of substance use in pregnancy than peripartum PWH in studies from previous decades [2, 14, 15, 16, 17, 18, 19]. This change over time in substance use may correlate with a population shift in PWH from women who primarily acquired HIV through injection‐substance use, which is associated with use of other substances, to those who primarily acquired HIV through heterosexual contact [17]. The epidemiologic evolution underscores the importance of examining substance use among a contemporary cohort of PWH.
The increase in cannabis use between 2007 and 2019 may reflect a shift in cannabis use in the general population as cannabis became legal in several states [16]. One previous study from 2010 to 2011 found that PWH were more likely than matched pregnant people without HIV to use tobacco, alcohol, and cannabis during pregnancy and more likely to have comorbid mental health conditions [18]. Our findings suggest this gap may be closing, with the population in our study demonstrating a lower overall rate of alcohol consumption during pregnancy than the general obstetric population (13.5% [26]) and a similar rate of cannabis use (7.1% in one California study [27] and 7% in another national study [28]). The deviations of patterns among PWH from patterns seen among similar populations without HIV support the importance of continuing to investigate peripartum substance use among PWH. A nuanced understanding of both alcohol and cannabis consumption in the peripartum period among PWH can inform policies, resources, and interventions to reduce disparities and help mitigate the unique vulnerabilities associated with substance use among pregnant PWH.
The increase in alcohol use postpartum follows anticipated trends, as alcohol is a broadly understood teratogen with subsequent avoidance in the antepartum period. Furthermore, public health messaging around alcohol avoidance in pregnancy is now deeply ingrained in US culture. In contrast, as data on antepartum and postpartum cannabis effects on the maternal–child dyad are much newer, this public health messaging is less ubiquitous. Still, alcohol and cannabis use patterns postpartum may merit particular attention for PWH because of the unique vulnerabilities of the postpartum period for this population [29, 30]. The postpartum period represents a unique and clinically important transition, characterized by shifts in physiologic state, mental health, health care engagement, and social support. In this context, substance use patterns may reflect not only individual risk factors but also structural and care‐related dynamics, including loss of pregnancy‐associated behavioral modification, reduced frequency of clinical contact, and competing psychosocial stressors. From a clinical perspective, these findings reinforce the postpartum period as a critical window for targeted screening, counseling, and intervention. Notably, during the years of the study, breastfeeding was not recommended for PWH due to concerns about transmission of HIV to the neonate [31]. As substance use is often discouraged when breastfeeding, exclusive formula feeding may be associated with postpartum substance use behaviors for the study cohort. Infant feeding recommendations have evolved in recent years in the United States to provide options for breastfeeding while minimizing risks of transmission [31], however, any associated changes in postpartum substance use among PWH remain areas for further research.
The literature documents decreased ART adherence and care engagement in the postpartum period for PWH [29]. Use of both alcohol and cannabis has been associated with inconsistencies in ART adherence [8]. The initiation of substance use in the postpartum period could contribute to more irregular ART schedules at a time when PWH are managing ART for both themselves and antiretroviral prophylaxis for their newborn. This challenge could be further compounded for PWH who breastfeed as alcohol can decrease milk supply and make exclusive breastfeeding, which is medically imperative for PWH who feed infants their own milk, more difficult.
PWH are at increased risk compared to those without HIV for postpartum mental health conditions [30]. Psychiatric comorbidity emerged as a strong predictor of postpartum cannabis use, though the overall prevalence of diagnosed psychiatric conditions in this cohort was low. This may represent underdiagnosis rather than true absence of disease, particularly given the well‐documented burden of depression, anxiety, and trauma among people with HIV. These findings underscore the importance of comprehensive mental health screening during pregnancy and the postpartum period, as well as integration of behavioral health services into perinatal care. Both cannabis and alcohol can contribute to symptoms of depression and anxiety and can serve as maladaptive coping mechanisms for individuals with mood disorders. Like persons with HIV, individuals who engage in alcohol and cannabis use in the postpartum period are subject to stigma, which may decrease their likelihood of disclosing their substance use to their support system or medical providers [32]. This combination of risk factors at a vulnerable period accentuates the value of maintaining rapport between provider and PWH in order to optimize chances of acceptance of mental health resources and medication support when indicated.
The findings demonstrate several trends that may be beneficial in allocating resources. Alcohol use was more common in the Midwest. Younger age, tobacco use, and pregnancy during more recent years were associated with cannabis use and use of both substances. These criteria could inform screening questions or patient‐centered dialogues on the risks of substance use. Education may be needed about the harms of cannabis use during and after pregnancy, particularly following the legalization of cannabis and subsequent misinformation regarding the “safety” and “therapeutic benefits” of cannabis in pregnancy perpetuated by the cannabis industry [33]. Unlike alcohol, cannabis use was more similar during pregnancy and postpartum among PWH, which may reflect a misunderstanding of the risks of cannabis to fetal development.
Strengths of this study include its use of contemporary data from a large, diverse population of PWH who experienced pregnancy and postpartum across the United States. Its use of longitudinal data on substance use patterns is a novel addition to the scant literature on this topic. However, this study has limitations. The study did not differentiate between prescription and recreational cannabis. A key limitation of this study is the reliance on self‐reported substance use, which is particularly vulnerable to underreporting during pregnancy due to social desirability bias and perceived stigma. Substance use during pregnancy was assessed retrospectively at delivery, while postpartum use was assessed at one year with recall of the prior six months, introducing both recall bias and potential differential misclassification between time periods. As such, the observed increase in substance use from pregnancy to postpartum may in part reflect differences in reporting rather than true behavioral change. However, previous studies have demonstrated that self‐reporting regarding substance use among PWH correlates with meconium substance testing [19]. We were unable to distinguish between true initiation of substance use postpartum and resumption following abstinence during pregnancy, as prepregnancy substance use was not collected. This distinction is clinically meaningful, as many individuals may discontinue use during pregnancy and subsequently resume postpartum. Accordingly, the outcome of “initiation or resumption” may reflect a return to prepregnancy behaviors rather than new uptake. The findings from this study may not be generalizable to other substances or to all pregnant and postpartum PWH; the individuals in the SMARTT study represent those with regular engagement with health care providers, both for themselves and their child, with willingness to participate in research. Particularly given the regional trends elicited in this study, efforts should be made to investigate other populations of PWH in future work [16, 17, 19].
5. CONCLUSION
This study examined alcohol and cannabis use patterns among pregnant and postpartum PWH. Substance use patterns differ for alcohol and cannabis use in the perinatal period among PWH. The postpartum period represents a uniquely vulnerable time during the life course of PWH, and substance use can exacerbate postpartum challenges such as the emergence of viremia, risk of infant HIV acquisition through breastfeeding, and perinatal mental health conditions including postpartum depression. Public health initiatives like the extension of postpartum Medicaid open a window of opportunity to elevate women's health and reduce risk for lifelong chronic conditions such as substance use disorder. Our findings provide the foundation for a nuanced approach to supporting PWH with peripartum substance use and can be applied to develop public policy and interventions for this population.
CONFLICT OF INTEREST STATEMENT
J.J. receives funding for a Gilead study unrelated to this manuscript. All other authors declare no conflicts of interest.
ETHICS STATEMENT
The study was approved by the institutional review board at the Harvard T.H. Chan School of Public Health and each study site.
Supporting information
Supporting Information
ACKNOWLEDGMENTS
We thank the participants and families for their participation in PHACS, and the institutions involved in the conduct of the PHACS SMARTT study. A complete list of the institutions, clinical site investigators and staff which participated can be found at https://phacsstudy.org/publications/smartt-acknowledgements/. The study was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD), Office of the Director, National Institutes of Health (OD), National Institute of Dental & Craniofacial Research (NIDCR), National Institute of Allergy and Infectious Diseases (NIAID), National Institute of Neurological Disorders and Stroke (NINDS), National Institute on Deafness and Other Communication Disorders (NIDCD), National Institute of Mental Health (NIMH), National Institute on Drug Abuse (NIDA), National Cancer Institute (NCI), National Institute on Alcohol Abuse and Alcoholism (NIAAA), and National Heart, Lung, and Blood Institute (NHLBI) through grants to the Harvard T.H. Chan School of Public Health (P01HD103133, Principal Investigators: Ellen Chadwick, Sonia Hernandez‐Diaz, Jennifer Jao, Paige Williams; Program Director: Liz Salomon and HD052102: Principal Investigator: George R Seage III; Program Director: Liz Salomon) and with Tulane University School of Medicine (HD052104) (Principal Investigators: Russell Van Dyke, Ellen Chadwick; Project Director: Patrick Davis). The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
This abstract was presented by Dr. Lynn Yee as a virtual poster at the Society for Maternal‐Fetal Medicine 42nd Annual Meeting in February 2022.
DATA AVAILABILITY STATEMENT
All SMARTT data were submitted to NICHD DASH as public use datasets. They are not available currently due to DASH transition to a new platform, but should be under this link: https://dash.nichd.nih.gov/study/227210
REFERENCES
- 1. Brown, Q. L. , Sarvet A. L., Shmulewitz D., Martins S. S., Wall M. M., and Hasin D. S.. 2017. “Trends in Marijuana Use among Pregnant and Nonpregnant Reproductive‐Aged Women, 2002–2014.” JAMA 317(2): 207–9. 10.1001/jama.2016.17383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Singh, S. , Filion K. B., Abenhaim H. A., and Eisenberg M. J.. 2020. “Prevalence and Outcomes of Prenatal Recreational Cannabis Use in High‐Income Countries: A Scoping Review.” BJOG 127(1): 8–16. 10.1111/1471-0528.15946. [DOI] [PubMed] [Google Scholar]
- 3. Hasin, D. S. , Shmulewitz D., and Keyes K.. 2019. “Alcohol Use and Binge Drinking among U.S. Men, Pregnant and Non‐Pregnant Women Ages 18–44: 2002–2017.” Drug and Alcohol Dependence 205: 107590. 10.1016/J.DRUGALCDEP.2019.107590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. England, L. J. , Bennett C., Denny C. H., Honein M. A., Gilboa S. M., Kim S. Y., Guy G. P., et al. 2020. “Alcohol Use and Co‐Use of Other Substances among Pregnant Females Aged 12–44 Years—United States, 2015–2018.” MMWR Morbidity and Mortality Weekly Report 69(31): 1009–14. 10.15585/mmwr.mm6931a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Chander, G. , Himelhoch S., and Moore R. D.. 2006. “Substance Abuse and Psychiatric Disorders in HIV‐Positive Patients: Epidemiology and Impact on Antiretroviral Therapy.” Drugs 66(6): 769–89. 10.2165/00003495-200666060-00004. [DOI] [PubMed] [Google Scholar]
- 6. Neblett, R. C. , Hutton H. E., Lau B., McCaul M. E., Moore R. D., and Chander G.. 2011. “Alcohol Consumption among HIV‐Infected Women: Impact on Time to Antiretroviral Therapy and Survival.” Journal of Women's Health 20(2): 279–86. 10.1089/JWH.2010.2043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Petry, N. M. 1999. “Alcohol Use in HIV Patients: What We Don't Know May Hurt Us.” International Journal of STD and AIDS 10(9): 561–70. 10.1258/0956462991914654. [DOI] [PubMed] [Google Scholar]
- 8. Springer, S. A. , Dushaj A., and Azar M. M.. 2012. “The Impact of DSM‐IV Mental Disorders on Adherence to Combination Antiretroviral Therapy Among Adult Persons Living with HIV/AIDS: A Systematic Review.” AIDS and Behavior 16(8): 2119–43. 10.1007/s10461-012-0212-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bailey, H. , Zash R., Rasi V., and Thorne C.. 2018. “HIV Treatment in Pregnancy.” Lancet HIV 5(8): e457–e67. 10.1016/S2352-3018(18)30059-6. [DOI] [PubMed] [Google Scholar]
- 10. Dyer, T. P. , Stein J. A., Rice E., and Rotheram‐Borus M. J.. 2012. “Predicting Depression in Mothers With and without HIV: The Role of Social Support and Family Dynamics.” AIDS and Behavior 16(8): 2198–208. 10.1007/S10461-012-0149-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Leenerts, M. H. , 1999. “The Disconnected Self: Consequences of Abuse in a Cohort of Low‐Income White Women Living With HIV/AIDS.” Health Care for Women International 20(4): 381–400. 10.1080/073993399245674. [DOI] [PubMed] [Google Scholar]
- 12. Machtinger, E. L. , Wilson T. C., Haberer J. E., and Weiss D. S.. 2012. “Psychological Trauma and PTSD in HIV‐Positive Women: A Meta‐Analysis.” AIDS and Behavior 16(8): 2091–100. 10.1007/S10461-011-0127-4. [DOI] [PubMed] [Google Scholar]
- 13. Wyatt, G. E. , Myers H. F., and Loeb T. B.. 2004. “Women, Trauma, and HIV: An Overview.” AIDS and Behavior 8(4): 401–3. 10.1007/S10461-004-7324-3. [DOI] [PubMed] [Google Scholar]
- 14. D'Souza, G. , Matson P. A., Grady C. D., Nahvi S., Merenstein D., Weber K. M., Greenblatt R., Burian P., and Wilson T. E.. 2012. “Medicinal and Recreational Marijuana Use Among HIV‐Infected Women in the Women's Interagency HIV Study (WIHS) Cohort, 1994–2010.” Journal of Acquired Immune Deficiency Syndromes 61(5): 618–26. 10.1097/QAI.0b013e318273ab3a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Yee, L. M. , Powis K. M., Jao J., Haddad L. B., Yao T., Barr E. A., Siminski S., Broadwell C., Chadwick E. G., and Kacanek D.. 2024. “Substance Use During Pregnancy and Postpartum Among Individuals With Perinatally‐Acquired HIV in the United States.” AIDS 38(15): 2103–7. 10.1097/QAD.0000000000004031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Yee, L. M. , Kacanek D., Brightwell C., Haddad L. B., Jao J., Powis K. M., Yao T., et al. 2021. “Marijuana, Opioid, and Alcohol Use Among Pregnant and Postpartum Individuals Living With HIV in the US + Invited Commentary + Supplemental Content.” JAMA Network Open 4(12): 2137162. 10.1001/jamanetworkopen.2021.37162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Rough, K. , Tassiopoulos K., Kacanek D., Griner R., Yogev R., Rich K. C., and Seage G. R. 3rd, & Pediatric HIVAIDS Cohort Study . 2015. “Dramatic Decline in Substance Use by HIV‐Infected Pregnant Women in the United States from 1990 to 2012.” AIDS 29(1): 117–23. 10.1097/QAD.0000000000000503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Malee, K. M. , Mellins C. A., Huo Y., Tassiopoulos K., Smith R., Sirois P. A., Allison S. M., et al. 2014. “Prevalence, Incidence, and Persistence of Psychiatric and Substance Use Disorders among Mothers Living with HIV.” Journal of Acquired Immune Deficiency Syndromes 65: 526–34. 10.1097/QAI.0000000000000070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tassiopoulos, K. , Read J. S., Brogly S., Rich K., Lester B., Spector S. A., Yogev R., and Seage G. R., 2010. “Substance Use in HIV‐Infected Women during Pregnancy: Self‐Report versus Meconium Analysis.” AIDS and Behavior 14(6): 1269–78. 10.1007/s10461-010-9705-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Aidala, A. , Havens J., Mellins C., Dodds S., Whetten K., Martin D., Gillis L., and Ko P.. 2004. “Development and Validation of the Client Diagnostic Questionnaire (CDQ): A Mental Health Screening Tool for Use in HIV/AIDS Service Settings.” Psychology, Health and Medicine 9(3): 362–80. 10.1080/13548500410001721927. [DOI] [Google Scholar]
- 21. 2005. Helping Patients Who Drink Too Much: A Clinician's Guide. National Institute on Alcohol Abuse and Alcoholism, Bethesda, Maryland; U.S. Department of Health & Human Services, National Institutes of Health. [Google Scholar]
- 22. Cook, R. L. , Zhu F., Belnap B. H., Weber K., Cook J. A., Vlahov D., Wilson T. E., et al. 2009. “Longitudinal Trends in Hazardous Alcohol Consumption among Women with Human Immunodeficiency Virus Infection, 1995‐2006.” American Journal of Epidemiology 169(8): 1025–32. 10.1093/aje/kwp004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Ko, J. Y. , Tong V. T., Bombard J. M., Hayes D. K., Davy J., and Perham‐Hester K. A.. 2018. “Marijuana Use During and After Pregnancy and Association of Prenatal Use on Birth Outcomes: A Population‐Based Study.” Drug and Alcohol Dependence 187: 72–78. 10.1016/j.drugalcdep.2018.02.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Haley, D. F. , Bullington B. W., Tien P., Knittel A. K., Bobitt J., Kempf M. C., Philbin M., et al. 2025. “Patterns of Cannabis Use among Women with HIV in the United States.” AIDS and Behavior 29(6): 2022–2032. 10.1007/S10461-025-04669-Z/TABLES/3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Kelso‐Chichetto, N. E. , Plankey M., Abraham A. G., Ennis N., Chen X., Bolan R., and Cook R. L.. 2018. “Association Between Alcohol Consumption Trajectories and Clinical Profiles among Women and Men Living with HIV.” American Journal of Drug and Alcohol Abuse 44(1): 85–94. 10.1080/00952990.2017.1335317;PAGE:STRING:ARTICLE/CHAPTER. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Gosdin, L. K. , Deputy N. P., Kim S. Y., Dang E. P., and Denny C. H.. 2022. “Alcohol Consumption and Binge Drinking during Pregnancy among Adults Aged 18–49 Years—United States, 2018–2020.” MMWR Morbidity and Mortality Weekly Report 71(1): 10–13. 10.15585/mmwr.mm7101a2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Young‐Wolff, K. C. , Slama N. E., Avalos L. A., Padon A. A., Silver L. D., Adams S. R., Does M. B., et al. 2024. “Cannabis Use During Early Pregnancy Following Recreational Cannabis Legalization.” JAMA Health Forum 5(11): e243656. 10.1001/JAMAHEALTHFORUM.2024.3656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Thompson, R. , Dejong K., and Lo J.. 2019. “Marijuana Use in Pregnancy: A Review.” Obstetrical & Gynecological Survey 74(7): 415. 10.1097/OGX.0000000000000685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Momplaisir, F. M. , Storm D. S., Nkwihoreze H., Jayeola O., and Jemmott J. B.. 2018. “Improving Postpartum Retention in Care for Women Living with HIV in the United States.” AIDS 32(2): 133–42. 10.1097/QAD.0000000000001707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Kapetanovic, S. , Dass‐Brailsford P., Nora D., and Talisman N.. 2014. “Mental Health of HIV‐Seropositive Women During Pregnancy and Postpartum Period: A Comprehensive Literature Review.” AIDS and Behavior 18(6): 1152–73. 10.1007/S10461-014-0728-9/FIGURES/1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Powell, A. , and Agwu A.. 2024. “In Support of Breast‐/Chestfeeding by People With HIV in High‐Income Settings.” Clinical Infectious Diseases 79(1): 202–7. 10.1093/CID/CIAE027. [DOI] [PubMed] [Google Scholar]
- 32. Weber, A. , Miskle B., Lynch A., Arndt S., and Acion L.. 2021. “Substance Use in Pregnancy: Identifying Stigma and Improving Care.” Substance Abuse and Rehabilitation 12: 105–121. 10.2147/SAR.S319180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Dickson, B. , Mansfield C., Guiahi M., Allshouse A. A., Borgelt L. M., Sheeder J., Silver R. M., and Metz T. D.. 2018. “Recommendations from Cannabis Dispensaries about First‐Trimester Cannabis Use.” Obstetrics and Gynecology 131(6): 1031–8. 10.1097/AOG.0000000000002619. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Data Availability Statement
All SMARTT data were submitted to NICHD DASH as public use datasets. They are not available currently due to DASH transition to a new platform, but should be under this link: https://dash.nichd.nih.gov/study/227210