Medications used during the preconception period and during pregnancy, which are critical periods for maternal health and fetal development, require further attention.1 Although an estimated two thirds of women use prescription medications during pregnancy,2 up-to-date information on the full extent of the types of drugs prescribed to pregnant women in recent years is limited. One common condition during pregnancy, diabetes, is associated with increased risk of congenital abnormalities and other adverse pregnancy outcomes.3,4
Medications that reduce hyperglycemia or increase insulin sensitivity are often grouped as “antidiabetic medications” and are used to treat type 1 or type 2 diabetes and other conditions associated with insulin resistance, and are used to treat gestational diabetes mellitus (GDM). With the increasing prevalence of GDM5–8 and the increase in the overall proportion of women who have diabetes and become pregnant,9–12 more women may use these medications during their pregnancies to control hyperglycemia. This descriptive epidemiologic study was conducted as part of the Medication Exposure in Pregnancy Risk Evaluation Program13 to describe the prevalence, trends, and patterns of use of antidiabetic medications in the preconception period and during pregnancy based on maternal demographic characteristics.
MATERIALS AND METHODS
The Medication Exposure in Pregnancy Risk Evaluation Program study is a collaborative effort between the U.S. Food and Drug Administration and researchers from 11 health plan–affiliated research institutions: Group Health Research Institute (Washington); Harvard Pilgrim Health Care Institute (Massachusetts); HealthPartners Research Foundation (Minnesota); Kaiser Permanente Colorado, Georgia, Northern California, Southern California, and Northwest (Oregon, Washington); LCF Research (New Mexico); Meyers Primary Care Institute (Massachusetts); and Tennessee State Medicaid (through the auspices of Vanderbilt University School of Medicine). Administrative health plan data are linked with clinical data systems to facilitate studies of medication use and pregnancy outcomes.13 These research institutions extracted information on maternal and newborn enrollment, demographic characteristics, outpatient pharmacy dispensing, and diagnosis codes from outpatient and inpatient health care encounters from their clinical and administrative data systems and linked them to newborn birth certificates to obtain information on maternal race and ethnicity, parity, and gestational age at birth for the newborn.14 All data were deidentified and standardized across centers and summary data tables, but no individual-level data were shared across the centers to conduct this study. The institutional review boards of each organization and the state Departments of Public Health (when applicable) approved the study.
The source population for this study included 684,635 deliveries to females age 12–50 years with one or more live births between January 1, 2001 and December 31, 2007, while enrolled in any of the 10 health maintenance organizations that provided data for these analyses. Tennessee State Medicaid did not participate in this study. To be eligible for inclusion, women had to be continuously insured with pharmacy benefits from 180 days before pregnancy through their delivery date. The final sample was 437,950 deliveries (64% of the source population); the pharmacy benefit and continuous insurance requirements resulted in the exclusion of 4.5% and 31.5% of the deliveries, respectively.
All antidiabetic medications approved for use by the U.S. Food and Drug Administration during the study period were included in these analyses (Appendix 1, available online at http://links.lww.com/AOG/A336). These medications were identified from the outpatient pharmacy dispensing data. Periods of drug exposure in relation to pregnancy and trimester were calculated from dispense dates and days supplied. For women with multiple refills of the same prescription, a 14-day grace period after the expected exhaustion of the days supplied was incorporated for each dispensing. A 120-day period before the estimated date of conception was used to identify dispensing before pregnancy because some health plans allow for up to a 100-day supply per dispensing.
Women with a health care encounter occurring in the 180 days before pregnancy were identified as potentially having diabetes mellitus, polycystic ovaries, female infertility, or being overweight or obese using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes for these conditions (Appendix 2, available online at http://links.lww.com/AOG/A337) during the 180 days before pregnancy. Additionally, diagnosis codes indicative of diabetes or GDM occurring during the second or third trimester of pregnancy were identified. Maternal age, calendar year of delivery, and number of deliveries per woman during the study were derived from the health plan data systems; maternal race and ethnicity and education were from newborn birth certificate data.
Gestational periods were defined using the last menstrual period or gestational age information recorded in the newborn birth certificates, when available (95% of deliveries). When last menstrual period date was available in the birth certificate, it was used as the first day of the pregnancy. If the last menstrual period was missing or invalid, day 0 was defined as the date of delivery minus the gestational age based on clinical or obstetric estimates. This definition was consistent with the approach used by the National Center for Health Statistics.15 When gestational age information was missing from the birth certificates, trimesters were estimated using the delivery date and ICD-9-CM codes (Appendix 2, http://links.lww.com/AOG/A337).16,17 Trimesters were categorized as first (days 0–90), second (days 91–180), and third (days 181–delivery).
Descriptive statistics are presented to estimate the number and proportion of deliveries to women who were dispensed antidiabetic medications in the preconception period (120 days before pregnancy), and for the second and third trimesters combined. Medications were considered individually and presented in reference to maternal age categories and then combined into broader categories that included both monotherapy (insulin only, biguanide only [only metformin was available], other oral agents [other than metformin] only), combinations of medications from these classes, and a summary category of any antidiabetic medication use. The unadjusted associations between antidiabetic medication use and maternal sociodemographic characteristics were assessed using χ2 tests. The unadjusted trends in the use of medications were assessed using the Cochran-Armitage trend test. Because one of the objectives of this study was to assess trends in prevalence of antidiabetic medication use, and because medication use can change within and between pregnancies, all women with one or more deliveries were retained in the final cohort instead of using the first, last, or randomly selected delivery.
Analysis of medication initiation during pregnancy and switching of medications after conception was conducted by cross-tabulating antidiabetic medications used in the prepregnancy period (when applicable) with those used in the second or third trimester of pregnancy. We chose the second and third trimesters of pregnancy because treatment for GDM is most commonly initiated in the late second trimester and the early third trimester. The underlying reason for antidiabetic medication use was explored using the ICD-9-CM codes.
The final analytic sample for this descriptive study comprised 437,950 deliveries to 372,543 women who had one or more live births during the study period. Of these deliveries, 42.4% were to non-Hispanic white women, 28.3% were to Hispanic women, 12.3% were to Asian women, 8.2% were to black women, and 8.5% were to women of other or unknown race or ethnicity. Most deliveries (63.4%) were to women with more than a high school education. The majority (85.1%) of the deliveries occurred at term (37–41 completed weeks of gestation), whereas 9.4% were preterm (less than 37 weeks of gestation) and 5.5% were postterm (42 weeks of gestation or more).
A little more than 1% (1.21%) of the deliveries were to women dispensed one or more antidiabetic medications in the 120 days before conception (Table 1). Medications used most commonly during this period were metformin (0.84%) and insulin (0.33%). Antidiabetic medication use before pregnancy was significantly associated with increasing maternal age and education and race or ethnicity (Table 2; all P values <.001). Native-American women were most likely to be using antidiabetic medications (1.76%), and black women were least likely to be using them (1.05%). Of the 5,299 deliveries to women using antidiabetic medications before pregnancy, 20.4% were preterm deliveries (less than 37 weeks of gestation) compared with 9.2% of the 432,651 deliveries to women who were not using these medications (P<.001).
Antidiabetic medication use before pregnancy increased from 0.66% of deliveries in 2001 to 1.66% of deliveries in 2007 (P<.001), representing a 2.5-fold increase (Table 2). The use of metformin in the prepregnancy period increased annually, from 0.24% of deliveries in 2001 to 1.16% of deliveries in 2007, representing a 3.8-fold increase over this 7-year period (P<.001). In contrast, the use of insulin alone and other oral agents alone remained relatively stable. The majority of the women who used insulin alone (n=1,233) or in combination with oral agents (n=212) or metformin in combination with any other oral antidiabetic drug (n=299) had a diagnosis code for type 1 or type 2 diabetes (83.5%, 94.3%, and 78.6%, respectively) before pregnancy. Of the women who used metformin alone (n=3,208), only 13.6% had a diagnosis code for diabetes alone or in combination with other conditions of interest, whereas 67.2% had a code for polycystic ovaries or infertility. Specifically, 7.4% had codes for polycystic ovaries alone; 23.6% had codes for polycystic ovaries and infertility; 1.8% had codes for polycystic ovaries and overweight; 10.2% had codes for polycystic ovaries, infertility, and overweight; 20.6% had codes for infertility only; 3.6% had codes for infertility and overweight; 2.2% had codes for overweight only; and 16.4% did not have any of these codes.
One or more antidiabetic medications were dispensed to women in the first trimester for 1.35% of deliveries, and in the second or third trimester (combined) 3.24% of deliveries. The most commonly used antidiabetic medications in the second or third trimester were insulin (2.45%) and sulfonylureas (0.83%), whereas the use of metformin (0.14%) during pregnancy was limited (Table 1). The use of antidiabetic medication was associated with increasing maternal age, decreasing maternal education, and maternal race or ethnicity. Native-American women were most likely to be using antidiabetic medications during pregnancy (4.67%), whereas non-Hispanic white women were least likely to be using these medications (2.40%). Increasing duration of gestation was negatively associated with antidiabetic medication use: 5.6% of the preterm deliveries (less than 37 weeks of gestation) were to women who used antidiabetic medications compared with 3.0% of term deliveries (37 weeks of gestation or more).
The use of any antidiabetic medication during the second or third trimester of pregnancy increased significantly from 2.77% of all deliveries in 2001 to 3.62% of all deliveries in 2007 (P for trend <.001), representing a 29% increase during this period (Table 3). The proportion of deliveries to women using insulin during their pregnancies decreased from 2.41% in 2001 to 2.09% in 2007, whereas the use of other agents, most commonly sulfonylureas, increased from 0.29% to 1.09%, and metformin use increased from 0.01% to 0.19%. Among the 14,185 deliveries to women who used one or more antidiabetic medications during pregnancy, 9.2% had ICD-9-CM codes for type 1 or type 2 diabetes only, 47.6% had codes for GDM only, 41.3% had codes for both GDM and diabetes, and 2.0% did not have a code for either condition. Among deliveries to women who used insulin only, 11.9% had codes for type 1 or type 2 diabetes only, 39.7% had codes for GDM only, 48.2% had codes for diabetes and GDM, and 0.2% did not have a code for either condition. In contrast, for deliveries to women who used oral agents, 76.9% had codes for GDM only, 0.8% had codes for diabetes only, 21.9% had codes for diabetes and GDM, and 0.4% had no diabetes-related codes.
Of the 5,299 deliveries to women who were dispensed an antidiabetic drug during the preconception period, 2,995 (56.5%) used an antidiabetic drug during the second or third trimester of pregnancy. Of the 1,233 deliveries to women who used only insulin in the preconception period, 95.3% continued to use only insulin, 0.8% had an oral agent added, 0.4% switched from insulin to an oral agent, and 3.5% had no evidence of any antidiabetic medication dispensed during pregnancy. Of the 3,208 deliveries to women who used metformin alone in the preconception period, 8.8% continued to use metformin and 16.8% switched to insulin, 3.0% switched to another oral agent, 3.2% switched to other drug combinations, and 68.0% did not use any antidiabetic medication during pregnancy. Of the 341 women who used an oral agent other than metformin, most commonly a sulfonylurea, before pregnancy, 5.3% continued to use an oral agent only, 71.3% switched to insulin only, and 16.7% did not use any antidiabetic medication. Of the 432,651 deliveries to women with no antidiabetic medication use before pregnancy, 1.8% initiated insulin only, 0.1% used insulin in combination with another antidiabetic drugs, 0.7% used an oral agent other than metformin, less than 0.1% used metformin only, and the remaining 97.4% did not use any antidiabetic medications during the second or third trimester.
The use of antidiabetic medications in the preconception period increased significantly from 2001 to 2007, resulting in a 2.5-fold increase. The increase in the use of antidiabetic medications during the second or third trimester of pregnancy was more gradual, increasing by approximately 31% over the 7-year period. The increase in the use of antidiabetic medications before pregnancy is almost entirely attributable to increase in metformin use. Of the women who used metformin, less than 15% had a diagnosis code for diabetes before pregnancy, whereas two thirds had a code for polycystic ovaries or infertility, suggesting that most women using metformin are being treated for polycystic ovaries or infertility associated with polycystic ovaries.18 Metformin increases the rate of ovulation among women with polycystic ovaries,19 but given that the majority (79%) of women with polycystic ovaries also had a code for infertility suggests that most were trying to conceive, whereas a minority of the women may have experienced an unplanned pregnancy resulting from the side effect of the increased fertility.
During the second and third trimesters of pregnancy, we observed a small decrease in the use of insulin, with a concomitant increase in the use of sulfonylureas and, to a lesser extent, metformin. Of the women who used metformin in the preconception period and who continued to use an antidiabetic drug during pregnancy, approximately two thirds (64.1%) switched to insulin or other oral agents by the second trimester, whereas the remainder continued to use metformin. Sulfonylureas are the most commonly used oral agents to treat GDM in the United States, whereas metformin is rarely used because of concerns that it crosses the placenta.20 Two meta-analyses of observational studies did not show an increase in congenital malformations or neonatal deaths associated with metformin use in pregnancy.21,22 The Metformin in Gestational Diabetes Trial, published in 2008 after the deliveries in the present study, reported that metformin (alone or in combination with insulin) was not associated with increased perinatal complications and women reported preferring metformin to insulin treatment.23 Based on previous studies, approximately one quarter of women with GDM diagnosed in the Kaiser Permanente Southern and Northern California regions, which combined comprise 74% of the current study sample, were treated with insulin or oral agents during their pregnancies during this period.24,25
The Medication Exposure in Pregnancy Risk Evaluation Program study cohort comprises deliveries resulting in live births to women who were insured with pharmacy benefits for at least 6 months before conception through delivery. The prepregnancy insurance criteria, deemed necessary to report prepregnancy medication use and switching, excluded the majority of women insured by Medicaid and other women who became insured after conception. Only 3.2% of the deliveries in this study were to women insured through Medicaid. Although this report is based on prescription drug dispensing data, we were unable to assess whether women adhered to the regimen as prescribed by their physicians. Results of oral glucose challenge tests and oral glucose tolerance tests during pregnancy, which are not included in the Medication Exposure in Pregnancy Risk Evaluation Program dataset, would have allowed us to better-differentiate between women with diabetes and GDM.9,26 Additionally, we did not have information on maternal height and weight across the 7-year study period in the Medication Exposure in Pregnancy Risk Evaluation Program dataset to calculate body mass index. Strengths of the study include the large racially and ethnically and geographically diverse population of insured women with more than 400,000 live births over a 7-year period, that deliveries occurred in 10 health plans across eight different states with varying models of health care, and that medications were based on actual dispensing and not self-report.
In the past decades, hyperglycemia during pregnancy was primarily a result of type 1 diabetes and GDM, but type 2 diabetes has emerged in adolescents and has become more prevalent in young adults over the past decade27,28 and the prevalence of GDM has increased.5 Additionally, the recently published criteria for the diagnosis of GDM,29 if adopted, will result in a significant increase in the prevalence of GDM.30,31 The size and the scope of our study allows us to describe the exposure to antidiabetic medications in a contemporary cohort, but it is only the first step in the process toward understanding the potential public health affect of using antidiabetic medications during pregnancy. Critical next steps for the Medication Exposure in Pregnancy Risk Evaluation Program study involve assessing fetal harm, including low birth weight, and prediction of women at highest risk for adverse outcomes in relation to their antidiabetic medication use, taking into account the affect of their hyperglycemia.
1. 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;157:247–50.
2. 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.
3. Platt MJ, Stanisstreet M, Casson IF, Howard CV, Walkinshaw S, Pennycook S, et al.. St Vincent’s declaration 10 years on: outcomes of diabetic pregnancies. Diabetic Med 2002;19:216–20.
4. Balsells M, García-Patterson A, Gich I, Corcoy R. Maternal and fetal outcome in women with type 2 versus type 1 diabetes mellitus: a systematic review and metaanalysis. J Clin Endocrinol Metab 2009;94:4284–91.
5. Ferrara A. Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes Care 2007;30:S141–6.
6. Ferrara A, Kahn HS, Quesenberry CP, Riley C, Hedderson MM. An increase in the incidence of gestational diabetes mellitus: Northern California, 1991-2000. Obstet Gynecol 2004;103:526–33.
7. Dabelea D, Snell-Bergeon JK, Hartsfield CL, Bischoff KJ, Hamman RF, McDuffie RS. Increasing prevalence of gestational diabetes mellitus (GDM) over time and by birth cohort: Kaiser Permanente of Colorado GDM Screening Program. Diabetes Care 2005;28:579–84.
8. Getahun D, Nath C, Ananth CV, Chavez MR, Smulian JC. Gestational diabetes in the United States: temporal trends 1989 through 2004. Am J Obstet Gynecol 2008;198:525.
9. Lawrence JM, Contreras R, Chen W, Sacks DA. Trends in the prevalence of preexisting diabetes and gestational diabetes mellitus among a racially/ethnically diverse population of pregnant women, 1999-2005. Diabetes Care 2008;31:899–904.
10. Feig DS, Razzaq A, Sykora K, Hux JE, Anderson GM. Trends in deliveries, prenatal care, and obstetrical complications in women with pregestational diabetes: a population-based study in Ontario, Canada, 1996-2001. Diabetes Care 2006;29:232–5.
11. Albrecht SS, Kuklina EV, Bansil P, Jamieson DJ, Whiteman MK, Kourtis AP, et al.. Diabetes trends among delivery hospitalizations in the U.S., 1994–2004. Diabetes Care 2010;33:768–73.
12. Hayes DK, Fan AZ, Smith RA, Bombard JM. Trends in selected chronic conditions and behavioral risk factors among women of reproductive age, behavioral risk factor surveillance system, 2001-2009. Prev Chronic Dis 2011;8:A120.
13. Andrade SE, Davis RL, Cheetham TC, Cooper WO, Li DK, Amini T, et al.. Medication exposure in pregnancy risk evaluation program. Matern Child Health J 2012;16:1349–54.
14. Andrade SE, Scott PE, Davis RL, Li DK, Getahun D, Cheetham TC, et al.. Validity of health plan and birth certificate data for pregnancy research. Pharmacoepidemiol Drug Saf 2012 Jul 3. doi: 10.1002/pds.3319 [Epub ahead of print].
15. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Mathews TJ, Kirmeyer S, et al.. Births: final data for 2007. Natl Vital Stat Rep 2010;58:1–85.
16. Raebel MA, Ellis JL, Andrade SE. Evaluation of gestational age and admission date assumptions used to determine prenatal drug exposure from administrative data. Pharmacoepidemiol Drug Saf 2005;14:829–36.
17. Toh S, Mitchell AA, Werler MM, Hernandez-Diaz 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.
18. Lord JM, Flight IH, Norman RJ: Metformin in polycystic ovary syndrome: systematic review and meta-analysis. BMJ 2003;327:951–3.
19. Lord JM, Flight IH, Norman RJ. Insulin-sensitising drugs (metformin, troglitazone, rosiglitazone, pioglitazone, D-chiro-inositol) for polycystic ovary syndrome. The Cochrane Database of Systematic Reviews 2003, Issue 3. Art. No.:CD003053. DOI: 10.1002/14651858.CD003053.
20. Charles B, Norris R, Xiao X, Hague W. Population pharmacokinetics of metformin in late pregnancy. Ther Drug Monit 2006;28:67–72.
21. Gutzin SJ, Kozer E, Magee LA, Feig DS, Koren G. The safety of oral hypoglycemic agents in the first trimester of pregnancy: a meta-analysis. Can J Clin Pharmacol 2003;10:179–83.
22. Gilbert C, Valois M, Koren G. Pregnancy outcome after first-trimester exposure to metformin: a meta-analysis. Fertil Steril 2006;86:658–63.
23. Rowan JA, Hague WM, Gao W, Battin MR, Moore MP. Metformin versus insulin for the treatment of gestational diabetes. N Engl J Med 2008;358:2003–15.
24. Lawrence JM, Hsu J-W, Chen W, Black MH, Sacks DA. Prevalence and timing of postpartum glucose testing and sustained glucose dysregulation after gestational diabetes mellitus. Diabetes Care 2010;33:569–76.
25. Ferrara A, Peng T, Kim C. Trends in postpartum diabetes screening and subsequent diabetes and impaired fasting glucose among women with histories of gestational diabetes mellitus: a report from the Translating Research Into Action for Diabetes (TRIAD) Study. Diabetes Care. 2009;32:269–74.
26. Andrade SE, Moore Simas TA, Boudreau D, Raebel MA, Toh S, Syat B, et al.. Validation of algorithms to ascertain clinical conditions and medical procedures used during pregnancy. Pharmacoepidemiol Drug Saf 2011;20:1168–76.
27. Dabelea D, Bell RA, D'Agostino RB Jr, Imperatore G, Johansen JM, Linder B, et al.. Incidence of diabetes in youth in the United States. JAMA 2007;297:2716–24.
28. Liese AD, D'Agostino RB Jr, Hamman RF, Kilgo PD, Lawrence JM, Liu LL, et al.. The burden of diabetes mellitus among US youth: prevalence estimates from the SEARCH for Diabetes in Youth Study. Pediatrics 2006;118:1510–8.
29. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, et al.. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010;33:676–82.
30. Black MH, Sacks DA, Xiang AH, Lawrence JM. Clinical outcomes of pregnancies complicated by mild gestational diabetes mellitus differ by combinations of abnormal oral glucose tolerance test values. Diabetes Care 2010;33:2524–30.
31. Holt RI, Coleman MA, McCance DR. The implications of the new International Association of Diabetes and Pregnancy Study Groups (IADPSG) diagnostic criteria for gestational diabetes. Diabetic Med 2011;28:382–5.
Supplemental Digital Content
© 2013 The American College of Obstetricians and Gynecologists