Racial/Ethnic Differences in Prenatal Supplement and Medication Use in Low-Risk Pregnant Women

Abstract

Objective

This study aimed to describe the overall quantity and type of supplements and medications used during pregnancy in a low-risk cohort and to examine any racial/ ethnic differences in intake.

Study Design

We used data from 2,164 racially/ethnically diverse, nonobese, and low-risk pregnant women participating without pre-pregnancy chronic conditions in a prospective cohort study at 12 sites across the United States. Medication data were self-reported as free text in enrollment, follow-up visit questionnaires, and abstracted from medical records at delivery. Supplements and medications data were mapped to their active ingredients and categorized into corresponding classes using the Slone Drug Dictionary. The total number and classes of supplements and medications consumed during pregnancy were calculated. Modified Poisson regression models were used to estimate the racial/ethnic differences in supplements and medications intake. All models were adjusted for maternal sociodemographic factors and study site.

Results

98% of women took at least one supplement during pregnancy, with prenatal vitamins/multivitamins being most common. While only 31% reported taking no medications during pregnancy, 23% took one, 18% took two, and 28% took three or more. The percentage of women taking at least one medication during pregnancy was highest among non-Hispanic white women and lowest among Asians (84 vs. 55%, p < 0.001). All racial/ethnic groups reported taking the same top four medication classes including central nervous system agents, gastrointestinal drugs, anti-infective agents, and antihistamines. Compared with non-Hispanic white women, Hispanic (adjusted relative risk [aRR]: 0.84, 95% confidence interval [CI]: 0.71–0.98), and Asian women (aRR: 0.83, 95% CI: 0.70–0.98) were less likely to take central nervous system agents, as well as gastrointestinal drugs (Hispanics aRR: 0.79, 95% CI: 0.66–0.94; Asians aRR = 0.75, 95% CI: 0.63–0.90), and antihistamines (Hispanics aRR: 0.65, 95% CI: 0.47–0.92).

Conclusion

Supplement intake was nearly universal. Medication use was also common among this low-risk pregnancy cohort and differed by race/ethnicity.

ClinicalTrials.gov Identifier

NCT00912132.

The safety of many medications for use during pregnancy is unknown.^1,2 Additionally, physiologic changes during pregnancy alter medications’ pharmacokinetics³ with unknown consequences. Nevertheless, medication consumption during pregnancy is common.^4–7 In fact, prenatal medication use has more than doubled in the past four decades.⁶ Similarly, consumption of dietary supplements is prevalent among pregnant women.⁴ Prenatal vitamins, analgesics, antibiotics, and antiemetics are the most common types of supplements and medications used during pregnancy.^4,8–10 Despite the high prevalence of use, maternal factors associated with supplement and medication consumption are not well defined. Being married, higher levels of education and annual income greater than $50,000 have been shown to be associated with higher supplement use (e.g., multivitamins and folic acid) during pregnancy.^4,11 Similarly, older age at the time of pregnancy and increasing prevalence of chronic conditions requiring medication treatment,¹² as well as smoking, drug, and alcohol consumption during pregnancy,¹³ may in part explain the greater use of medications during pregnancy. However, the association of maternal race/ethnicity with prenatal supplement and medication use is less clear which is important given the racial/ethnic disparities in chronic conditions requiring medications.¹⁴

Previous observational studies have reported on maternal race/ethnicity associated with preconception and prenatal patterns of supplement and medication intake.^{6,10,12,13,15–17} Overall, these studies found higher medication intake (prescription and/or nonprescription) and dietary supplement use in non-Hispanic white women compared with other racial/ethnic groups in the month before or during pregnancy.^{6,7,10,12,15–19} However, many studies were based on retrospective design with the recall of supplement/medication use after knowledge of birth outcomes, whereas our study is strengthened by its prospective data collection and longitudinal design. The Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies-Singleton cohort is ideal for investigating race/ethnic differences in prenatal supplement and medication use because it is a well-characterized, racially/ethnically diverse, and low-risk cohort of pregnant women without pre-pregnancy chronic conditions, which reduces some of the issues with potential confounding that exist in prior literature. Using data from this cohort, we first examined the quantity and types of supplements and medications used at different time points across pregnancy and then investigated the potential racial/ethnic differences in supplement and medication intake during pregnancy.

Materials and Methods

The NICHD Fetal Growth Studies-Singletons cohort (2009–2013) was a multisite and prospective study with the goal of establishing standards for fetal growth.²⁰ The study’s original exclusion and inclusion criteria were established to allow for the recruitment of women who were nonsmokers with low-risk antenatal profiles and without pre-pregnancy chronic conditions. A full list of these conditions can be found in the cohort profile.²¹ Hence, the cohort comprised of 2,334 pregnant women from four self-reported racial/ethnic backgrounds who had a pre-pregnancy body mass index of 19 to <30 kg/m². Details of the study can be found elsewhere.²¹ Women were enrolled between 8^0/7 and 13^6/7 weeks of gestation, with a screening ultrasound scan performed at 10 to 13 weeks to confirm gestational age. At enrollment, detailed demographic and health information were obtained from participants through in-person interviews by trained research nurses. Women were interviewed in-person for up to five follow-up visits at 16 to 22, 24 to 29, 30 to 33, 34 to 37, and 38 to 41 weeks of gestation. Human subjects’ approval was obtained from all participating sites, the NICHD, and the data coordinating center. All participating women also provided informed consent prior to data collection.

For the current analysis, women were excluded if they were determined to be ineligible after enrollment (n = 15), lost to follow-up (n = 29), refused to continue the study (n = 86), moved (n = 26), or did not have medical records (n = 13) or were missing medication information (n = 1). After exclusion, the analytic sample included 2,164 women.

The outcomes of interest were the quantity and types of supplements and medications (prescription and over the counter) consumed at any time during pregnancy and by study visit. During the in-person enrollment interview women were asked, “since you became pregnant, have you used or taken any medications and nonprescription vitamins, minerals, herbals, and supplements?” At each follow-up in-person visit women were asked “since your last visit with us, have you used or taken any medications and nonprescription vitamins, minerals, herbals, and supplements?” If participants answered yes, they were asked to provide the free text list of all their supplements and medications. Information on prenatal medication intake was also abstracted from women’s medical records after delivery by research personnel.

The Slone Drug Dictionary was used with the Coding Engine program to link free text data to the active ingredients or the specific components of the brand and generic names for every reported supplement and medication.^22–25 Supplements and medications were then categorized into their corresponding classes as defined by the Slone Drug Dictionary.

The information for the main exposure, self-reported maternal race/ethnicity was recorded at enrollment and categorized into four groups: non-Hispanic white (NHW), non-Hispanic black (NHB), Hispanic, and Asian/Pacific Islanders (Asian/PIs).

Covariates in this analysis were maternal characteristics including age, pre-pregnancy weight, height, parity (nulliparous, multiparous), marital status (married/living with a partner, unmarried), current job status (at least one job/full-time student and no job), health insurance (private/managed care, Medicaid/self-pay), education (less than high school, high school, some college/associate degree, and bachelor’s degree or higher), annual family income (less than $30,000, $30,000– $39,999, $40,000–$49,999, $50,000–$74,999, $75,000– $99,999, $100,000, or more), and study site.

Statistical Analysis

Descriptive statistics were used to compare baseline maternal characteristics among women from the four racial/ethnic groups using Chi-square tests, Fisher’s exact test, or one-way analysis of variance for categorical and continuous variables, respectively.

We computed the percentage of women taking supplements and medications at any point during pregnancy and by study visit. We used modified Poisson regression models²⁶ to estimate risk ratios (RR) with 95% confidence intervals (CI) using robust error variance for the association between race/ ethnicity and total number of supplements and medications consumed during pregnancy, and to estimate the racial/ ethnic differences in the percentage of women taking the top four medication classes. We also fit adjusted regression models with generalized estimating equations under Poisson distribution to estimate the racial/ethnic differences in risk for taking specific supplement and medication classes during pregnancy using NHW, as the reference group and to account for multiple observations from each participant in the study. An interaction term between race/ethnicity and median gestational week at each of the six visits (including enrollment) was used to determine the differences in race/ethnicity in the consumption of medication classes. A statistically significant (p < 0.05) interaction between race/ethnicity and time is interpreted, as the intake of medication classes among NHB, Hispanic, or Asian/PI women differs from that in NHW women at each designated time point during pregnancy. Finally, we calculated the percent of women taking different types of supplements and medications in the top four supplement and medication classes, for any time during pregnancy and by each study visit. Since the supplement and medication classes specified in the Slone Dictionary are broad, we provide a detailed list of specific supplements and medications in each of the top four classes in Supplementary Tables S1 and S2 (available in the online version), respectively. We grouped prenatal vitamins with and without docosahexaenoic acid (DHA) with multivitamins, as a separate category within their designated class of supplements.

For the regression analyses, we used multivariate imputation via the chained equation using MICE package in R to impute the missing information. Visits that were missed because the participant did not attend the visit were considered intermittent missingness and imputed. In contrast, visits missed due to pregnancy termination or delivery prior to the expected visit date were not considered as missing and were not included in the imputation set. The primary analysis of this dataset was completed in 2018. All analyses were conducted using SAS version 9.4 or R version 3.4.4.

Results

Among 2,164 women, there was nearly equal distribution of women in each racial/ethnic group except for fewer Asian/PI women (Table 1). Overall, the average age of women at enrollment was 28 ± 5.5 years, and the majority of women were married. Compared with NHB and Hispanic women, NHW and Asian/PI women were more likely to be older, nulliparous and married, to have reached a higher educational level, and to have higher incomes (p-values < 0.0001). Only 3.5% of women developed gestational diabetes and 5.8% any hypertensive disorder of pregnancy. Hispanic and NHB women were more likely to develop these conditions than NHW women (Table 1).

Table 1.

Maternal characteristics at enrollment by race/ethnicity, National Institute of Child Health and Human Development Fetal Growth Studies-Singletons^a

Maternal characteristics	Overall	Non-Hispanic White	Non-Hispanic Black	Hispanic	Asian/Pacific Islanders
Sample size, % (n)	n = 2,164	26 (571)	26 (568)	28 (611)	19 (414)
Age, mean ± SD (y)	28.0 ± 5.5	30 ± 4.5	25 ± 5.4	27 ± 5.5	31 ± 4.5
Pre-pregnancy weight, Mean ± SD (kg)	63 ± 9.5	64 ± 9.3	65 ± 9.9	62 ± 8.5	57 ± 8.5
Self-reported height, Mean ± SD (cm)	163 ± 7.1	166 ± 7.1	164 ± 6.8	160 ± 6.5	160 ± 6.0
Parity (%)
Nulliparous	48.8	54.6	50.7	38.6	53.1
Marital status (%)
Married/living with a partner	75.5	93.9	47.9	73.0	91.6
Currently paid job (%)
At least one job/full-time student	71.2	83.2	73.8	60.1	67.4
Health insurance (%)
Private/managed care	57.2	93.5	44.0	32.2	61.8
Medicaid/other	39.7	5.3	52.8	62.2	35.8
Self-pay	2.3	1.1	1.8	4.3	1.7
Unknown	0.9	0.2	1.4	1.3	0.7
Annual family income (%)
<$30,000	24.1	3.8	41.6	34.4	12.8
$30,000–$39,999	7.8	3.1	7.6	13.9	5.6
$40,000–$49,999	6.8	4.0	10.2	8.2	4.1
$50,000–$74,999	10.5	12.3	8.3	10.8	10.9
$75,000–$99,999	11.6	18.0	8.8	5.7	15.0
$100,000 or more	25.2	55.2	10.4	9.2	27.8
Missing	14	3.5	13.2	17.8	23.9
Hypertensive disorders of pregnancyb (%)
Yes	5.8	5.3	9.9	5.1	1.9
Gestational diabetes mellitusc (%)
Yes	3.5	2.8	1.8	3.9	6.0

Abbreviations: GED, general educational development; OGTT, oral glucose tolerance test; SD, standard deviation.

^aDifferences and the corresponding p-values were obtained using Chi-square test of independence for parity, maternal marital status, employment, health insurance, education, last year income, and diagnosis of gestational diabetes and hypertension and one-way ANOVA were used to obtain the p-values for the mean differences for maternal age, pre-pregnancy weight, and self-reported height. All p-values <0.01.

^bHypertensive disorders of pregnancy included a discharge diagnosis of severe or mild preeclampsia, severe or mild gestational hypertension, or unknown hypertension as recorded in the medical record.

^cGestational diabetes mellitus was defined using a combination of OGTT results and the discharged diagnosis of gestational diabetes in the medical records. Women with OGTT results or an indication of gestational diabetes in the medical records were considered as developing gestational diabetes during pregnancy.

Nearly all women (98%) reported taking at least one supplement. The most commonly reported supplement was prenatal vitamins (with and without DHA)/multivitamins (97%), followed by iron (32%) and essential fatty acids (12%) (Supplementary Table S1 [available in the online version]). Most women (77%) reported taking one to two supplements, 17% took three to four, and 7% took five or more (Table 2). There were no significant differences in the proportion of women taking supplements during pregnancy by racial/ethnic groups (p > 0.05) (Fig. 1).

Table 2.

Number of dietary supplements taken by women at any point during pregnancy and by study visit, National Institute of Child Health and Human Development Fetal Growth Studies-Singletons

	Enrollment	Visit 1	Visit 2	Visit 3	Visit 4	Visit 5	Overall
Sample size, n	2,164	2,109	2,075	2,047	1,910	999	2,164
Gestational weeks, median (range)	13 (10–14)	20 (15–28)	28 (22–32)	32 (28–36)	36 (33–40)	39 (37–41)
Supplement quantity (%)
0	10.8	11.1	8.9	7.5	7.6	8.1	2.2
1	66.3	68.7	66.2	60.4	59.5	57.4	43.1
2	17.8	15.8	19.6	25.4	26.2	27.6	34.6
3	3.4	3.0	3.7	4.7	4.8	5.0	12.6
4	1.0	1.0	1.0	1.3	1.5	1.1	4.3
5	0.5	0.2	0.5	0.5	0.3	0.5	1.7
6	0.1	0.05	0.05	0.05	0.05	0.2	1.0
7		0.1	0.05	0.05	0.05	0.1	0.3
8+		0.05	0.10	0.05	0.1		0.3

Fig. 1.

Percentage of women taking 0, 1, 2, or 3 or more dietary supplements during pregnancy by race/ethnicity, National Institute of Child Health and Human Development Fetal Growth Studies Singletons. Modified Poisson regression model was used to find differences in the number of supplements used by race/ethnicity. Non-Hispanic white was the reference group. Adjusted differences were not statistically significant (p > 0.05). Regression models were adjusted for maternal age, pre-pregnancy weight, self-reported height, parity, marital status, current employment status, health insurance status, educational level, income level, and study site. NHB, non-Hispanic black; NHW, non-Hispanic white.

The majority of women (69%) reported taking at least one nonsupplemental medication at any point during pregnancy. The most commonly reported medication was acetaminophen (Tylenol) (39%), followed by antacids (26%) and anticonstipation medications (11%) (Supplementary Table S2 [available in the online version]). Overall, 23% of women took one, 18% took one, and 28% took three or more medications (Table 3). Fig. 2 illustrates the percentage of women taking medications overall and by race/ethnicity. Compared with 28% of NHW women, only 17% of Hispanic women and 16% of Asian/PI women reported taking three or more medications during pregnancy. Compared with NHW women, Hispanic (aRR: 0.85, 95% CI: 0.74–0.97), and Asian/PI (aRR: 0.86, 95% CI: 0.75–0.99) women, but not NHB women, were less likely to consume more medications.

Table 3.

Number of medications (prescription and over the counter) taken by women at any time during pregnancy and by study visit, National Institute of Child Health and Human Development Fetal Growth Studies-Singletons

	Enrollment	Visit 1	Visit 2	Visit 3	Visit 4	Visit 5	Overall
Sample size, n	2,164	2,109	2,075	2,047	1,910	999	2,164
Gestational week, median (range)	13 (10–14)	20 (15–28)	28 (22–32)	32 (28–36)	36 (33–40)	39 (37–41)
Medication quantity (%)
0	60.1	60.2	62.3	63.7	63.1	65.4	30.8
1	25.5	25.1	24.9	22.5	22.5	22.9	22.8
2	9.3	9.7	8.6	8.3	9.8	7.3	17.9
3	3.0	3.2	2.9	3.6	2.3	3.2	11.7
4	1.3	1.2	0.8	1.4	1.6	0.8	6.5
5	0.4	0.4	0.5	0.2	0.4	0.3	4.3
6	0.3	0.1	0.05	0.1	0.3	0.1	2.5
7	0.05		0.10	0.1			1.6
8+				0.2			1.8

Fig. 2.

Percentage of women taking 0, 1, 2, or 3 or more medications during pregnancy by race/ethnicity, National Institute of Child Health and Human Development Fetal Growth Studies Singletons. Modified Poisson regression model was used to find differences in the number of medications used by race/ethnicity. Non-Hispanic white, reference group. *p-value < 0.05; p-values indicates that lower percentage of Hispanic and Asian women took more medications compared with non-Hispanic white women. Regression models were adjusted for maternal age, pre-pregnancy weight, self-reported height, parity, marital status, current employment status, health insurance status, educational level, income level, and study site. NHB, non-Hispanic black; NHW, non-Hispanic white.

A total of three classes of dietary supplements (Supplementary Table S1 [available in the online version]) and 17 classes of medications were identified (Supplementary Table S3 [available in the online version]). The percentages of women using any supplements within each of the three supplement classes are presented in Supplementary Table S1 (available in the online version). The topmost frequently reported class of supplements was “vitamins” (98%) with prenatal vitamins (with and without DHA)/multivitamins (97%) being the single most commonly reported category in this class. When examining the intake of this category (i.e., prenatal vitamins (with and without DHA)/multivitamins) over time, we found a significant interaction between race/ethnicity and median gestational week (p < 0.05). In the stratified analysis, at enrollment compared with Asian women, a lower proportion of the other three racial/ethnic groups (NHW, NHB, and Hispanic women) reported taking prenatal vitamins (with and without DHA)/multivitamins (aRR: 0.93, 95% CI: 0.90–0.97).

The top four medication classes included central nervous system (CNS) agents (43%), gastrointestinal (GI) drugs (37%), anti-infective agents (20%), and antihistamines (17%). The percentages of specific medications in the top four medication classes consumed at each study visit are presented in Supplementary Table S2 [available in the online version]). The CNS agents included painkillers, nonsteroidal anti-inflammatory drugs, and antidepressants and other mood disorder medications with acetaminophen being the single most commonly used (39%). The GI drug class included antacids/proton pump inhibitors, anticonstipation medications, antiemetics, antidiarrhea medications, and other GI-directed medications with antacids being most common (27%). Anti-infective agents were comprised of different types of antibiotics, antiviral and antifungal medications with unspecified antibiotics (6%), and azithromycin (3%) most common. Antihistamines class included different types of antihistamines, allergy medications with decongestants, and cold medications containing antihistamines with diphenhydramine (7%) being reported most frequently. Fig. 3 illustrates the percentage of women taking the top four medication classes during pregnancy by race/ethnicity. There were statistically significant racial/ ethnic differences in the intake of each medication class during pregnancy. In general, a higher percentage of NHW women took these medication classes compared with women of other racial/ethnic groups. Specifically, only 8% of Hispanic women (aRR: 0.65, 95% CI: 0.47–0.91) reported taking antihistamines compared with 26% of NHW women. Similarly, 31% of Hispanic women (aRR: 0.84, 95% CI: 0.71– 0.98) and 28% of Asian/PI women (aRR: 0.83, 95% CI: 0.70– 0.98) reported taking medications in the CNS agents class compared with more than half of NHW women (58%). Also, 36% of NHB (aRR: 0.84, 95% CI: 0.72–0.98), 27% of Hispanic (aRR: 0.79, 95% CI: 0.66–0.94), and 26% of Asian/PI (aRR: 0.75, 95% CI: 0.63–0.90) women reported taking medications in the GI drugs class compared with 59% of NHW women. Twenty-eight percent of NHB women reported taking anti-infective agents compared with 18% NHW, 18% Hispanic, and 12% Asians, but these differences were not statistically significant (Fig. 2).

Fig. 3.

Percentage of women taking the top medication classes during pregnancy by race/ethnicity, National Institute of Child Health and Human Development Fetal Growth Studies Singletons. Modified Poisson regression models were used to estimate the differences in medication class intake by racial/ethnic groups. *p-value < 0.05, **p-value < 0.01; non-Hispanic white, reference group. Regression models were adjusted for maternal age, pre-pregnancy weight, self-reported height, parity, marital status, current employment status, health insurance status, educational level, income level, and study site. CNS, central nervous system; GI, gastrointestinal; NHB, non-Hispanic black; NHW, non-Hispanic white.

We examined whether the observed racial/ethnic differences persisted over the entire pregnancy at median gestational weeks of 13, 20, 28, 32, 36, and 39 (Supplementary Table S4 [available in the online version]). The intake of medications in three classes (antihistamines, anti-infective agents, and CNS agents) decreased for all women from enrollment to the final study visit. In contrast, the consumption of GI drugs increased for all women during the course of pregnancy. Differences by race/ethnicity throughout pregnancy appeared to be stable, and we did not detect a significant interaction between gestational week and the association of race/ethnicity and medication use (interaction p-value for all comparisons > 0.05).

Discussion

In a prospective cohort of nonobese and racially/ethnically diverse pregnant women without pre-pregnancy chronic conditions, supplement and medication consumption was highly prevalent. Specifically, the intake of dietary supplements at any time during pregnancy was nearly ubiquitous in women of all racial/ethnic groups with prenatal vitamins being the most commonly used. Moreover, almost 70% of women took at least one medication during pregnancy with racial/ethnic differences observed in the quantity of medications taken. Hispanic and Asian/PI women took fewer medications during pregnancy than NHW women even after accounting for maternal sociodemographic factors. The top medication classes consumed during pregnancy for all racial/ ethnic groups included GI drugs, CNS agents, anti-infectives, and antihistamines although the percentage of women taking a medication in each class varied by race/ethnicity. The racial/ethnic differences observed in the overall intake of antihistamines, CNS agents, and anti-infectives remained similar regardless of gestational age.

The nearly universal supplement use and the frequent use of medications during pregnancy documented in our study are consistent with results from previous studies, although our prevalence of supplement use was higher and medication use was lower than previously reported.^{4,6,7,10,27,28} This discrepancy is likely due to our inclusion criteria restricting our study to nonobese and nonsmoking women without chronic conditions. The prevalent intake of prenatal vitamins, GI medications, CNS medications, and antihistamines reported during pregnancy in the current study is also consistent with findings previously identified.^4,8,10,29

Several previous studies have reported on racial/ethnic differences in prenatal supplement and medication intake. Similar to previous findings, maternal race/ethnicity was not independently associated with overall use of dietary supplements during pregnancy.^4,11 However, we found that a higher percentage of Asian women than the other racial/ ethnic groups combined (NHW, NHB, and Hispanics) reported intake of prenatal vitamins/multivitamins in the first trimester. We did not find differences among other racial/ethnic groups, which is in contrast to prior findings where African American women were more likely to report dietary supplement nonuse.^15,16 The racial/ethnic diversity of our sample together with the healthy profile of our cohort compared with those from previous studies may explain the differences in findings.

Our study offers a unique contribution to the existing literature by demonstrating the racial/ethnic differences in the quantity and types of medications taken by a cohort of women with low-risk pregnancies, even after adjustment for multiple covariates. Speculations on whether higher medication use among NHW women would place them at a higher risk for experiencing possible negative consequences of taking and mixing medications during pregnancy is outside the scope of the present study; however, these racial/ethnic differences in the quantity and types of medication consumption identified in this study, highlight the potential role of culture surrounding medication intake and health care utilization described in medical literature.^30,31

Previous studies have attributed the increased medication intake during pregnancy to the rise in chronic health conditions among women³² and unhealthy behaviors during pregnancy.¹³ However, our study indicates a high prevalence of medication use even among pregnant women with no pre-pregnancy chronic conditions. Furthermore, this use cannot be attributed to health conditions developed during pregnancy, as only small percentages of women in our cohort developed gestational diabetes (3.5%) or a hypertensive disorder of pregnancy (6%).Nevertheless, there are still other factors that may be responsible for the high prevalence of prenatal medication intake that we did not account for in our study.

It is important for clinicians to be aware that a large percentage of women even those with low-risk pregnancies consume at least one medication, and that a significant proportion consumed three or more. Moreover, the most common medications used during pregnancy in this low-risk cohort can be purchased over the counter without a prescription or physician consultation. Given that all medications can have interactions or other side effects, it is important for clinicians to be aware that their patients may be taking several medications which they have not prescribed. The results of this study should be used to stimulate more studies to explore the potential mechanisms through which race/ethnicity, independent of other confounding factors, may affect prenatal medication use.

Pharmacokinetic changes in medications’ metabolism due to pregnancy³ accompanied by insufficient information on the safety profile of many medications, and the absence of information on the safety of medication combinations, present a challenge to pregnant women and health care professionals who provide advice regarding prenatal medication use. Health care professionals prescribe medications weighing risks and benefits to maternal–fetal health.³³ However, women’s decisions to ultimately take medications during pregnancy may not only be influenced by these patient–physician interactions but also by their knowledge, attitude,³⁴ and perception of risks and benefits associated with certain medications² as well as their family and culture.³⁵ Therefore, understanding how different racial/ethnic backgrounds may affect women’s decisions surrounding medication intake may help physicians improve their counseling.

Although pregnancy is a sensitive time in maternal and fetal health, data on prenatal medication intake and its correlates are limited. Recommendations by the department of health and human services call for more comprehensive data collection and research on addressing knowledge gaps in medication use and safety during the prenatal period.³⁶ We need to better understand the physiological changes during pregnancy that influence the pharmacokinetics of different medications, and the metabolic and transport pathways for many medications and their appropriate dosing during pregnancy.³⁷

A previous study comparing nonpregnant to pregnant women found a lower prevalence of prescription medication intake among pregnant compared with nonpregnant women.²⁹ Alternatively, another study has reported a higher prevalence in over-the-counter medication use during pregnancy compared with 3 months prior to pregnancy.⁷ It is possible that women assume prescription medications to be harmful, while other medications that can be purchased over the counter to be safe for consumption during pregnancy.^38,39 Since the prevalence of over-the-counter medication use is high during pregnancy,⁷ future studies should focus on safety of these medications and their combined use.

The prospective design of the current study was a major advantage compared with most previous studies as it enabled us to reduce recall bias. The longitudinal design also allowed for a more in-depth assessment of the patterns of supplement and medication intake through pregnancy. Furthermore, our study was racially/ethnically diverse which allowed for statistically meaningful comparisons. The characteristics of the study population which included nonobese and nonsmoking women with no pregestational chronic conditions enabled us to eliminate differences in medication intake due to the existence of these conditions. Furthermore, in contrast to previous studies, we adjusted for other maternal characteristics such as education and insurance that could confound the association of race/ethnicity with prenatal medication intake.

While this study provides a more comprehensive analysis of the racial/ethnic differences in prenatal medication intake, it also has potential limitations. Medication data collection was partly based on women’s self-report at in-person interviews at different times during pregnancy and may be subject to recall bias, in particular underreporting.⁴⁰ Information on the clinical indication for taking medication was also not collected. Furthermore, the database was unable to distinguish certain over the counter versus prescription medications. Therefore, we were unable to determine whether there are racial/ethnic differences in intake of either types of medications.

The NICHD Fetal Growth Study was conducted intentionally among low-risk pregnant women without major pre-pregnancy chronic conditions, which consequently limited the impact of these confounders on the association between maternal race/ethnicity and prenatal supplement and medication use. For instance, although racial/ethnic disparities in access to certain types of medications such as antidepressants have been identified previously,^41,42 we did not observe such differences given the study’s primary exclusion criteria and consequently low prevalence of reported use. However, these findings may underestimate medication use in the general population and consequently may not be generalizable to all pregnant women. Nevertheless, this low-risk cohort with minimal confounding has an advantage over prior studies that have attributed the increased prenatal medication use to the rise in chronic conditions and unhealthy behaviors among women during pregnancy.^13,32

In conclusion, the findings indicate high medication use during prenatal period even among women without chronic conditions having low-risk pregnancies. The findings emphasize the need for more research on medication safety and consumption during pregnancy, and a better understanding of how maternal race/ethnicity may influence medication intake.

Key Points.

• In women without chronic conditions, medication use is common.

• Racial/ethnic differences exist in prenatal medications use.

• Almost all women use supplements during pregnancy.