Although rates of teenage pregnancy and birth in the United States have reached historic lows, they remain the highest among industrialized countries.1–3 Pregnancy among females younger than 15 years in the United States is understudied, despite data indicating that 16% of females are sexually experienced by their 15th birthday4 and that 14,000 females younger than 15 years (1.4/1,000) nationwide became pregnant in 2008.2 There is a paucity of published data regarding correlates and outcomes associated with early teenage pregnancy. Correlates for all teenage pregnancy include being part of a racial or ethnic minority group, living in poverty, being raised by a single parent, having a mother who gave birth as a teenager, and not using contraception at sexual debut.5–8
Mothers aged younger than 15 years experience worse obstetric outcomes and higher rates of pregnancy-related morbidity than mothers aged 15 to 19 years.9,10 Compared with females aged 15 to 19 years, they have higher rates of preterm delivery (21.3% compared with 13.0%), fetal death (4.1 compared with 3.3%), low birth weight (12.6 compared with 8.5%), infant mortality (15.4 compared with 9.1/1,000), and late or no prenatal care (16.1 compared with 7.0%).11 Pregnancy among females aged 10 to 14 years is also associated with stillbirth.12 Young black teenagers are disproportionately affected by these adverse outcomes.13
The purpose of this study is to characterize females with pregnancies occurring before age 15 years in the United States and to compare known correlates of teenage pregnancy and pregnancy outcomes to females reporting their first pregnancy between ages 15 and 19 years using nationally representative data.
MATERIALS AND METHODS
We analyzed data from the 2006 to 2010 National Survey of Family Growth, a nationally representative cross-sectional survey of males and females of reproductive age (15–44 years) in the United States conducted by the National Center for Health Statistics.14 The National Survey of Family Growth uses a multistage area probability sample with oversampling of black, Hispanic, female, and teenage respondents. Data are collected using in-person, in-depth interviews that cover topics including demographic characteristics, pregnancy history, contraceptive use, sexual behavior, sexual relationships, and use of reproductive health services. The survey instrument was extensively pretested in previous National Survey of Family Growth cycles 1995 and 2002, with additional piloting during the first 2 months of sampling in 2006. Participants complete additional and more sensitive questions (eg, history of oral and anal intercourse, human immunodeficiency virus status) using Audio Computer-Assisted Self-Interview. Data from the National Survey of Family Growth are publicly available online for download (www.cdc.gov/nsfg). The accompanying Audio Computer-Assisted Self-Interview data file was provided by the National Center for Health Statistics on request with a signed user agreement. Both data files included designated survey weighting variables for use in analysis.
The 2006 to 2010 National Survey of Family Growth includes responses from 12,279 women (78% response rate) and 10,403 men (75% response rate).15 We restricted our sample to female respondents aged 20 to 44 years who reported first becoming pregnant at age 19 years or younger (n=3,384). Age 20 years was chosen as our lower limit for age at time of interview to capture all respondents who became pregnant before this age and to ensure that the data contained pregnancy outcomes. We used respondents' reported age at first conception to classify them into our two-level outcome variable: pregnancy before age 15 years and pregnancy between ages 15 and 19 years.
We selected independent variables from the National Survey of Family Growth file based on risk factors for teenage pregnancy described, including sociodemographic characteristics (race or ethnicity, nativity, level of education, maternal level of education, poverty level, geographic residence), circumstances of sexual debut (age, partner age, relationship type, contraceptive use, wanting sex), and family characteristics (religion of upbringing, residence with both biological parents at age 14 years, mother's age at first pregnancy). Current poverty level and geographic residence, measured at the time of survey, served as proxies for respondents' family poverty and residence during their teenage years because these variables were not available for the time of conception. We also included the intendedness of the index pregnancy as well as its outcome (live birth, abortion, miscarriage), which were assessed during the core National Survey of Family Growth interview and validated later through Audio Computer-Assisted Self-Interview when women may have felt more comfortable reporting outcomes other than live births.
Correlates of pregnancy before age 15 years, compared with pregnancy between ages 15 and 19 years, were first evaluated in bivariate analysis using Pearson χ2 test and Student t test to assess statistical significance (P<.05) for categorical and continuous variables, respectively. We then performed multivariable logistic regression to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for pregnancy before age 15 years compared with pregnancy between ages 15 and 19 years. We first included all covariates that were significantly associated (P<.05) with the outcome variable in bivariate analysis, plus additional adjustment variables of clinical significance. The final model retained all covariates that remained significant at P<.05 and was adjusted for age at time of interview. Observations with missing data were excluded from analysis. Analyses were performed using Stata 11 using the “svy” prefix to account for National Survey of Family Growth sampling and survey design effects.
This project was submitted to the Institutional Review Board of the University of Chicago Biological Sciences Division, who deemed it exempt from review under 45CFR 46.101(b).4
The 2006 to 2010 National Survey of Family Growth included 7,835 female respondents who were aged 20 to 44 years and had ever been pregnant: 289 (3.4%) had a first pregnancy that occurred before age 15 years; 3,095 (6%) had a first pregnancy that occurred between ages 15 and 19 years; and 4,451 (57.1%) had a first pregnancy that occurred at older ages. Median ages at first pregnancy within the pregnancy before age 15 years group and the pregnancy during age 15 to 19 years group were 14.2 years and 17.9 years, respectively.
In bivariate analysis (Table 1), there were significant group differences in demographic and family characteristics. Women with pregnancies occurring before age 15 years were less likely to attain a high school degree compared with women with pregnancies occurring between ages 15 and 19 years (n=157 [41.7%] compared with n=2,079 [69.0%]; P=.025). There was no difference in proportion of respondents whose mothers attained a high school education (n=172 [62.4%] compared with n=2,018 [65.5%]; P=.46). Women with pregnancies occurring before age 15 years were significantly less likely to report living with both biological parents at age 14 years (n=88 [33.3%] compared with n=1,535 [53.4%]; P<.001) and more likely to have mothers who became pregnant at an age younger than 18 years (n=102 [35.8%] compared with n=824 [25.7%]; P=.02). Compared with females with pregnancies occurring between ages 15 and 19 years, they were also less likely to report using contraception during their first sexual intercourse (n=91 [25.3%] compared with n=1,729 [56.0%]; P<.001) and to report the pregnancy as unintended (n=257 [88.9%] compared with n=2,333 [74.9%]; P=.005). Pregnancy outcomes also varied significantly between groups, with a lower proportion of live births in women with pregnancies before age 15 years.
In multivariable analysis (Table 2), pregnancies occurring before age 15 years were associated with women being Hispanic (OR 1.84; 95% CI 1.07–3.20) or black (OR 2.24; 95% CI 1.47–3.42) compared with white non-Hispanic women. Odds of pregnancies occurring before age 15 years were lower among women who reported being raised in a religion compared with no religion (Catholic: OR 0.32; 95% CI 0.17–0.59; Protestant: OR 0.52; 95% CI 0.27–0.98) or living with both biological parents at age 14 years (OR 0.49; 95% CI 0.33–0.71). Behavior differences at first sexual intercourse remained significant: women with pregnancies occurring before age 15 years had lower odds of contraceptive use at sexual debut (OR 0.29; 95% CI 0.18–0.46) and were three times as likely to have a much older (by 6 years or more) sexual partner (OR 3.34; 95% CI 1.71–6.51).
Pregnancy occurring before age 15 years is an understudied problem. Here, we provide population-level data regarding risk factors and outcomes for these young women demonstrating that they are distinct from those with pregnancy occurring between ages 15 and 19 years. It is particularly important to focus on these young teenagers. Although they comprise only a small fraction of conceptions and births among teenagers in the United States, pregnancies occurring before age 15 years are more likely to end in abortion compared with pregnancies occurring between ages 15 and 19 years. In our study, half of women with pregnancies before age 15 years went on to have a live birth compared with more than two thirds with pregnancies occurring between ages 15 and 19 years. These results are similar to birth rates for teenagers published from vital statistics data.2 Females who deliver at 10 to 14 years of age have the highest rates of inadequate prenatal care, pregnancy-induced hypertension, eclampsia, preterm delivery, low birth weight, and infant death of any age group in the United States.11 Understanding the risk factors that precede these high-risk pregnancies are critically important to developing prevention strategies.
The social effect of pregnancy on women who experience pregnancy before age 15 years compared with pregnancy between ages 15 and 19 years is difficult to assess, because many of the same factors that put women at risk for teenage pregnancy may also be a result of the social and economic strains of a teenage pregnancy. In our study, both groups had similar rates of poverty at the time of interview. However, females with pregnancy before age 15 years had lower rates of attaining a high school education (58.3%) or higher compared with those with a pregnancy between ages 15 and 19 years (69.9%). Although beyond the scope of this study, it is possible that the outcomes of pregnancy before age 15 years (ie, live birth, miscarriage, termination) affect the subsequent socioeconomic trajectories of women.
Many of the established risk factors for pregnancy occurring between ages 15 and 19 years are increased for the women who reported pregnancies occurring before age 15 years. Nationwide, Hispanic and black women have the highest pregnancy rates for 15- to 19-year-olds.2 In our study, females with pregnancy before age 15 years were more likely to be Hispanic or black. There are significant racial and ethnic differences in age of sexual debut, with 3.9% of white and 13.9% of black youth reporting their first intercourse before age 13 years, which may contribute to disparities in birth and pregnancy rates.4 Our findings further align with several studies that have noted that causes of teenage pregnancy are multifactorial and influenced by individual, family, peer, and environmental factors.7,16 In our study, the family environment, specifically religious upbringing and living with two biological parents, conferred a lower risk of pregnancy before age 15 years. More than one quarter of all women in our sample reported that their mother had a pregnancy during her teenage years, and our bivariate analyses echoed the intergenerational pattern of teenage pregnancy observed in previous literature.7 However, this association was no longer statistically significant on adjustment for other model covariates.
Behaviors and circumstances at time of sexual debut have been shown to correlate with risk behaviors, including teenage pregnancy and acquisition of sexually transmitted infections.8 Our analysis found that women with pregnancy before age 15 years are far more likely to engage in high-risk behaviors at sexual debut than women with pregnancies between ages 15 and 19 years. We found a strong relationship between pregnancies occurring before age 15 years and lack of contraceptive use at sexual debut, which has been shown to increase risk of overall teenage pregnancy.17 Alarmingly, nearly 36% of women reporting pregnancies occurring before age 15 years reported that their first sexual partner was older than them by 6 years or more. In all states in the United States, this occurrence qualifies as statutory rape. Equally as concerning is the fact that one in five women in our total sample reported that first intercourse was involuntary or unwanted.
Because pregnancy in young adolescents is a relatively rare event, a large nationally representative data set such as the National Survey of Family Growth provides a robust tool for examining this phenomenon. Our study also has important limitations. Respondents were asked about index pregnancies that occurred from 1 to 35 years before interview, which may affect participant recall. There is potential for nondifferential misclassification because women whose first sex was unwanted or unintended—experiences that were more frequently reported by our cohort of women whose pregnancies occurred before age 15 years—may be more likely to recall age of first partner, contraception use, and relationship with first partner than those whose first intercourse was wanted. The National Survey of Family Growth did not ascertain several salient variables (ie, socioeconomic, family and partner characteristics) at time of first conception—our outcome of interest. Thus, we were unable to assess their relationship with pregnancy before age 15 years. In the analyses in which data reported at the time of interview were used as a proxy for time of first pregnancy, findings should be interpreted with caution, particularly for poverty, which may serve as a risk factor for both teenage pregnancy as well lower education attainment and lower socioeconomic status postpregnancy.
Unlike for births, there is no single national surveillance system for pregnancies. Rather, national reports rely on data reported from vital statistics (births), abortion surveillance, and the National Survey of Family Growth (fetal loss, miscarriage, and abortions) to derive statistics on pregnancies in the United States.2 Because younger women are more likely to have an abortion or miscarriage than older women, national pregnancy data may underestimate pregnancy among the youngest teenagers. Because pregnancy is partially based on National Survey of Family Growth data, and because underreporting of abortion among National Survey of Family Growth participants is a known problem,18 we may be underestimating the number of females younger than 15 years who are becoming pregnant.
We believe that national efforts to quantify teenage pregnancy should be inclusive of those occurring among females younger than 15 years. We can hope to address a problem adequately only after we understand the full extent of it. Unfortunately, most interventions targeting teenage pregnancy do not often stratify by age of the teenager.8 Identification of the youngest females who are at high risk for pregnancies, especially those involved in relationships with older men, is a challenge and may be difficult because these particularly high-risk relationships may be hidden from family and school officials. Although reasons for age disparities regarding first sex are complex, it is incumbent on professionals—physicians, nurses, teachers, social workers—to educate young girls about the dangers of these predatory relationships, create an environment in which young girls have tools to avoid these relationships, and provide support and resources to young women who find themselves in such relationships. Understanding the factors that predict pregnancy among those aged younger than 15 years and those aged 15 to 19 years may help public health, social work, and medical personnel to better identify and target the youngest females to prevent these high-risk pregnancies.
1. Hamilton B, Mathews T, Ventura S. Declines in state teen birth rates by race and Hispanic origin. NCHS Data Brief, No. 123. Hyattsville (MD): National Center for Health Statistics; 2013.
2. Ventura SJ, Curtin SC, Abma JC, Henshaw SK. Estimated pregnancy rates and rates of pregnancy outcomes for the United States, 1990-2008. Natl Vital Stat Rep 2012;60:1–21.
3. Singh S, Darroch JE. Adolescent pregnancy and childbearing: levels and trends in developed countries. Fam Plann Perspect 2000;32:14–23.
4. Finer LB, Philbin JM. Sexual Initiation, contraceptive Use and pregnancy among young adolescents. Pediatrics 2012;131:886–91.
5. Vital signs: teen pregnancy—United States, 1991-2009. MMWR Morb Mortal Wkly Rep 2011;60:414–20.
6. Martinez G, Copen CE, Abma JC. Teenagers in the United States: sexual activity, contraceptive use, and childbearing, 2006-2010 national survey of family growth. Vital Health Stat 2011;23:1–35.
7. Meade CS, Kershaw TS, Ickovics JR. The intergenerational cycle of teenage motherhood: an ecological approach. Health Psychol 2008;27:419–29.
8. Moore K, Miller B, Glei D, Morrison D. Adolescent sex, contraception and childbearing: a review of recent research. Washington, DC: Child Trends, Inc.; 1995.
9. Cooper LG, Leland NL, Alexander G. Effect of maternal age on birth outcomes among young adolescents. Soc Biol 1995;42:22–35.
10. Malabarey OT, Balayla J, Klam SL, Shrim A, Abenhaim HA. Pregnancies in young adolescent mothers: a population-based study on 37 million births. J Pediatr Adolesc Gynecol 2012;25:98–102.
11. Menacker F, Martin JA, MacDorman MF, Ventura SJ. Births to 10-14 year-old mothers, 1990-2002: trends and health outcomes. Natl Vital Stat Rep 2004;53:1–18.
12. Salihu HM, Sharma PP, Ekundayo OJ, Kristensen S, Badewa AP, Kirby RS, et al.. Childhood pregnancy (10-14 years old) and risk of stillbirth in singletons and twins. J Pediatr 2006;148:522–6.
13. Leland NL, Petersen DJ, Braddock M, Alexander GR. Variations in pregnancy outcomes by race among 10-14-year-old mothers in the United States. Public Health Rep 1995;110:53–8.
14. Groves RM, Mosher WD, Lepkowski JM, Kirgis NG. Planning and development of the continuous National Survey of Family Growth. Vital Health Stat 2009;1:1–64.
15. Lepkowski JM, Mosher WD, Groves RM, West BT, Wagner J, Gu H. Responsive design, weighting and variance estimation in the 2006-2010 National Survey of Family Growth. Vital Health Stat 2013;2:1–52.
16. Woodward L, Horwood L, Fergusson D. Risk factors and life processes associated with teenage pregnancy: results from a prospective study from birth to 20 years. J Marriage Fam 2001;63:1170–84.
17. Sexual experience and contraceptive use among female teens—United States, 1995, 2002, and 2006-2010. MMWR Morb Mortal Wkly Rep 2012;61:297–301.
18. Fu H, Darroch JE, Henshaw SK, Kolb E. Measuring the extent of abortion underreporting in the 1995 National Survey of Family Growth. Fam Plann Perspect 1998;30:128–33.