Original Study

Pregnancies in Young Adolescent Mothers: A Population-Based Study on 37

Million Births
Ola T. Malabarey MD 1, Jacques Balayla 1, Stephanie L. Klam MD 1, Alon Shrim MD 2,
Haim A. Abenhaim MD, MPH 1,*
1
Department of Obstetrics and Gynecology, Jewish General Hospital, McGill University, Montreal, Quebec, Canada
2
Department of Obstetrics and Gynecology, Royal Victoria Hospital, McGill University, Montreal, Quebec, Canada

a b s t r a c t
Objectives: Pregnancy in young adolescents is often understudied. The objective of our study was to evaluate the effect of young maternal
age on adverse obstetrical and neonatal outcomes.
Methods: We conducted a population-based cohort study using the Center for Disease Control and Prevention’s Linked Birth-Infant Death
and Fetal Death data on all births in the US between 1995 and 2004. We excluded all births of gestational age under 24 weeks and those
with reported congenital malformations or chromosomal abnormalities. Maternal age was obtained from the birth certificate and relative
risks estimating its effect on obstetrical and neonatal outcomes were computed using unconditional logistic regression analysis.
Results: 37,504,230 births met study criteria of which 300,627 were in women aged !15 years with decreasing rates from 11/1,000 to
6/1,000 over a 10-year period. As compared to women 15 years and older, women !15 were more likely to be black and Hispanic, less
likely to have adequate prenatal care, and more likely to not have had any prenatal care. In adjusted analysis, births to women !15 were
more likely to be IUGR, born under 28, 32, and 37 weeks’ gestation and to result in stillbirths and infant deaths. Prenatal care was
protective against infant deaths in women ! 15 years of age.
Conclusion: Although public health initiatives have been successful in decreasing rates of young adolescent pregnancies, these remain high
risk pregnancies that may benefit from centers capable of ensuring adequate prenatal care.
Key Words: Adolescent pregnancy, Stillbirth, Growth restriction, Contraception

Introduction
In the United States today, 9% of women aged 15 to 19
years become pregnant each year of which 5% will deliver
a baby, 3% will choose to have an induced abortion, and 1%
end in miscarriage or stillbirth.1 Of concern, 12% of adoles-
cent pregnancies are reported to be in females younger than
14 years of age.2 There have been several cohort studies that
addressed adverse perinatal outcomes in adolescent preg-
nancies but none were large enough to address specific
adverse outcomes such as perinatal mortality. Furthermore,
most studies evaluating adolescent pregnancies group all
women under 20 as adolescent in order to ensure sufficient
sample size to establish the effect on obstetrical outcomes.
The objective of our study was thus to use a large dataset
with sufficient power to measure adverse outcomes in
pregnancies of mothers under the age of 15.
Materials and Methods
We conducted a retrospective cohort study using data
from the National Center for Health Statistics (Centers for
Disease Control and Prevention) - Linked Birth-Infant Death
The authors indicate no conflicts of interest.
* Address correspondence to: Haim A. Abenhaim, MD, Jewish General Hospital,
Obstetrics and Gynecology, 5790 Cote-Des-Neiges Road, Pav H, Room 325, Mon-
treal, Quebec, H3S 1Y9
E-mail address: haim.abenhaim@gmail.com (H.A. Abenhaim).
1083-3188/$ - see front matter Ó 2012 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc.
doi:10.1016/j.jpag.2011.09.004
and Fetal Death data files. The data files contain information
on approximately 4 million annual live births of residents
and non-residents in the United States (US). We created
a 10-year cohort using data for the years 1995 to 2004 with
records for the contiguous US, Hawaii, and Alaska. Puerto
Rico, Guam and the Virgin Islands were excluded. We
excluded any delivery prior to 24 weeks of gestation and
those with a congenital anomaly or chromosomal abnor-
mality. This protocol was approved by the Medical Research
Ethics Department of the Jewish General Hospital.
We defined our exposure on the basis of maternal age
which we set as being !15 years of age or 15 years and
older. This information is based on records from the birth
certificate. Among the different available baseline charac-
teristics, we used the Kessner Index,3 which uses the
number and timing of prenatal visits in relation to the
gestational age as markers of adequacy of prenatal care. Our
outcomes included birthweight, small for gestational age
(SGA), intrauterine growth restriction (IUGR), stillbirths,
infant deaths, preterm labor and rates of cesarean section
(CS) deliveries. We categorized birthweights as being very
low birthweight (VLBW) if the birthweight was 1500 gm or
less; low birthweight (LBW) if the birthweight was 2500 gm
or less; macrosomia for birthweights greater and equal
to 4000 gm; or normal birthweight (NBW) for all other
birthweights. SGA was defined as birthweight ! 10th
percentile, whereas IUGR was defined as birthweight ! 3rd
percentile using previously published curves standardized
for gestational age and gender.4 For purposes of our study,
 O.T. Malabarey et al. / J Pediatr Adolesc Gynecol 25 (2012) 98e102
stillbirth was defined as any death before expulsion from
the mother beyond 24 week gestation and infant death was
defined as any postnatal fatality taking place between birth
and the first year of life. Prematurity was defined as any
delivery before 37 completed weeks of pregnancy.
Our analysis was conducted in 4 steps. First, we carried
out descriptive statistics to summarize the characteristics of
our population stratified according to maternal age group.
We then investigated whether there was a trend in the rate
of pregnancies below the age of 15 over the 10-year study
period by calculating the change in rate of young adolescent
pregnancies over time and used a chi-square analysis to
obtain the corresponding P value. To examine the rela-
tionship between maternal age and outcomes, we used
unconditional logistic regression models to estimate odds
ratios (OR) along with the 95% confidence intervals (CI) of
our outcomes of interest. We examined the adjusted effects
of young adolescent pregnancies on the aforementioned
outcomes through the use of a model that adjusted for
variables which were considered true confounders to the
exposure-outcome relationship, specifically race, twins and
number of prior births. Finally, in order to identify predic-
tors of fetal and infant death among young adolescents, we
examined the effect of specific characteristics on fetal and
infant death in an analysis restricted to young adolescents.
All analyses were conducted using SAS enterprise guide 4.2
(SAS Institute, Cary, NC).
Results
Over the 10 year period between 1995 and 2004 there
were 40,529,306 deliveries of which 37,504,230 remained
after exclusion of births to infants with congenital malfor-
mations and birth to infants of gestational age less than 24
weeks. Of all births, 0.80% were born to women aged less
than 15 years of age while 99.19% were in women aged 15
years and older (Fig. 1). Over the study period, the rate of
teenage pregnancies below the age of 15 dropped steadily
from 1.1% in 1995, to 0.61% in 2004 for a reduction in early
adolescent birth rates of 0.005 births/year (Fig. 2).
All births and fetal deaths between
1995 and 2004
(n = 40,529,306)
Eligible Subjects
(n = 37,504,230)
≤15 years of age >15 years of age
(n = 300,627) (n = 37,203,603)
† Overlap of 292,126 cases which consisted of anomalies below 24 weeks of gestation.
Fig. 1. Study flow chart.Ă
99
Table 1 compares baseline characteristics between
women aged less than 15 with those aged 15 years and
older. Women less than 15 years of age were more likely to
be Black and Hispanic than were women aged 15 and older.
Pregnant women less than 15 years of age were also more
likely not to have had any prenatal care and more likely to
have intermediate or inadequate care, as per the Kessner
Index, than women aged 15 years and older.
In adjusted analysis, women less than 15 years of age
were significantly more likely to have a baby that was SGA,
IUGR, VLBW, LBW, and born prematurely. These women
were also more likely to have stillbirths and infant deaths as
compared to women aged 15 years and older. Rates of
cesarean delivery were lower in women less than 15 years
of age as compared to women aged 15 and older (Table 2).
A total of 131,896 stillbirths occurred over the study
period with the incidence being higher in women less than
15 years of age (0.56% vs 0.35%, P ! 0.0001). There were also
154,456 infant deaths over the 10-year period with the
incidence among women less than 15 years of age being
significantly higher than in women aged 15 and older (0.86
vs 0.41, P ! 0.0001).
Baseline and clinical characteristics were then evaluated
for their independent effect on stillbirths and infant deaths
among women aged under 15 years (Table 3). Our results
suggested that among women aged !15 years, that being
Black, having twin pregnancies, and having had prior births
increased the risk of stillbirths and infant deaths. On the
other hand, we found that having had any prenatal care was
protective against having stillbirths and fetal deaths.
Discussion
Teen birth rates have decreased steadily over the past
decade5 and although the exact reason is uncertain, it has
been speculated that the increasing accessibility of effective
contraception for adolescents is the main reason for the
decline in adolescent births.6 Nevertheless, adolescents,
particularly young adolescents, typically represent
a different demographic and thus the risks of obstetrical
Exclusions†:
-All births and fetal deaths below 24
weeks of gestation (n = 501,913)
-All births and fetal deaths with a
reported congenital or chromosomal
abnormality (n = 2,815,289)
100 O.T. Malabarey et al. / J Pediatr Adolesc Gynecol 25 (2012) 98e102

1200 that among all adolescent mothers, 22.73% were unwanted/

unplanned pregnancies. The effect of having an unwanted
1000
pregnancy has been linked to lack of access to health
services provided by public medical insurance.7 It has also
800
been found that young women who give birth as adoles-
600 cents are more likely to be of low socioeconomic status and
end up having poor school performance, and low maternal
400 education.5 Of concern, in more than one study, childhood
sexual abuse has been found to increase the incidence of
200 subsequent adolescent pregnancy.8,9
Our study showed a decrease in the incidence of young
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 adolescent pregnancies between the years 1995 and 2004.
Year The decline in the incidence of young adolescent pregnan-
Fig. 2. Rate of pregnancies under 15 in the United States between 1995 and 2004.Ă cies demonstrated during this 10-year study period may be
attributed to 2 main factors: increasing availability of
contraceptives and increasing accessibility to therapeutic
and neonatal outcomes may be different than older women abortions. In a study looking at the use of contraception
of childbearing age. Given that only a small proportion of all between the years 1995 and 2002, investigators reported
births are to adolescent women, the availability of large that the contraceptive risk index declined 34% overall and
enough cohorts of young women for studies on pregnancies 46% among adolescents. The overall pregnancy risk index
in young adolescent women has been limited by inadequate declined 38%, with 86% of the decline attributable to
sample size. The purpose of our study was to evaluate the improved contraceptive use.10 Although these results are
effect of young maternal age on birthweight, risk of small promising, not all sexually active adolescents use contra-
for gestational age infants, intrauterine growth restriction, ception, and even those who do use contraception some-
stillbirth, infant death, prematurity, and rate of cesarean times use it incorrectly.11 In a cross-sectional study, out of
deliveries. Our results suggest that the rate of adolescent 220 adolescent mothers, 91.3% were familiar with “the pill”
pregnancies has declined over the 10-year study period as a contraceptive method; 84.72% knew about the IUD and
from 1995 to 2004, and that pregnancies among young 63.68% knew about the condom. However, only 35% of them
adolescents are associated with an increased risk of fetal were actually using an effective contraceptive method six
and infant mortality. weeks after delivery.7
Young adolescent pregnancies are an important subset of Another explanation for the declining rates of teenage
all pregnancies as they are not particularly reflective of the pregnancies, is the increasing accessibility to therapeutic
general population. In 2003, Nu n~ ez-Urquiza et al3 reported abortions which has been reported to be as high as 56% of
all teenage pregnancies.12 It has previously been shown that
Table 1
increased ease of access to therapeutic abortions has been
Cohort Baseline Characteristics According to Maternal Age associated with declining rates of births among young
!15 years old $15 years old
women age 14 and less.13 It has also been shown that teens
n 5 300,627 (%) n 5 37,203,603 (%) who terminated their pregnancies were more likely to be in
Age school or working one year later, and this also resulted in an
!15 100 0 increase in reported self-esteem.14
15e19 0 11.07 Over the last decade, multiple programs have been
20e24 0 25.10
25e29 0 27.25 implemented to educate adolescents, and in turn, reduce the
30e34 0 23.34 incidence of teenage pregnancies. An example is the Cal-
35e39 0 10.97 ifornia’s Adolescent Sibling Pregnancy Prevention Program,
40e44 0 2.16
O45 0 0.11 which targets what is believed to be a high risk group:
Race siblings of adolescents who have been pregnant or are
White 29.97 60.62
parents. Female program clients had a significantly lower
Black 37.51 14.64
Native 1.61 0.86 pregnancy rate than comparison females during the evalu-
Hispanic 29.30 19.11 ation period (4% vs 7%), as well as a lower rate of sexual
Other 1.58 4.75
initiation (7% vs 16%).15 The TeenSTAR Program is another
Plurality
Singleton 98.92 96.98 example of an effective method implemented in an effort to
Multiple 1.06 3.00 reduce unintended adolescent pregnancies that was applied
Prenatal visit 92.25 95.90
successfully in a high school in Santiago, Chile.16 Founded in
Adequacy of care*
Adequate 43.20 72.48 1979, the Massachusetts Alliance on Teen Pregnancy is
Intermediate 35.04 17.76 another example. It is a private, non-profit advocacy agency
Inadequate 14.7 5.08
working on issues associated with early child bearing,
Unknown 7.03 4.66
No. of prior births including teen pregnancy prevention and services for teen
0 93.46 32.61 parents and their children. The Alliance is a statewide
1 6.07 29.52
organization with a small, highly effective staff that provides
2 0.45 37.86
a casual, flexible, and supportive working environment.
 O.T. Malabarey et al. / J Pediatr Adolesc Gynecol 25 (2012) 98e102 101

Table 2
Effect of Maternal Age !15 years on Obstetrical and Neonatal Outcomes Adjusting for Baseline Characteristics

!15 years old N 5 300,627 (%) $15 years old N 5 37,203,603 (%) Unadjusted OR (95% CI) Adjusted OR (95% CI) P value

Birthweight
VLBW 2.31 1.28 1.76 (1.72e1.80) 1.47 (1.43e1.50) !0.0001
LBW 9.24 6.14 1.47 (1.45e1.49) 1.37 (1.36e1.39) !0.0001
NBW 84.62 82.84 Reference Reference Reference
Macrosomic 3.83 9.74 0.38 (0.37e0.39) 0.51 (0.51e0.52) !0.0001
SGA 11.44 7.69 1.60 (1.59e1.62) 1.14 (1.12e1.15) !0.0001
IUGR 7.10 4.27 1.80 (1.77e1.82) 1.21 (1.20e1.23) !0.0001
Stillbirth 0.56 0.35 1.60 (1.53e1.68) 1.31 (1.24e1.38) !0.0001
Infant Death 0.86 0.41 2.10 (2.02e2.18) 1.87 (1.80e1.95) !0.0001
Stillbirth/Infant Death 1.42 0.76 1.88 (1.82e1.94) 1.68 (1.63e1.73) !0.0001
#28 weeks 1.30 0.53 2.48 (2.40e2.56) 1.90 (1.83e1.96) !0.0001
#32 weeks 4.06 1.72 2.40 (2.36e2.45) 2.01 (1.97e2.05) !0.0001
#37 weeks 18.41 11.33 1.76 (1.75e1.78) 1.71 (1.69e1.72) !0.0001
Delivery method
Vaginal 85.25 74.08 Reference Reference Reference
CS 14.36 23.33 0.85 (0.84e0.86) 0.54 (0.53e0.55) !0.0001

VLBW, Very Low Birthweight (#1500 grams); LBW, Low Birthweight (O 1500 ! 2500 grams); NBW, Normal Birthweight ($ 2500, ! 4000 grams); Macrosomic, $ 4000
grams; SGA, Small for Gestational Age; IUGR, Intrauterine Growth Restriction; CS, Cesarean section

The concern about the adolescent pregnant patient is
the noted higher incidence of poor obstetrical and neonatal
outcomes that has been reported in multiple studies.17e21
It is unclear if the increased risk of adverse events is due
to differences in access to care or to an inherent increased
risk in adolescent pregnancies. Our study showed an
increased risk of stillbirths and newborn deaths in the
pregnant adolescent and we believe that the observed
adverse maternal and neonatal outcomes are likely
a combination of both differences in risk factors as well as
an inherent increased risk. Our reasoning behind this is
that our observed association between young adolescent
pregnancies and adverse outcomes not only was signifi-
cant when adjusting for baseline risk factors but also when
adjusting for downstream variables such as prenatal care
and Kessner index suggesting that there is an inherent risk
of adverse outcomes that is seen in young adolescents. This
idea has been suggested in previous studies. Orvos et al17
showed that although 65% of adolescent pregnant
patients had adequate antenatal care, 18% delivered before
37 weeks, and 16% showed signs of intrauterine growth
restriction. In a similar study, Eure et al18 showed that
young adolescents were significantly more likely to have
preeclampsia, eclampsia, preterm delivery, low-birth-
weight delivery, and very-low-birth-weight delivery. The
adverse outcomes have not been limited to prematurity
and neonatal events. In a study on postpartum complica-
tions, adolescent pregnant patients were found to have
higher rates of endometritis, dehiscence of surgical
wounds, and pyelonephritis.21
There are several limitations to our study. Our population
cohort included females over 40 years of age in the (older
than 15 years old) group. One of the concerns that may arise
from including this segment of the population is that we
may attenuate the effect of adverse outcomes since these
women are known to be particularly at high risk for adverse
events. However, we chose not to remove them from our
cohort, since we felt that by removing any specific group we
might inadvertently introduce bias. Moreover, the incidence
of pregnancy beyond the age of 40 years was very low in our
cohort; thus the inclusion in our cohort likely did not have
considerable effect on our outcomes. Another limitation is
that we did not have growth curves for the classification of
SGA and IUGR that were specific to the United States, rather,
we used a reference based on a population-based Canadian
study specific for singleton pregnancies. While this may
have a minor effect on generalizability, we do not believe
that it is a major limitation of our study.

Table 3
Predictors of Fetal and Infant Mortality among 300,627 Women Aged ! 15 Years

Stillbirths (n 5 1669) % Adjusted OR (95% CI) Infant deaths (n 5 2595) Adjusted OR (95% CI) Composite (n 5 4264) Adjusted OR (95 % CI)
%

Race
White 29.24 Reference 31.68 Reference 30.72 Reference
Black 46.91 1.14 (1.01e1.28) 44.28 1 (0.90e1.11) 45.31 1.08 (1.00e1.16)
Native 1.14 0.72 (0.45e1.15) 1.43 0.94 (0.66e1.34) 1.31 0.78 (0.60e1.03)
Hispanic 21.39 0.67 (0.58e0.78) 21.27 0.62 (0.55e0.71) 21.32 0.65 (0.60e0.71)
Other 1.32 0.77 (0.49e1.19) 1.35 0.65 (0.43e0.97) 1.34 0.77 (0.59e1.01)
Plurality
Singleton 96.35 Reference 93.76 Reference 94.77 Reference
Multiple 3.66 2.12 (1.61e2.78) 6.24 6.0 (4.9e7.3) 5.23 4.00 (3.44e4.65)
Prenatal visit
None 30.08 Reference 18.38 Reference 22.96 Reference
Present 69.92 0.21 (0.19e0.24) 81.62 0.45 (0.37e0.53) 77.04 0.29 (0.27e0.31)
Prior births
0 82.83 Reference 88.80 Reference 86.51 Reference
1 14.72 2.37 (2.05e2.74) 10.03 1.26 (1.10e1.45) 11.83 1.68 (1.52e1.86)
2 2.45 5.03 (3.61e7.00) 1.17 1.74 (1.19e2.53) 1.66 2.97 (2.30e3.82)
102 O.T. Malabarey et al. / J Pediatr Adolesc Gynecol 25 (2012) 98e102
On the other hand, this study has a number of strengths.
First, to our knowledge, this is the largest study to have
investigated the effect of young maternal age on specific
pregnancy and infancy outcomes. With 37.5 million births,
this study has unprecedented power for the detection of the
risk of stillbirth and infant mortality that no other study has
been able to appropriately evaluate. Second, the data that
was used is population based making the information
collected unlikely to be biased with respect to our study
question.
In conclusion, the rates of pregnancies in young adoles-
cents have been on the decline in the United States.
Although early adolescence itself is likely a risk factor for
adverse outcomes, these pregnancies also tend to have
suboptimal prenatal care which may potentiate the adverse
risks that are observed. In conjunction with current public
health initiatives focusing on providing access to both
contraception and therapeutic abortions, efforts should be
made towards the development of programs that focus on
ensuring adequate care and support for those young
adolescents who do choose to continue their pregnancies in
order to decrease the risk of adverse outcomes that can
occur in this high risk population.
References
1. Darroch JE: Adolescent pregnancy trends and demographics. Curr Womens
Health Rep 2001; 1:102
2. Goldenberg P, Figueiredo Mdo C, Silva Rde S: [Adolescent pregnancy, prenatal
care, and perinatal outcomes in Montes Claros, Minas Gerais, Brazil]. Cad
Saude Publica 2005; 21:1077
3. 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
4. Kramer MS: A new and improved population-based Canadian reference for
birth weight for gestational age. Pediatrics 2001; 108:E35
5. Elfenbein DS, Felice ME: Adolescent pregnancy. Pediatr Clin North Am 2003;
50:781
6. Treffers PE: [Teenage pregnancy, a worldwide problem]. Ned Tijdschr Geneeskd
2003; 147:2320
7. Nun~ ez-Urquiza RM, Herna ndez-Prado B, Garcıa-Barrios C, et al: [Unwanted
adolescent pregnancy and post-partum utilization of contraceptive methods.].
Salud Publica Mex 2003; 45(Suppl 1):S92
8. Noll JG, Shenk CE, Putnam KT: Childhood sexual abuse and adolescent
pregnancy: a meta-analytic update. J Pediatr Psychol 2009; 34:366
9. Pallitto CC, Murillo V: Childhood abuse as a risk factor for adolescent pregnancy
in El Salvador. J Adolesc Health 2008; 42:580
10. Santelli JS, Lindberg LD, Finer LB, et al: Explaining recent declines in adolescent
pregnancy in the United States: the contribution of abstinence and improved
contraceptive use. Am J Public Health 2007; 97:150
11. Lindberg CE: Emergency contraception for prevention of adolescent pregnancy.
MCN Am J Matern Child Nurs 2003; 28:199
12. Deligeoroglou E, Christopoulos P, Creatsas G: Pregnancy and abortion in Greek
adolescent gynecologic clinics. Akush Ginekol (Sofiia) 2004; 43(Suppl 4):37
13. Shelton JD: Very young adolescent women in Georgia: has abortion or
contraception lowered their fertility? Am J Public Health 1977; 67:616
14. Bailey PE, Bruno ZV, Bezerra MF, et al: Adolescent pregnancy 1 year later: the effects
of abortion vs. motherhood in Northeast Brazil. J Adolesc Health 2001; 29:223
15. East P, Kiernan E, Chavez G: An evaluation of California’s Adolescent Sibling
Pregnancy Prevention Program. Perspect Sex Reprod Health 2003; 35:62
16. Cabezon C, Vigil P, Rojas I, et al: Adolescent pregnancy prevention: An
abstinence-centered randomized controlled intervention in a Chilean public
high school. J Adolesc Health 2005; 36:64
17. Orvos H, Nyirati I, Hajdu J, et al: Is adolescent pregnancy associated with
adverse perinatal outcome? J Perinat Med 1999; 27:199
18. Eure CR, Lindsay MK, Graves WL: Risk of adverse pregnancy outcomes in young
adolescent parturients in an inner-city hospital. Am J Obstet Gynecol 2002; 186:918
19. Mariotoni GG, Barros Filho AA: [Is adolescent pregnancy a risk factor for low
birth weight?]. J Pediatr (Rio J) 1998; 74:107
20. Pavlova-Greenfield T, Sutija VG, Gudavalli M: Adolescent pregnancy: positive
perinatal outcome at a community hospital. J Perinat Med 2000; 28:443
21. Pereira LS, Lira Plascencia J, Ahued Ahued R, et al: [Maternal morbidity in
adolescent pregnancy]. Ginecol Obstet Mex 2002; 70:270