Variations in body mass index (BMI, calculated as weight (kg)/[height (m)]2) are associated with menstrual cycle disturbances and diminished reproductive function.1 Although this is a recognized phenomenon, relatively little attention has been paid to how variations in BMI might be associated with the timing of the recognition of pregnancy and subsequent entry into prenatal care. If menstrual cycle irregularities cause a woman to delay home pregnancy testing, that could in turn result in a delay in seeking early prenatal care.
In 2012, 29.6% of women in the United States were overweight; an additional 27.5% were obese.2 The prevalence of menstrual cycle irregularities increases with increasing BMI, even among women who do not carry the diagnosis of polycystic ovarian syndrome3,4; 30–47% of overweight or obese women are estimated to have menstrual cycle irregularities.1
Overweight and obese women may also experience higher rates of contraceptive failure,5,6 which, coupled with their increased prevalence of menstrual disturbances, could put them at increased risk of experiencing a delay in pregnancy recognition. Given their increased number of medical comorbidities coupled with their heightened risk for excess gestational weight gain and pregnancy-related complications, early prenatal care for overweight and obese women might be particularly important.1 To address this data gap, we investigated whether prepregnancy BMI is associated with the timing of pregnancy recognition and initiation of prenatal care.
MATERIALS AND METHODS
We used cross-sectional population-based national data from the 2009–2010 Pregnancy Risk Assessment Monitoring System. Data were available from 30 states: Alaska, Arkansas, Colorado, Delaware, Georgia, Hawaii, Illinois, Massachusetts, Maryland, Maine, Michigan, Minnesota, Missouri, Mississippi, Nebraska, New Jersey, New York, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, Tennessee, Texas, Utah, Vermont, Washington, Wisconsin, West Virginia, and Wyoming. We included only women not missing data on prenatal care or BMI.
In brief, participating states drew a sample of 100–250 birth certificates a month and attempted to contact the mothers 2–4 months after delivery to complete the standardized Pregnancy Risk Assessment Monitoring System questionnaire. Selected women were first contacted by mail and if there was no response, an attempt was made to interview them by telephone. Questionnaires received in the mail were data-entered in a proprietary data entry system, which is overseen by the Centers for Disease Control and Prevention. A 10% sample of mail questionnaires was double-entry-verified to ensure accuracy. Questionnaires that were completed over the phone were done with the help of a proprietary computer-assisted telephone interviewing program. Mother's responses were then linked to extracted items from the birth certificate and data were assigned a sampling weight to adjust for nonresponse to produce population-based estimates. Further details regarding study design and data collection are available elsewhere.7,8
Body mass index was available from the birth certificate and from the prepregnancy height and weight as self-reported by the mother on the Pregnancy Risk Assessment Monitoring System questionnaire. Both were categorized into underweight, normal weight, overweight, and obese according to World Health Organization standards.9 A BMI less than 18.5 was classified as underweight, 18.5–24.9 as normal, 25.0–29.9 as overweight, and 30.0 or greater as obese. As a result of the uncertainty regarding the validity of the source of the prepregnancy BMI reported on the birth certificate, we used the BMI category as calculated from the mother's self-reported data for this analysis. For the 1,997 (2.9%) women for whom self-reported prepregnancy height and weight were unavailable, we used the BMI as reported on the birth certificate.
For the primary outcome, the Pregnancy Risk Assessment Monitoring System questionnaire queried women regarding how many weeks or months pregnant they were when they had their first visit for prenatal care. Late prenatal care was defined as initiating care at or after 14 weeks of gestation. Women were also asked how many weeks or months pregnant they were when they were first sure that they were pregnant. All responses were converted to weeks.
Covariates were chosen by review of the literature and examination of a directed acyclic graph.10–12 Covariates included: maternal race (black, white, and other), Hispanic ethnicity (yes or no), smoking status in the 3 months before pregnancy, pregnancy intendedness, maternal age (categories), marital status (married or not), maternal educational attainment (years), and parity. We also examined the role of insurance status in relation to prenatal care initiation. However, because of the way that the questions were asked, only women who received prenatal care were asked about insurance coverage for that care. To ensure that we could examine the effect of insurance status (private insurance, Medicaid, and uninsured) for everyone, for women who reported receiving no prenatal care (0.7%), we used the insurance status at delivery as a proxy for insurance status at the time of prenatal care initiation.
Deidentified data were examined for missing or extreme values or both. Differences in distributions of variables by prenatal care status and BMI were examined using χ2 tests and analysis of variance. In instances in which pairwise comparisons were made, these were accomplished with pairwise contrasts after regression analysis. We used ordinary least squares regression to examine the association among maternal prepregnancy BMI, the week of pregnancy recognition, and the week of prenatal care initiation after adjustment for covariates. Unconditional logistic regression was used to calculate the adjusted odds of receiving late prenatal care. Participants who were missing data on any of the covariates were dropped from the regression analyses. Data were weighted using the sampling weights to produce population-based estimates. A P value of <.05 was considered statistically significant. All data were analyzed using the survey procedure in Stata 12.0. The study was reviewed and approved by the biomedical institutional review board at The Ohio State University.
Among the 72,913 Pregnancy Risk Assessment Monitoring System participants in 2009–2010, 69,872 (96%) met the eligibility criteria for analysis. Among these women, 88.8% reported initiating prenatal care in the first trimester of pregnancy. Women who initiated prenatal care late were more likely to be nonwhite, Hispanic, younger, to have less education, to be parous, to have smoked in the 3 months before pregnancy, to have Medicaid as their source of health insurance at the time of prenatal care initiation, and to be unmarried. Those who initiated care late were also less likely to report that they were trying to get pregnant at that time. Women who reported having no prenatal care were more likely to be nonwhite, younger, unmarried, and uninsured. They were also less likely to report that they were trying to get pregnant at that time (Table 1).
Women completed the Pregnancy Risk Assessment Monitoring System questionnaire a mean of 117 days after delivery. The time at which women reported that they knew that they were pregnant differed by timing of prenatal care initiation. Women who initiated prenatal care early knew at 5.4 weeks of gestation on average, whereas women who initiated care late or received no prenatal care reported that they knew for sure that they were pregnant at 10.7 and 11.0 weeks of gestation, respectively (Table 1). Women who initiated care early had their first prenatal care visit at 7.8 weeks of gestation on average compared with 18.8 weeks of gestation for women who initiated care late.
Underweight women were least likely to report that they were trying to get pregnant. Among women who reported that they were not trying to get pregnant, obese women were the least likely to report use of contraception when they became pregnant. Both underweight and obese women were more likely to report that they were not using contraception because they thought that they or their partner was sterile (9.6% and 11.3%, respectively). These groups were also more likely to report that they did not think that they could get pregnant at that time (30.2% and 30.2%, respectively) (Table 2).
Underweight women also reported that they knew for sure that they were pregnant slightly later (6.3 weeks of gestation) than normal-weight women, but the difference was not clinically important (Table 2). There were no differences by BMI regarding when women initiated prenatal care. Underweight women were more likely to report that they did not receive care as early as they wanted (Table 3). Among those who reported that they did not receive prenatal care as early as they wanted, underweight women were more likely to report that this was the result of no transportation (20.8%) and that they were too busy (22.7%). This finding is likely related to the overrepresentation of teen mothers in the underweight category, 19.5% compared with 11.5% of normal-weight women. Normal-weight women who reported that they did not receive care as early as they wanted were more likely to report that was because they wanted to keep the pregnancy a secret (15.6%).
Table 4 presents the estimates of association between prepregnancy BMI status and timing of pregnancy recognition and prenatal care initiation after adjustment for maternal race, Hispanic ethnicity, smoking status in the 3 months before pregnancy, pregnancy intentions, insurance status, maternal age, marital status, maternal education, and parity. We also examined if there was any confounding by geographic region, but there was not. There was no association between prepregnancy BMI status and the week of pregnancy recognition with the mean time of pregnancy recognition being approximately 6 weeks of gestation regardless of BMI. In contrast, after adjustment for confounding, obese women initiated prenatal care 0.20 weeks earlier on average compared with normal-weight women, although the difference was not clinically important (mean difference −0.20, 95% confidence interval [CI] −0.38 to −0.03). When examining the odds of receiving late or no prenatal care, there was no association with prepregnancy BMI.
One factor that was found to have a significant association with the timing of pregnancy recognition and prenatal care initiation was being uninsured. Uninsured women reported that they found out that they were pregnant slightly later on average (6.4 weeks of gestation compared with 5.7 weeks of gestation, mean difference 0.67, 95% CI 0.24–1.09) than privately insured women after adjustment for confounders. Uninsured women reported initiating prenatal care almost 3 weeks later on average than privately insured women (11.5 weeks of gestation compared with 8.6 weeks of gestation, mean difference 2.83, 95% CI 2.27–3.38), which translated into a more than fourfold increased odds of receiving late (at 14 weeks of gestation or later) or no prenatal care (odds ratio 4.04, 95% CI 3.13–5.23). Women who reported having Medicaid when they found out that they were pregnant did not report any clinically important delays compared with privately insured women (9.3 weeks of gestation compared with 8.6 weeks of gestation).
Not surprisingly, women who reported that they were trying to get pregnant recognized their pregnancy approximately 1 week earlier than women who were not trying after adjustment for confounding (5.4 weeks of gestation compared with 6.5 weeks of gestation, mean difference −1.08, 95% CI −1.19 to −0.97). They also initiated prenatal care a little more than 1 week earlier than women who were not trying (8.4 weeks of gestation compared with 9.6 weeks of gestation, mean difference −1.24, 95% CI −1.37 to −1.10). Finally, women who were trying to get pregnant had a much lower odds of receiving late or no prenatal care (odds ratio 0.37, 95% CI 0.33–0.42).
Both early recognition of pregnancy and early initiation of prenatal care afford women the opportunity to influence their pregnancy outcomes in a positive way. Studies have shown that pregnancy recognition alone is enough for women to adopt healthier behaviors, even before the first prenatal visit.13,14 Late recognition of pregnancy is associated with an increased odds of prematurity and low birth weight.15 This study found no evidence that prepregnancy BMI was associated with clinically meaningful delays in pregnancy recognition or initiation of prenatal care.
Like others, we found that recognition of pregnancy and initiation of prenatal care varied by pregnancy intention and health insurance status.16,17 We found that women who were not trying to get pregnant recognized their pregnancy and initiated prenatal care 1 week later on average compared with women who were trying to get pregnant. Similarly, we found that uninsured women recognized their pregnancy approximately 1 week later than privately insured women and initiated prenatal care more than 3 weeks later than privately insured women. Their odds of receiving late or no prenatal care were fourfold higher than that of insured women. Using 2000–2004 Pregnancy Risk Assessment Monitoring System data, Ayoola and colleagues15 found that women with no health insurance were much less likely to recognize their pregnancy before 7 weeks of gestation compared with women with private insurance. Rosenberg and colleagues18 used Pregnancy Risk Assessment Monitoring System data from 1998–2000 to examine the effect of prepregnancy Medicaid coverage on initiation of prenatal care among unmarried women receiving public assistance as their sole source of income. They found that women who had prepregnancy Medicaid coverage were more likely that those without coverage to initiate prenatal care in the first trimester (70.0% compared with 47.3%) and that the effect was largest for women with less than a high school education.
Although a major strength of this study is the population-based nature of the data, this work does have some limitations. First, prepregnancy BMI was calculated based on self-reported weight and height before pregnancy; however, these data have been shown to be valid, showing substantial agreement in 95% of women.19 The timing of pregnancy recognition and initiation of prenatal care were also self-reported. Although misclassification of these variables is possible, we would not expect it to be differential by BMI. Furthermore, given that the rate of prenatal care initiation in the first trimester was in the range that we would expect based on national data, social desirability bias (reporting initiation of prenatal care in the first trimester because it is socially desired) was unlikely to have played a major role.20 Finally, women were not queried regarding their menstrual cycle characteristics; so, we were unable to assess whether cycle irregularities among underweight or obese women might have affected the timing of pregnancy recognition.
In summary, we found that prepregnancy BMI was not meaningfully associated with the timing of pregnancy recognition or the initiation of prenatal care. This was despite the fact that women at the extremes of gestational age were less likely to have reported that they planned their pregnancy. These findings are reassuring, because early initiation of prenatal care is the first step in ensuring adequate prenatal care, something that may be particularly important for overweight and obese women who are at heighted risk of adverse maternal and neonatal outcomes.21
1. Practice Committee of American Society for Reproductive Medicine. Obesity and reproduction: an educational bulletin. Fertil Steril 2008;90(suppl):S21–9.
2. Centers for Disease Control and Prevention. Behavioral risk factor surveillance system survey data. Atlanta (GA): U.S. Department of Health and Human Resources, Centers for Disease Control and Prevention; 2012.
3. Castillo-Martinez L, López-Alvarenga JC, Villa AR, González-Barranco J. Menstrual cycle length disorders in 18- to 40-y-old obese women. Nutrition 2003;19:317–20.
4. Douchi T, Kuwahata R, Yamamoto S, Oki T, Yamasaki H, Nagata Y. Relationship of upper body obesity to menstrual disorders. Acta Obstet Gynecol Scand 2002;81:147–50.
5. Grimes DA, Shields WC. Family planning for obese women: challenges and opportunities. Contraception 2005;72:1–4.
6. Holt VL, Scholes D, Wicklund KG, Cushing-Haugen KL, Daling JR. Body mass index, weight, and oral contraceptive failure risk. Obstet Gynecol 2005;105:46–52.
7. Centers for Disease Control and Prevention. PRAMS model surveillance protocol. Atlanta (GA): Centers for Disease Control and Prevention; 2005.
8. Centers for Disease Control and Prevention, RTI International. Pregnancy risk assessment monitoring system (PRAMS) implementation manual. Atlanta (GA): Centers for Disease Control and Prevention; 2007.
9. World Health Organization. Physical status: the use and interpretation of anthropometry, report of a WHO expert Committee. Geneva (Switzerland): World Health Organization; 1995.
10. Hernan MA, Hernández-Diaz S, Robins JM. A structural approach to selection bias. Epidemiology 2004;15:615–25.
11. Howards PP, Schisterman EF, Poole C, Kaufman JS, Weinberg CR. “Toward a clearer definition of confounding” revisited with directed acyclic graphs. Am J Epidemiol 2012;176:506–11.
12. Glymour MM, Greenland S. Causal diagrams. In: Rothman KJ, Greenland S, editors. Modern epidemiology. 3rd ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2008:183–208.
13. Kost K, Landry DJ, Darroch JE. Predicting maternal behaviors during pregnancy: does intention status matter? Fam Plann Perspect 1998;30:79–88.
14. Pirie PL, Lando H, Curry SJ, McBride CM, Grothaus LC. Tobacco, alcohol, and caffeine use and cessation in early pregnancy. Am J Prev Med 2000;18:54–61.
15. Ayoola AB, Nettleman MD, Stommel M. Time from pregnancy recognition to prenatal care and associated newborn outcomes. J Obstet Gynecol Neonatal Nurs 2010;39:550–6.
16. Pagnini DL, Reichman NE. Psychosocial factors and the timing of prenatal care among women in New Jersey's HealthStart program. Fam Plann Perspect 2000;32:56–64.
17. Wiemann CM, Berenson AB, Pino LG, McCombs SL. Factors associated with adolescents' risk for late entry into prenatal care. Fam Plann Perspect 1997;29:273–6.
18. Rosenberg D, Handler A, Rankin KM, Zimbeck M, Adams EK. Prenatal care initiation among very low-income women in the aftermath of welfare reform: does pre-pregnancy Medicaid coverage make a difference? Matern Child Health J 2007;11:11–7.
19. Shin D, Chung H, Weatherspoon L, Song WO. Validity of prepregnancy weight status estimated from self-reported height and weight. Matern Child Health J 2013 [Epub ahead of print].
20. U.S. Department of Health and Human Services. Healthy people 2020. Washington (DC): U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion; 2014.
21. Kotelchuck M. An evaluation of the Kessner Adequacy of Prenatal Care Index and a proposed Adequacy of Prenatal Care Utilization Index. Am J Public Health 1994;84:1414–20.