Data regarding the risks of most prescription medications during pregnancy are inadequate, posing serious public health concerns.1 Identifying medications commonly used in pregnancy is critical to inform research priorities regarding safety.2 This is particularly relevant considering the U.S. Food and Drug Administration's (FDA) December 2014 Pregnancy and Lactation Labeling Rule, which made historic changes to prescription medication labeling requirements regarding use during pregnancy and breastfeeding.3
Several studies from the United States within the past two decades have reported on medication use during pregnancy within broad classes.4–84–84–84–84–8 Fewer have identified the most commonly used prescription medications during pregnancy.9–119–119–11 These studies were based on maternal recall after delivery (National Birth Defects Prevention Study and Slone Epidemiology Center Birth Defects Study)10,1110,11 and on commercial health plan databases (Health Maintenance Organization [HMO] Research Network).9 Furthermore, the use of medications with potential for fetal harm has been described using medical records and claims from commercial health plans and the Tennessee Medicaid program.12–1412–1412–14
Medicaid insures more than 40% of all U.S. deliveries,15 and pregnant women enrolled in Medicaid are low income, racially–ethnically diverse, and tend to be relatively young.16 The Medicaid Analytic eXtract, which contains health care utilization data from Medicaid enrollees, provides an opportunity to study this large population of women whose demographics differ from previous descriptive studies.9–119–119–11 We aimed to characterize the 20 most common prescription medications dispensed at least once (referred to hereafter as “most commonly dispensed”), the 10 most commonly dispensed prescription medications classified in the former FDA categories D or X, and the 10 most commonly dispensed classes among pregnant women enrolled in Medicaid.
MATERIALS AND METHODS
The Medicaid Analytic eXtract contains beneficiary enrollment information and claims from inpatient and outpatient procedures and diagnoses and outpatient pharmacy dispensings reimbursed by Medicaid. Medicaid data on medication type and dispensing dates are generally believed to be accurate17 and have been used to study medication use during pregnancy.8,18–208,18–208,18–208,18–20 We used Medicaid Analytic eXtract data from 2000 to 2007 because these years were available to our research group and overlapped with years from the previous studies.9–119–119–11 As previously described, more than 7 million women aged 12–55 years with delivery-related health care claims were identified from 2000 to 2007 Medicaid Analytic eXtract data and were linked with newborns.16 We estimated the date of the last menstrual period to be 245 days before the newborn's birth date for pregnancies with maternal or newborn International Classification of Diseases, 9th Revision, codes indicative of preterm birth (644.0, 644.2, and 765.x), and 270 days before the newborn's birth date for other pregnancies.21 This method correctly classified gestational age at delivery within 2 weeks for 75% of preterm and nearly all term deliveries in a similar database.21 We required that women have continuous enrollment in Medicaid, no private insurance, no restricted benefits, and appropriate enrollment type from 3 months before the last menstrual period through delivery. The study cohort contained 1,106,757 pregnancies ending in live birth from 45 U.S. states and Washington, DC. This project was approved by the Brigham and Women's Hospital, Harvard School of Public Health, and University of California, San Diego's institutional review boards and a data use agreement was in place with the Centers for Medicare & Medicaid Services.
We used the Medicaid Analytic eXtract Prescription Drug file to identify medication dispensings excluding vitamins (including iron preparations), caloric agents, electrolytic replacement preparations, diagnostic agents, dental agents, and devices and medical supplies. The same woman could have multiple dispensings for the same medication. We defined medication prevalence as the proportion of pregnancies with at least one dispensing date for medications of interest during the pregnancy periods of interest including: pregnancy: last menstrual period date until the day before the delivery date; prepregnancy: 90 days before last menstrual period date until the day before the last menstrual period date; first trimester: last menstrual period date until 89 days after the last menstrual period date; second trimester: 90 days after the last menstrual period date until 179 days after the last menstrual period date; and third trimester: 180 days after the last menstrual period date until the day before the delivery date.
There were 2,506 unique generic medication names dispensed during pregnancy. Generic medications that were identical except for the salt were merged. Medications with combination ingredients were not merged with single-ingredient formulations. We identified the 20 medications most commonly dispensed during pregnancy and reported prevalence overall and by pregnancy period, maternal age, and race–ethnicity. Using prevalence ratios and Wald 95% confidence intervals, we compared proportions by age and race–ethnicity, two major demographic characteristics that differentiate Medicaid from other data sources. We stratified by birth year combining 2000 with 2001 because few pregnancies were available from 2000 as a result of the enrollment requirements. We plotted year-specific proportions for those with at least a 40% increase or decrease between any years.
We identified medications assigned to the former FDA D or X categories published in Briggs, Freeman, and Yaffe.22 Although this classification system is limited for conveying teratogenic risk for clinical practice, in part because designations also depend on maternal benefits,11,23,2411,23,2411,23,24 we used this method to classify medications with the potential to cause fetal harm. When the manufacturer's designated category differed from that assigned by Briggs et al or by trimester of use, we classified the medication in the least favorable category (eg, medications with both C and D designations were classified as D). Clomiphene, follitropin, and chorionic gonadotropin, identified as commonly used X medications in previous studies,10,1110,11 were excluded because fertility treatments are rarely covered by Medicaid.25 After combining hormonal contraceptives, there were 180 different D or X medications dispensed during pregnancy. We reported the prevalence of all medications, D or X medications, and X medications overall and by pregnancy period, age, race–ethnicity, and year. Additionally, we reported proportions overall and by pregnancy period, age, and race–ethnicity for the 10 most commonly dispensed D and X medications during the entire pregnancy.
We grouped medications into classes according to the second tier of the American Hospital Formulary Service Pharmacologic-Therapeutic Classification system or the first tier for medications lacking a more granular classification.26 Medications may have multiple classes because of multiple indications, mechanisms of action, routes of administration, or ingredients. Because all American Hospital Formulary Service classes for a medication are considered equally valid,26 medications with multiple classes were included in more than one class. We identified the 10 most commonly dispensed classes during the entire pregnancy. We reported the prevalence overall and stratified by pregnancy period, age, and race–ethnicity. We also stratified by year and plotted the year-specific proportions for those with at least a 20% increase or decrease between any years.
We identified medications used by more than 1.5% of women during the first trimester or by more than 3% of women during the entire pregnancy excluding fertility treatments and unspecified medications (eg, not otherwise specified antibiotic) in at least one of the previous U.S. studies with overlapping study years: Slone Epidemiology Center Birth Defects Study 1999–2003, National Birth Defects Prevention Study 2004–2007, Slone Epidemiology Center Birth Defects Study 2004–2008, or HMO Research Network 1996–2000.9–119–119–11 We restricted to oral formulations as described in the publications. We identified the corresponding medication prevalence in the current study population.
In this cohort, the mean maternal age was 23.2 years (standard deviation 5.8 years). Overall, 39.9% of women were white, 33.7% were black, 16.3% were Hispanic, 3.5% were Asian, 1.8% were Native American, and 4.8% were classified as having other or unknown race–ethnicity. A total of 11.2% of women had preterm delivery.
During pregnancy, 82.5% of the cohort was dispensed at least one prescription medication; the proportion was lowest among Asian women (69.7%) and did not differ meaningfully by age and year (Appendices 1–4, available online at http://links.lww.com/AOG/A667). The proportion dispensed at least one prescription medication by trimester increased slightly from before pregnancy (52.3%) through the third trimester (57.3%) (Appendix 1, http://links.lww.com/AOG/A667).
The most commonly dispensed prescription medications during pregnancy were nitrofurantoin (21.6%), metronidazole (19.4%), amoxicillin (18.0%), azithromycin (16.9%), and promethazine (13.5%) (Table 1). Nitrofurantoin had the greatest relative increase from 1.4% before pregnancy to 9.8% in the third trimester. Compared with other pregnancy periods, the prevalence of promethazine was highest in the first trimester (8.4%) as was metoclopramide (2.9%) and clindamycin (2.0%).
The prevalence of some of the most common medications differed by age and race–ethnicity (Tables 2 and 3). Women ages younger than 20 years had the highest proportions of several medications. Compared with women 35 years old or older, sulfamethoxazole and trimethoprim was 51% higher, nitrofurantoin was 55% higher, metronidazole was 73% higher, and azithromycin was 93% higher during pregnancy among women younger than 20 years. White women had the highest proportions of common medications, with some exceptions. Notably, metronidazole was 107% higher among black women, miconazole was 126% higher among Hispanic women, and acetaminophen was more than 400% higher among Asian women than white women. Major changes in year-specific proportions are illustrated in Figure 1.
Excluding fertility treatments, 42.0% had at least one dispensing for a category D or X medication during pregnancy; the proportions were 39.6% for D and 5.3% for X medications. The proportion with D or X medications decreased from 39.8% before pregnancy to 18.2% in the third trimester (Appendices 1–4, http://links.lww.com/AOG/A667). When stratifying by age, women 25–29 years had the highest proportion of D or X medications (44.9%) and women younger than 20 years had the lowest proportion (36.2%). White women had the highest proportion of D or X medications (50.5%), and Asian women had the lowest proportion (30.6%). The proportion of dispensed X medications increased between 2000 and 2001 (4.1%) and 2007 (5.4%).
During pregnancy, the most commonly dispensed D medications included: codeine (11.9%), hydrocodone (10.2%), ibuprofen (4.9%), sulfamethoxazole (4.0%), and hydrocortisone (4.0%), and the most prevalent X medications included hormonal contraceptives (4.9%), temazepam (0.11%), atorvastatin (0.07%), warfarin (0.04%), and simvastatin (0.04%) (Table 4). Age and race–ethnicity-stratified proportions of the most common D and X medications are listed in Appendices 5–6 (available online at http://links.lww.com/AOG/A667).
The most commonly dispensed classes during pregnancy included antibacterials (49.7%), analgesics and antipyretics (29.6%), skin and mucous membrane anti-infectives (28.7%), urinary anti-infectives (21.7%), and first-generation antihistamines (18.6%) (Appendix 7, available online at http://links.lww.com/AOG/A667). Respiratory antihistamines (18.1%) were frequently captured by other common classes including first-generation antihistamines, sympathomimetic agents, and bronchodilators. Bronchodilators (14.3%), the tenth most commonly dispensed class, included sympathomimetic agents, antihistamines, and other agents. In terms of age, younger women had the highest proportions of several classes (Appendix 8, available online at http://links.lww.com/AOG/A667). White women had the highest and Asian women had the lowest proportions of most of the common classes (Appendix 9, available online at http://links.lww.com/AOG/A667). Major changes in year-specific class proportions are illustrated in Appendix 10 (available online at http://links.lww.com/AOG/A667).
Prevalence estimates of the most commonly used medications during the first trimester and during pregnancy from previous U.S. studies are reported in Appendices 11 and 12 (available online at http://links.lww.com/AOG/A667).9–119–119–11
Prescription medication use during pregnancy in this low-income cohort was common; four of five pregnancies had at least one dispensing and two of five had at least one dispensing for D or X medications. In agreement with previous U.S. studies using self-report, health care claims, and nationally representative survey data, medications used to treat infections and allergies or asthma were the most common prescription medications.9–11,279–11,279–11,279–11,27 The prevalence of several medications varied by age and race–ethnicity. These differences could partially explain differences in use between this cohort, which was relatively young and racially–ethnically diverse, compared with women from previous studies.
Of the 20 most commonly dispensed medications, nine are rated by the Teratogen Information System as having limited to fair data quality and quantity to inform human teratogenic risk assessments: azithromycin, cephalexin, terconazole, hydrocodone–acetaminophen, albuterol, clindamycin, miconazole, fluconazole, and sulfamethoxazole–trimethoprim.28 The other 11 received fair to good ratings. Lack of unambiguous safety information may lead to the use of medications with potential to cause adverse pregnancy outcomes, whereas beneficial medications may be avoided.29,3029,30
In this cohort, 82.5% were dispensed a prescription medication during pregnancy, which is higher than reports from the Slone Epidemiology Center Birth Defects Study (70.0%, 2006–2008), National Birth Defects Prevention Study (49.4%, 1997–2003),10 and the HMO Research Network (64%, 1996–2000).9 A caveat for these comparisons is that the Slone Epidemiology Center Birth Defects Study and National Birth Defects Prevention Study included medications available only by prescription, whereas the database studies also included over-the-counter medications dispensed at pharmacies. Moreover, we identified differences in specific medication estimates in this study compared with the Slone Epidemiology Center Birth Defects Study, National Birth Defects Prevention Study, and HMO Research Network study. We relied on medication dispensing records as did the HMO Research Network study and may overestimate actual use. However, studies using maternal recall, including the Slone Epidemiology Center Birth Defects Study and National Birth Defects Prevention Study, may underestimate use, especially for specific medications rather than classes, which may be difficult to remember months after delivery, for example, amoxicillin compared with unspecified antibiotic. Potential contributions of population composition to the observed differences in medication use across studies are discussed in Appendix 13 (available online at http://links.lww.com/AOG/A667). Finally, other studies reported less D or X use but used different classifications.12–1412–1412–14 Each of the 10 most common D medications are also classified as B or C depending on circumstances of use.22
Besides overestimation of medication use from dispensing information, additional sources of exposure misclassification should be considered. First, no information was available for medications purchased over the counter or prescribed during hospitalizations. Dispensing information in health care databases underestimates use of medications also available over the counter including analgesics and antipyretics as a class and acetaminophen, ibuprofen, and miconazole. The observed variability by age and race–ethnicity for these medications could be explained in part by demographic differences in over-the-counter compared with prescription use. Second, gestational age at delivery was unavailable, and last menstrual period and trimester dates had to be estimated. As such, misclassification of the dispensing timing by pregnancy period is possible but should be minor considering the long exposure windows (3–9 months long).31 Finally, the cohort was comprised of pregnancies ending in live births. Consequently, medications that cause fetal harm may be underestimated because women with spontaneous abortions, terminations, and stillbirths are not included.9
Only women enrolled in Medicaid before pregnancy were included in this study, that is, those who were Medicaid-eligible because they were classified as children, multiparae, or as having a disability.16 We expect results to generalize well to pregnant women enrolled in Medicaid with live births, including those who became eligible because of pregnancy given similarities in eligibility classification, age, and race–ethnicity distributions.16 However, if prepregnancy enrollees had greater medication use, exposure may be overestimated for all Medicaid pregnancies.
This study is the first to describe the most commonly dispensed medications during pregnancy in a nationwide Medicaid cohort. Medication information was collected prospectively and avoided recall problems. Furthermore, information on generic names was available and vague classifications, for example, unspecified antibiotic, were avoided.
The characterization of prescription medication use among pregnant women enrolled in Medicaid augments our knowledge of medication use during pregnancy by providing information from low-income women. Geographic variability in medication use and contributors to the identified age and racial–ethnic disparities should be investigated. Experts recommended frequency of medication use among pregnant women as one criterion to set priorities for the Centers for Disease Control and Prevention's “Treating for Two: Safer Medication Use in Pregnancy” initiative.32 Considering the differences observed between the current and previous studies, multiple data sources, including Medicaid, should be used to inform this criterion. The FDA's Pregnancy and Lactation Labeling Rule3 is an impetus for research to inform labeling of newer medications. Our study indicates, however, that many older medications are commonly used and also lack safety evidence to inform treatment decisions. Class and medication rankings from this study and others could be used by the FDA to prioritize medications for pregnancy risk narrative label updates.
1. Adam MP, Polifka JE, Friedman JM. Evolving knowledge of the teratogenicity of medications in human pregnancy. Am J Med Genet C Semin Med Genet 2011;157C:175–82.
2. Mitchell AA. Systematic identification of drugs that cause birth defects—a new opportunity. N Engl J Med 2003;349:2556–9.
3. U.S. Food and Drug Administration. Pregnancy and lactation labeling final rule. Available at: http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/Labeling/ucm093307.htm
. Retrieved January 26, 2015.
4. Toh S, Li Q, Cheetham TC, Cooper WO, Davis RL, Dublin S, et al.. Prevalence and trends in the use of antipsychotic medications during pregnancy in the U.S., 2001–2007: a population-based study of 585,615 deliveries. Arch Womens Ment Health 2013;16:149–57.
5. Avalos LA, Chen H, Yang C, Andrade SE, Cooper WO, Cheetham CT, et al.. The prevalence and trends of antiviral medication use during pregnancy in the US: a population-based study of 664,297 deliveries in 2001–2007. Matern Child Health J 2014;18:64–72.
6. Epstein RA, Bobo WV, Martin PR, Morrow JA, Wang W, Chandrasekhar R, et al.. Increasing pregnancy-related use of prescribed opioid analgesics. Ann Epidemiol 2013;23:498–503.
7. Bateman BT, Hernandez-Diaz S, Rathmell JP, Seeger JD, Doherty M, Fischer MA, et al.. Patterns of opioid utilization in pregnancy in a large cohort of commercial insurance beneficiaries in the United States. Anesthesiology 2014;120:1216–24.
8. Desai RJ, Hernandez-Diaz S, Bateman BT, Huybrechts KF. Increase in prescription opioid use during pregnancy among Medicaid-enrolled women. Obstet Gynecol 2014;123:997–1002.
9. Andrade SE, Gurwitz JH, Davis RL, Chan KA, Finkelstein JA, Fortman K, et al.. Prescription drug use in pregnancy. Am J Obstet Gynecol 2004;191:398–407.
10. Mitchell AA, Gilboa SM, Werler MM, Kelley KE, Louik C, Hernández-Díaz S; National Birth Defects Prevention Study. Medication use during pregnancy, with particular focus on prescription drugs: 1976–2008. Am J Obstet Gynecol 2011;205:51.e1–8.
11. Thorpe PG, Gilboa SM, Hernandez-Diaz S, Lind J, Cragan JD, Briggs G, et al.. Medications in the first trimester of pregnancy: most common exposures and critical gaps in understanding fetal risk. Pharmacoepidemiol Drug Saf 2013;22:1013–8.
12. Riley EH, Fuentes-Afflick E, Jackson RA, Escobar GJ, Brawarsky P, Schreiber M, et al.. Correlates of prescription drug use during pregnancy. J Womens Health (Larchmt) 2005;14:401–9.
13. Andrade SE, Raebel MA, Morse AN, Davis RL, Chan KA, Finkelstein JA, et al.. Use of prescription medications with a potential for fetal harm among pregnant women. Pharmacoepidemiol Drug Saf 2006;15:546–54.
14. Cooper WO, Hickson GB, Ray WA. Prescriptions for contraindicated category X drugs in pregnancy among women enrolled in TennCare. Paediatr Perinat Epidemiol 2004;18:106–11.
15. Maternal and child health (MCH) update: states increase eligibility for children's health in 2007. 2008. Available at: http://www.nga.org/files/live/sites/NGA/files/pdf/0811MCHUPDATE.PDF
. Retrieved March 3, 2015.
16. Palmsten K, Huybrechts KF, Mogun H, Kowal MK, Williams PL, Michels KB, et al.. Harnessing the Medicaid Analytic eXtract (MAX) to Evaluate Medications in Pregnancy: Design Considerations. PLoS One 2013;8:e67405.
17. Hennessy S, Palumbo Freeman C, Cunningham F. US government claims databases. In: Strom BL, Kimmel SE, Hennessy S, editors. Pharmacoepidemiology. 5th ed. Chichester (UK): Wiley-Blackwell; 2012. p. 209–23.
18. Huybrechts KF, Palmsten K, Mogun H, Kowal M, Avorn J, Setoguchi-Iwata S, et al.. National trends in antidepressant medication treatment among publicly insured pregnant women. Gen Hosp Psychiatry 2013;35:265–71.
19. Bateman BT, Hernandez-Diaz S, Huybrechts KF, Palmsten K, Mogun H, Ecker JL, et al.. Patterns of outpatient antihypertensive medication use during pregnancy in a Medicaid population. Hypertension 2012;60:913–20.
20. Phiri K, Fischer MA, Mogun H, Williams PL, Palmsten K, Seage GR III, et al.. Trends in antiretroviral drug use during pregnancy among HIV-infected women on Medicaid: 2000–2007. AIDS Patient Care STDS 2014;28:56–65.
21. Margulis AV, Setoguchi S, Mittleman MA, Glynn RJ, Dormuth CR, Hernández-Díaz S. Algorithms to estimate the beginning of pregnancy in administrative databases. Pharmacoepidemiol Drug Saf 2013;22:16–24.
22. Briggs GG, Freeman RK, Yaffee SJ, editors. Drugs in pregnancy and lactation: a reference guide to fetal and neonatal risk. 8th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2008.
23. Chambers CD, Polifka JE, Friedman JM. Drug safety in pregnant women and their babies: ignorance not bliss. Clin Pharmacol Ther 2008;83:181–3.
24. Ramoz LL, Patel-Shori NM. Recent changes in pregnancy and lactation labeling: retirement of risk categories. Pharmacotherapy 2014;34:389–95.
25. Ranji U, Salganicoff A, Stewart AM, Cox M, Doamekpor L. State Medicaid coverage of family planning services: summary of state survey findings November 2009. Available at: https://kaiserfamilyfoundation.files.wordpress.com/2013/01/8015.pdf
. Retrieved January 20, 2015.
26. AHFS pharmacologic-therapeutic classification. Available at: http://www.ahfsdruginformation.com/pt-classification-system.aspx
. Retrieved January 22, 2015.
27. Tinker SC, Broussard CS, Frey MT, Gilboa SM. Prevalence of prescription medication use among non-pregnant women of childbearing age and pregnant women in the United States: NHANES, 1999–2006. Matern Child Health J 2015;19:1097–106.
28. Friedman JM, Polifka JE. TERIS. In: Micromedex 2.0 (electronic version). Greenwood Village (CO): Truven Health Analytics. Available at: http://www.micromedexsolutions.com
. Retrieved January 7, 2015.
29. Lagoy CT, Joshi N, Cragan JD, Rasmussen SA. Medication use during pregnancy and lactation: an urgent call for public health action. J Womens Health (Larchmt) 2005;14:104–9.
30. Parisi MA, Spong CY, Zajicek A, Guttmacher AE. We don't know what we don't study: the case for research on medication effects in pregnancy. Am J Med Genet C Semin Med Genet 2011;157C:247–50.
31. Toh S, Mitchell AA, Werler MM, Hernández-Díaz S. Sensitivity and specificity of computerized algorithms to classify gestational periods in the absence of information on date of conception. Am J Epidemiol 2008;167:633–40.
32. Broussard CS, Frey MT, Hernandez-Diaz S, Greene MF, Chambers CD, Sahin L, et al.. Developing a systematic approach to safer medication use during pregnancy: summary of a Centers for Disease Control and Prevention–convened meeting. Am J Obstet Gynecol 2014;211:208–14.e1.