Original Study Periconceptional Risk Factors for Birth Defects among Younger and Older Teen Mothers Amy P. Case PhD 1,*, Adrienne T. Hoyt MPH 1, Mark A. Canfield PhD 1, Anna V. Wilkinson PhD 2 1 2
Texas Department of State Health Services, Austin, TX The University of Texas School of Public Health, Austin Regional Campus & Michael and Susan Dell Center for Healthy Living, Austin, TX
a b s t r a c t Study Objectives: We sought to determine whether selected periconceptional health behaviors that influence risk for birth defects differ between older and younger adolescents and whether pregnancy intention predicts more positive preconception health behaviors among teens. Design and Participants: We analyzed interview responses from 954 adolescent control group participants from the National Birth Defects Prevention Study who delivered live infants during 1997-2007. Main Outcome Measures: Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were calculated for factors of interest by age categories (13-15, 16-17, and 18 years, relative to 19 years). To construct a composite periconceptional behavior index, we summed the following healthy behaviors: nonsmoker, nondrinker, folic acid supplementation, and eating 5 or more servings of fruits and vegetables per day. Results: Analyses indicated that women in the youngest group (13-15 years of age) were more likely to be Hispanic (aOR 2.83, 95% CI 1.405.70) and less likely to engage in some unhealthy pregnancy-related behaviors compared with 19-year-olds, such as smoking (aOR 0.45, 95% CI 0.20-0.99) and being overweight or obese (aOR 0.32, 95% CI 0.16-0.61). However, they were also less likely to have taken periconceptional folic acid (aOR 0.44, 95% CI 0.21-0.90). About one-third of teen mothers indicated that their pregnancies had been intended. Among 18- and 19-year-olds, this predicted a higher mean value for the composite periconceptional behavior index (2.30 versus 1.94, P # .01). Conclusions: Teen mothers are not a homogeneous group. Each age subgroup presents varied demographic and behavioral factors that put them at varying levels of risk for birth defects. Furthermore, caregivers should not assume that teens do not plan pregnancies or that they need not be informed of the importance of periconceptional health. Key Words: Adolescent, Preconception care, Health behavior, Tobacco use, Adolescent behavior, Drinking behavior, Fruit, Vegetables, Folic acid, Congenital anomalies, Health surveys
Introduction
Although birth rates for US teens continue to fall, about 400,000 babies are delivered each year to mothers aged 20 years or younger; about 14% of those are born to mothers younger than 17 years.1 Several types of birth defects, including gastroschisis and renal agenesis, disproportionally affect the offspring of younger mothers,2e5 and there is some evidence to suggest that births to very young teens disproportionately contribute to this burden,6,7 underscoring the possibility that they may have different risk factors compared with older adolescents. Findings on the contribution of physiological immaturity to poorer birth outcomes among offspring born to adolescents have been inconsistent,8,9 suggesting that differing periconceptional health and behaviors play a part.10 However, while it is clear from studies of within-group differences of nonpregnant Funding for this study was provided in part by the Texas Center for Birth Defects Research and Prevention, at the Texas Department of State Health Services, through cooperative agreement No. U50/CCU613232 from the Centers for Disease Control and Prevention (CDC). No manifest or potential conflicts of interest are noted. * Address correspondence to: Amy P. Case, PhD, Birth Defects Epidemiology & Surveillance Branch, Texas Department of State Health Services, P.O. Box 149347, Austin, TX 78714-9347, USA; Phone: (512) 565-1687; fax: (512) 458-7330 E-mail address: [email protected] (A.P. Case).
teens that each age group presents a different behavioral and physiological profile, most etiological birth defect studies treat teen mothers as homogeneous regardless of their age. Therefore, it is important for future studies of birth defects risk factors to better understand whether and how relevant periconceptional health and behaviors vary within this age group. Several patterns about the health behaviors of interest have been observed in the literature that inform the basis for this study: First, these behaviors tend to differ between nonpregnant older and younger teens11,12; second, younger and older teen mothers and their adult counterparts exhibit different behaviors during pregnancy13; and, third, in general, mothers report several different periconceptional behaviors depending on whether the pregnancy was planned.14,15 However, virtually all studies of preconception health treat teens as a single stratum, and only occasionally are younger teens represented in these studies.16e18 Although there are notable exceptions,12,19 scant attention has been directed to preconception behaviors and exposures among the youngest mothers and whether they differ from older teens. Finally, the extent to which pregnancy intention exerts a similar influence among teens has not been explored.
1083-3188/$ - see front matter Ó 2015 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpag.2014.09.004
264
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Although it is often assumed that pregnancies among teen mothers are universally unintended, research indicates that between 7% and 35% of these conceptions were planned or wanted.20e22 Although the proportion varies by age group, even among the youngest teens, planned pregnancy is not unknown.20 Pregnancy intention is a particularly important characteristic when considering preconception behaviors,15,23,24 because the intent to become pregnant may act as a strong motivator to change unhealthy behaviors to improve the prospects of that pregnancy. Although teen mothers have exhibited improved self-care behaviors during and after pregnancy,21 it is not clear whether adolescents, whose motives, attitudes, and awareness toward pregnancy planning may be particularly complex,25 tend to adopt healthier behaviors before an intended conception occurs. Many health promotion strategies rely on tailoring to specific characteristics of a population, including attitudes and behavioral intentions. Further, health behaviors tend to cluster in the same individuals26; conversely, unhealthy behaviors co-vary considerably and may cluster in different patterns between adolescent age groups.27 Identifying how risk and protective factors cluster will further inform the development of health promotion strategies. The primary aim of this study was to determine whether periconceptional health conditions and behaviors, particularly those relevant to birth defects prevention, vary among teen mothers by age subgroups. An additional aim was to investigate whether selected behaviors vary according to whether teen mothers indicated having planned their pregnancies and whether those patterns also vary for older and younger teens.
Subjects and Methods Study Population
We conducted a nested cross-sectional analysis of responses given by control participants in the National Birth Defects Prevention Study (NBDPS). The NBDPS is a multicenter case-control study with 3 major components: (1) collection and categorization of information regarding the index infant and maternal demographics obtained from state birth defects registries, (2) a 1-hour computer-assisted telephone interview of the mother, and (3) collection of DNA from maternal, paternal, and child buccal cell samples. The interview, which must be concluded within 2 years after the index delivery, includes questions regarding periconceptional health, such as dietary habits, exposures during pregnancy, occupation, and health status around the periconceptional period. Control mothers gave birth to liveborn infants selected from the same base population as cases but their children had no major malformations and they had an estimated date of delivery occurring in the same year as cases. They are selected from either birth certificates or birth hospitals by using a stratified random sampling scheme.28 Responses from control mothers who were younger than 20 years and gave birth to a child without a birth defect covered by the NBDPS between 1997
and 2007 form the basis of the present analysis. The NBDPS has been described in greater detail elsewhere.28,29 The NBDPS was approved by the Office of Management and Budget and the appropriate institutional review boards at the Centers for Disease Control and Prevention and each participating site. Data Analyses
Variables of interest were behaviors or characteristics that are known risk factors for birth defects, with particular attention to behaviors that, based on a review of the literature, could be expected to vary among the age groups of interest (eg, alcohol use, high blood pressure). A priori power calculations were performed for the selected variables, which indicated that maternal smoking, environmental tobacco exposure, drinking, high blood pressure, obesity, folic acid use, and maternal nativity were within the capacity of this analysis to identify meaningful associations with the age groups of interest. In contrast, medications other than oral contraceptives and prepregnancy diabetes were eliminated from further analyses due to insufficient power. Responses regarding the 3 months before and 1 month after conception were included for behavioral variables (eg, folic acid supplementation, binge drinking); dietary recall, however, refers to average consumption during the 1 year before pregnancy. (A detailed description of variables is given in Table 1.) We calculated c2 values or 2-sided Fisher's exact estimates in cases where cell sizes were less than 5 for our risk factors of interest and for the 4 age categories (13-15, 16-17, 18, and 19 years). Three separate binary logistic regression models were run for each of the 3 teen subcategories against 19-year-olds (ie, 13- to 15-year-olds, 16- to 17-yearolds, and 18-year-olds versus 19-year-olds) for all maternal risk factors. Both crude and adjusted multivariate models were assessed adjusting for all other factors considered significant in at least 1 of the crude models. All statistical calculations were performed with SAS software, version 9.3 (SAS Institute Inc, 2011). Recognizing that adolescent risk behaviors often occur together, we constructed a composite periconceptional behavior index (CPBI) by computing mean behaviors by age group for the following reported behavioral characteristics, which are known to play a role in preventing certain birth defects. Cumulative responses to 4 known risk/protective factors for birth defects were grouped by age and number of positive responses (smoking and drinking were inverted so that the positive corollary could be compared to the protective factors). The 4 behaviors were (a) not smoking (3 months before to 1 month after conception), (b) not drinking any alcohol (3 months before to 1 month after conception), (c) consuming at least 5 servings of fruits or vegetables per day in the year before conception (proxy for a balanced diet),30 and (d) taking folic acid supplements (3 months before to 1 month after conception). We used the CPBI in 2 analyses used to determine if composite behavior patterns differed by age groups: (a) cumulative percent reporting each level of the CPBI (potential range 0-4,
Table 1 Description of Selected Periconceptional Health and Behavioral Variables Variable
Responses Used in Analysis of This Variable*
Were you ever told by a doctor that you had diabetes (including gestational diabetes), sometimes called sugar diabetes or diabetes mellitus? If Yes: What type of diabetes did you have?
High blood pressure during pregnancy
Did you have pregnancy-related high blood pressure when you were pregnant with (this pregnancy)?
Pregnancy Intention
At the time that you became pregnant (with this pregnancy), did you (a) want to become pregnant then; (b) want to wait until later, or (c) not want to become pregnant at all? 1. Did you drink any wine, beer, mixed drinks, or shots of liquor three months prior to conception to the end of pregnancy? If Yes: Which months/days did you drink alcoholic beverages? During which months did you drink any alcoholic beverages? 2. In the (third/second/first month before pregnancy, first/second/third month of pregnancy, second/third trimester), on average, how many days/month did you drink alcoholic beverages? On those days that you drank alcoholic beverages, on average, how many drinks did you have per day? What was the greatest number of drinks you had on 1 occasion? From 3 months before you became pregnant to the end of your pregnancy, did you smoke cigarettes? If Yes, during which months did you smoke? Did anyone in your household smoke cigarettes in your home between 3 months before you became pregnant to the end of your pregnancy? During which months did someone smoke in your home? From 3 months before you became pregnant to the end of your pregnancy, did you take any of the following single vitamins or minerals? (W)hat your usual diet was like (in the year) before you were pregnant? For seasonal foods, fruits, and vegetables, you can average over the 6 months prior to (this pregnancy). For foods that you ate less than once a month, you can report as never or none. (Interviewer reads list of food items, and asks, “How often, on average, did you use.”)
Was it: Gestational (diagnosed during pregnancy only) Insulin-dependent diabetes, also called type 1 or juvenile; noneinsulin-dependent diabetes, also called type 2 or adult onset (diagnosed before pregnancy only) Yes (diagnosed during pregnancy only) No Intendedy Unintended (Mistimed, ambivalent, or unwanted)y Any: Yes, B3-P1 Binge: $4 drinks on any occasion, B3-P1
Alcohol use
Smoking Environmental tobacco smoke (home) Folic acid $5 Fruits/vegetables per day
Yes: B3-P1 No Yes: B3-P1 Yes to folic acid: B3-P1
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Question Summary
Diabetes
Average 5 servings/day of any combination of fruit or vegetable items (including orange and tomato juices)
* B signifies month(s) prior to conception (eg, B1); P signifies month(s) after conception (eg, P3). y The coded variable actually uses responses to several questions to classify pregnancy intention, including questions about contraceptive use at the time of conception. See Dott et al, “Association between pregnancy intention and reproductive-health related behaviors before and after pregnancy recognition, National Birth Defects Prevention Study, 1997-2002” (2010) for more detail.
265
266
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Table 2 Selected Demographic and Periconceptional Characteristics by Age at Conception, NBDPS Control Mothers (1997-2007)* Younger than 20 Years at Conception (N 5 954) Demographic and Periconceptional Characteristics
All Teens 13-15 n
Totalz Maternal ethnicity NH white NH black Hispanic Other Nativity Foreign born US born Pregnancy intended Yes No Body mass index, kg/m2 Underweight (!18.5) Normal Overweight-obese (25þ) Diabetes Prepregnancy Yes No Gestational Yes No High blood pressure during pregnancy Yes No Alcohol usejj Any None Binge drinkingjj Binge None Cigarette smokingjj Any None Environmental tobacco smoke exposurejj Any None Folic acid supplementationjj Any None $5 Fruits/vegetables per day{ Yes No
Py
Age at Conception, y 16-17 %
n
18 %
308
n
19 %
247
n
%
954
110
289
333 163 401 57
26 22 55 7
23.6 20.0 50.0 6.4
91 49 148 20
29.5 15.9 48.1 6.5
90 44 98 15
36.4 17.8 39.7 6.1
126 48 100 15
43.6 16.6 34.6 5.2
.05
173 747
13 91
12.5 87.5
53 243
17.9 82.1
45 198
18.5 81.5
62 215
22.4 77.6
.31
306 648
24 86
21.8 78.2
90 218
29.2 70.8
90 157
36.4 63.6
102 187
35.3 64.7
.84
92 546 258
13 76 17
12.3 71.7 16.0
30 182 79
10.3 62.5 27.2
23 147 63
9.9 63.1 27.0
26 141 99
9.8 53.0 37.2
.02
4 950
1 109
0.9 99.1
1 307
0.3 99.7
2 245
0.8 99.2
0 289
0.0 100.0
.26x
31 923
1 109
0.9 99.1
11 297
3.6 96.4
9 238
3.6 96.4
10 279
3.5 96.5
.55x
94 860
14 96
12.7 87.3
31 277
10.1 89.9
23 224
9.3 90.7
26 263
9.0 91.0
.97
247 675
18 87
17.1 82.9
62 234
20.9 79.1
72 172
29.5 70.5
95 182
34.3 65.7
.12
122 695
8 91
8.1 91.9
32 235
12.0 88.0
37 177
17.3 82.7
45 192
19.0 81.0
.53
242 681
15 90
14.3 85.7
72 225
24.2 75.8
73 171
29.9 70.1
82 195
29.6 70.4
.86
389 533
48 56
46.2 53.8
104 193
35.0 65.0
113 130
46.5 53.5
124 154
44.6 55.4
.69
233 709
14 93
13.1 86.9
73 230
24.1 75.9
64 182
26.0 74.0
82 204
28.7 71.3
.66
204 750
24 86
21.8 78.2
63 245
20.5 79.6
45 202
18.2 81.8
72 217
24.9 75.1
.05
* Excludes Iowa, the Centers for Disease Control and Prevention, and New York for the years 2004-2007. y Mantel-Haenszel c2 or Fisher's exact test if indicated calculated for all age groups by respective demographic characteristics and risk factors. z Totals for some categories may vary due to missing values. x Two-sided Fisher's exact test. jj Maternal exposure 3 months before conception through the first month of pregnancy (see Table 1). { Participants questioned on how often they consumed a particular food during the year before becoming pregnant with index birth (see Table 1).
with 4 representing all behaviors) were calculated by age groups and (b) mean number of behaviors for each group was calculated for both intended and unintended pregnancies. A Student t statistic was computed to compare the mean values. Results
Participants for the current analysis were selected from the larger NBDPS database based on the following criteria: (a) live birth between 1997 and 2007, (b) no birth defect noted, and (c) the mother was younger than 20 years when she gave birth. Mothers from Iowa, Atlanta, and New York for the years 2004-2007 were excluded because teens were not included in data collection for these states. Two
additional records were excluded because of invalid data. Thus, responses from 954 teen control mothers (!20 years of age at conception) who gave birth between 1997 and 2007 were included in this analysis; 289 (30%) were age 19 at conception and formed the referent group for univariate and multivariate analyses. Several measures that could potentially introduce bias into our results were first assessed. Of teen mothers who were contacted to participate, 58.2% completed the interview, compared with 65.7% overall. Because of concern that recall bias might disproportionally impact responses from 1 or more age groups, we assessed the mean time to interview and found no differences among groups (data not shown). Likewise, we surmised that the month that pregnancy was recognized
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
267
Table 3 Multivariable Analysis of Selected Demographic and Periconceptional Characteristics by Age at Conception, NBDPS Control Mothers (1997-2007)* Younger than 20 Years at Conception (N 5 954) Age at Conception (!20), yy
Demographic and Periconceptional Characteristics*
13-15 OR (95% CI)
aOR (95% CI)z
x
Total Maternal ethnicity NH white NH Bback Hispanic Other Maternal nativity Foreign born US born Pregnancy intended Yes No Body mass index, kg/m2 Underweight (!18.5) Normal Overweight-obese (25þ) Diabetes Prepregnancy Yes No Gestational Yes No High blood pressure during pregnancy Yes No Alcohol use (B3-P1) Any None Binge drinking (B3-P1) Binge None Cigarette smoking (B3-P1) Any None Environmental tobacco smoking exposure (B3-P1) Any None Folic acid supplementation (B3-P1) Any None $5 Fruits/vegetables per day Yes No
16-17 n
OR (95% CI)
18
aOR (95% CI)z
110
n
OR (95% CI)
aOR (95% CI)z
308
n 247
1.00 2.22 (1.15-4.29) 2.66 (1.56-4.55) 2.26 (0.84-6.10)
1.00 1.84 (0.84-4.02) 2.83 (1.40-5.70) 1.27 (0.31-5.29)
25 21 50 3
1.00 1.41 (0.87-2.29) 2.05 (1.41-2.97) 1.85 (0.90-3.80)
1.00 1.41 (0.81-2.47) 2.58 (1.56-4.29) 1.95 (0.85-4.46)
87 48 129 18
1.00 1.28 (0.79-2.10) 1.37 (0.93-2.02) 1.40 (0.65-3.01)
1.00 1.37 (0.78-2.41) 1.55 (0.92-2.59) 1.42 (0.59-3.39)
90 43 82 15
0.50 (0.26-0.94) 1.00
0.25 (0.10-0.59) 1.00
11 88
0.76 (0.50-1.14) 1.00
0.42 (0.24-0.74) 1.00
42 240
0.79 (0.51-1.21) 1.00
0.64 (0.35-1.15) 1.00
33 197
0.51 (0.31-0.85) 1.00
0.52 (0.28-0.97) 1.00
21 78
0.76 (0.54-1.07) 1.00
0.72 (0.48-1.09) 1.00
81 201
1.05 (0.74-1.50) 1.00
0.98 (0.64-1.49) 1.00
81 149
0.93 (0.45-1.91) 1.00 0.32 (0.18-0.57)
1.01 (0.42-2.41) 1.00 0.32 (0.16-0.61)
12 70 17
0.89 (0.51-1.58) 1.00 0.62 (0.43-0.89)
1.05 (0.54-2.02) 1.00 0.61 (0.40-0.92)
29 175 78
0.85 (0.46-1.56) 1.00 0.61 (0.41-0.90)
0.81 (0.41-1.60) 1.00 0.61 (0.39-0.94)
23 145 62
N/A 1.00
N/A
N/A 1.00
N/A
N/A 1.00
N/A
0.26 (0.03-2.02) 1.00
1.03 (0.43-2.47) 1.00
1.48 (0.74-2.94) 1.00
N/A
0.40 (0.22-0.70) 1.00
0.97 (0.31-3.05) 1.00
0.37 (0.17-0.83) 1.00
1.05 (0.42-2.64) 1.00
1.13 (0.65-1.96) 1.00
N/A
1.04 (0.58-1.87) 1.00
N/A
18 81
0.51 (0.35-0.74) 1.00
0.54 (0.24-1.21) 1.00
61 221
0.80 (0.55-1.16) 1.00
0.69 (0.31-1.51) 1.00
72 158
0.45 (0.11-1.77) 1.00
8 85
0.58 (0.35-0.95) 1.00
0.87 (0.35-2.16) 1.00
30 223
0.89 (0.55-1.44) 1.00
1.22 (0.51-2.91) 1.00
37 163
0.40 (0.22-0.73) 1.00
0.45 (0.20-0.99) 1.00
14 85
0.76 (0.53-1.10) 1.00
1.04 (0.63-1.71) 1.00
69 213
1.01 (0.70-1.48) 1.00
1.02 (0.61-1.71) 1.00
72 158
1.06 (0.68-1.67) 1.00
1.59 (0.89-2.84) 1.00
44 55
0.67 (0.48-0.94) 1.00
0.88 (0.58-1.34) 1.00
102 180
1.08 (0.76-1.52) 1.00
1.08 (0.71-1.66) 1.00
109 121
0.37 (0.20-0.69) 1.00
0.44 (0.21-0.90) 1.00
14 85
0.79 (0.55-1.14) 1.00
0.90 (0.57-1.40) 1.00
70 212
0.87 (0.60-1.28) 1.00
0.97 (0.62-1.53) 1.00
63 167
0.84 (0.50-1.42) 1.00
N/A
0.77 (0.53-1.14) 1.00
N/A
0.67 (0.44-1.02) 1.00
N/A
N/A, Estimates not presented due to insufficient sample size * Excludes Iowa, the Centers for Disease Control and Prevention, and New York for the years 2004-2007. y Referent group: 19-year-olds. z Totals for some categories may vary due to missing values. All n values presented are based on adjusted estimates. See Table 2 for crude n values. x Adjusted for maternal race, nativity, body mass index, alcohol use (B3-P1), binge drinking (B3-P1), cigarette smoking (B3-P1), environmental tobacco smoking (B3-P1), folic acid supplementation (B3-P1), and/or pregnancy intention.
might vary by age and therefore affect behavior change in the periconceptional period (eg, that older teens might have better access to pregnancy confirmation tests and therefore be aware much earlier of the pregnancy, and thus be more likely to change their health during the early weeks of pregnancy). However, no differences by age groups were observed, and nearly one-half of interviews were missing a response to this question. Finally, age group distribution among the 8 study sites indicated that the only significant difference was that Texas had somewhat greater proportions of 13- to 15-year-olds and 15- to 16-year-olds (data not shown).
Demographic Characteristics
In our bivariate analysis (Table 2), compared with women who were 19 at the time of conception, younger teen mothers were significantly more likely to be Hispanic but less likely to be born outside of the United States. This association persisted in our multivariable models after adjusting for potential confounders, with both 13- to 15-year-olds and 16- to 17year-olds more likely to be Hispanic (adjusted odds ratio [aOR] 2.83 and 95% confidence interval [CI] 1.56-4.29, aOR 2.58 and 95% CI 1.40-5.70, respectively) and less likely to be foreign born (aOR 0.42 and 0.25) (Table 3).
268
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Table 4 Cumulative Percent Reporting 0-4 Selected Healthy Periconceptional Behaviors Cumulative Behaviors* Reported
$1y $2 $3 All 4
Mother's Age, y 13-15
16-17
18
19
96.6 78.9 22.4 5.9
92.6 71.2 28.8 4.3
91.8 67.2 30.3 3.6
89.5 68.3 33.1 4.9
* No smoking, no alcohol, consumed $5 servings fruits/vegetables per day, taking folic acid supplements. y Mantel-Haenszel c2 indicated P ! .05.
Health and Health Behaviors
In our multivariable analysis (Table 3), 13- to 15-yearolds presented a somewhat healthier picture than the oldest (referent) group in regard to cigarette smoking (aOR 0.45, 95% CI 0.20-0.99) and being either overweight or obese (aOR 0.32, 95% CI 0.16-0.61). However, the youngest teens (13- to 15-year-olds) were far less likely to have taken any folic acid supplements during the periconceptional period (aOR 0.44, 95% CI 0.21-0.90). Pregnancy Intention
Overall, about one-third of teens indicated that their pregnancies had been planned, but the youngest teens were about half as likely to indicate this compared with 19-yearolds (aOR 0.52) (Table 3). CPBI
Mean healthy behaviors for all teens on the CPBI was 2.15 (range 0-4) and did not significantly differ by age group. Reporting at least 1 positive periconceptional behavior was associated with a negative age gradient (ie, the younger the mothers, the more likely they were to report at least 1 positive behavior); otherwise, no difference in cumulative mean behaviors were indicated (Table 4). On the other hand, women who reported having intended pregnancy had a higher mean CPBI compared with their peers who did not (2.30 versus 1.94, P # .01) (Table 5), suggesting that older teens are significantly more likely to adopt positive behaviors when they are planning a pregnancy, but this pattern is not observed among younger teens. Discussion
This study used a population-based sample of women who had recently given birth and had become pregnant while still in their teens, to better understand how known risk factors for birth defects vary by age subgroups and Table 5 Mean Healthy Periconceptional Behaviors by Pregnancy Intention Mother's Age, y 13-15 16-17 18 19 All
Intended
Unintended
t
2.27 2.07 2.31 2.23 2.30
1.99 1.93 1.71 1.80 1.94
0.25 0.30 !0.01 !0.01 !0.01
pregnancy intention. Most studies of teen health behavior stratify either by age or by sex but not by both. Similarly, most studies of women's periconceptional behaviors do not differentiate among adolescent ages. To our knowledge, this is the first study to stratify periconceptional behaviors by age subgroups among teens using a nationally representative sample. The National Health and Nutrition Examination Surveys (NHANES) report of adolescent behaviors for 1999-2004 provides useful validation of our results. About 15% of the 13- to 15-year-olds in our study reported having smoked during the periconceptional period, which is consistent with the 14% of 14- to 15-year-old girls who indicated that they had smoked in the 30 days before responding to NHANES.31 Likewise, 24% of 16- to 17-year-old control subjects in the NBDPS had indicated smoking, compared with 25% among the NHANES sample (18- to 19-year-olds were not included). Reported binge drinking in the 2 studies were 8% in the 13- to 15-year-old group (NBDPS) and 9% for the 14- to 15-year-old girls in NHANES. However, in the 16- to 17-year-old group, rates among NHANES respondents were higher those than in the current study (19% and 12%, respectively). Similar alcohol consumption patterns between older and younger teen moms have been observed elsewhere.13 An important difference between NHANES and the current study (as well as the analysis by Harrison and Sidebottom) is that the NHANES respondents are not reporting behaviors during the periconceptional period. This suggests at least 2 possible implications. First, the lower rates of binge drinking among NBDPS participants may reflect behavior change associated with planning a pregnancy, taken with our observation that older teens were more likely to adopt certain good periconceptional behaviors. Second, NBDPS participants, having been pregnant, may be much more aware of the potential risks to the fetus of smoking and alcohol consumption and therefore may be more likely to give a more socially desirable response. Our finding of prepregnancy BMI in the overweight or obese range among teens (29%) was similar to that reported in NHANES (30-32% of females aged 12-19).32 However, in the present study, the youngest group (ages 13-15) reported prepregnancy mean BMI approximately one-half of their older counterparts. While cognitive approaches to healthy weight maintenance (ie, education) show little success in general,33 concerns for the health of future children may add sufficiently meaningful information to motivate some girls to maintain a healthy weight throughout adolescence. Moreover, knowledge of the effect of maternal prepregnancy weight on reproductive health is quite low,34,35 so advising adolescents that overweight and obesity could impair the health of their future children may provide completely new information. Although current attitudes and behaviors among adolescents regarding future childbearing is rarely discussed, much less studied, there is some evidence that teens can and do consider their desires for childbearing in light of their health.36 Folic acid intake among teens has been consistently lower than that of their older counterparts, although most studies do not distinguish between girls at risk for
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
pregnancy or who have been pregnant and those who are unlikely to become pregnant. Rosenberg at al23 observed that only 9% of teen respondents to the Oregon Pregnancy Risk Assessment Monitoring System (PRAMS) reported having supplemented with periconceptional folic acid, whereas NHANES reported that only 14% of adolescents aged 14-18 (including boys) reported taking multivitamins.37 These findings stand in contrast to our findings that nearly 25% of control teen mothers reported taking folic acid around the time of conception (22.6% among those with unintended pregnancies and 29.2% of those with intended pregnancies, data not shown). In addition to our sample containing women who were planning a pregnancy, it is possible that this discrepancy reflects dynamics similar to the questions about alcohol and tobacco, discussed earlier. Indeed, teens who have been pregnant may be much more likely to know that taking folic acid is beneficial and thus their responses may be more sensitive to social desirability bias. Using data from the National Vital Statistics system,7 researchers reported that 4.1% of 18- to 19-year-olds had experienced pregnancy-associated hypertension, whereas our subjects, including younger teens, reported more than 9%. Compared with the same study, we found much higher rates of gestational diabetes (3.2% for all teens, compared with 0.9%). These differences likely reflect methodological advantages conferred by using interview data in the NBDPS; thus, this study adds important data on pregnancy conditions among teens, previously unavailable from birth certificate data. The CPBI provides a picture of how periconceptional risk factors cluster. It is noteworthy that younger teens showed a slightly better periconceptional risk pattern compared with their older teen counterparts, primarily because they were less likely to have smoked or consumed alcohol in the weeks around conception. Most prior studies of adolescent risk behaviors have looked only at substance (ie, tobacco, alcohol, illicit drugs) exposure (although often in combination with sexual risk-taking), so their results are not directly comparable.38 However, in general, our results are consistent with those of other researchers in that potentially harmful preconceptual and periconceptional behaviors tend to cluster. For example, using a exploratory factor analysis, Santelli et al27 also observed that similar measures of nonsexual risk behavior are highly correlated (and are correlated with sexual risk behaviors associated with higher chance of pregnancy); they also found that risk behaviors were less common among younger teens. Likewise, in a study of pregnancies in Ireland, all 4 periconceptional behaviors included in our CPBI varied in a similar direction,39 although they examined aORs for each behavior individually. Further, as mentioned earlier, data for female subjects only by age are not readily available. Overall, nearly one-third of the adolescents in this study indicated that the index pregnancy was intended (ie, neither mistimed nor unwanted). This is considerably higher than has been observed in the past; previous studies of pregnancy intent have observed proportions of intended pregnancies among teens from 14%20 to 21%.25 The main contributor to this difference is found among the youngest
269
teensd22% of 13- to 15-year-old NBDPS control mothers indicated that the pregnancy was intended, compared to 5% of 12- to 15-year-olds interviewed by Phipps and Nunes.20 The National Survey of Family Growth found that only 9% of 15- to 17-year-olds and 19% of 18- to 19-year-olds reported that they would feel “pleased/very pleased” to become pregnant.40 However, measuring pregnancy intention, especially among teens, is particularly complex.41 This complexity, in addition to varying age groupings and question wording, makes meaningful comparison among studies difficult. At least 3 methodological limitations should be taken into account in interpreting these findings. First, because we used controls from a case-control study designed to look at risk factors for birth defects, these subjects by definition did not have offspring with any of the anomalies of interest, and therefore they may not represent the population in regard to known risk factors. Second, as with any study based on self-report of health behaviors, the answers may be affected by recall and social desirability bias. However, there is no evidence that these biases have a differential impact among groups within the study. Finally, we were unable to adequately assess whether month of recognition of pregnancy varied by age group, because nearly half of the interviewees did not provide a response to this question. We were looking at behaviors in the first 2 months after conception, so a high correlation between age and early pregnancy recognition could have affected multivariate analyses. We have demonstrated that several important periconceptional behaviors and characteristics, which have implications for the health of teen mothers' offspring and for the design of birth defects prevention programs, differ by age even among teens. Despite difficulties in measuring pregnancy intention, it is clear that a significant proportion of teen mothers, even very young ones, did plan their pregnancies. Our finding that healthy periconceptional behaviors were significantly higher among older teens who reported having planned their pregnancy compared with those who did not is comparable to findings for mothers of all ages.14 Thus, our results suggest that these young women are interested in learning about behaviors that can reduce the risk of birth defects and therefore highlight an opportunity for preventive interventions among this group. In addition, given that the younger teens were more likely to not drink or smoke, caregivers should consider that young women in this age group may be receptive to information affirming those choices and educating them about the positive benefits for future pregnancies. As with other age groups, knowledge about the need to begin healthy behaviors before a positive pregnancy test is essential. Acknowledgments
The authors thank the participating families, staff, and scientists from all NBDPS sites. References 1. Hoyert DL, Mathews TJ, Menacker F, et al: Annual summary of vital statistics: 2004. Pediatrics 2006; 117:168
270
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
2. Salihu HM, Pierre-Louis BJ, Druschel CM, et al: Omphalocele and gastroschisis in the State of New York,1992-1999. Birth Defects Res A Clin Mol Teratol 2003; 67:630 3. Grazi RV, Redheendran R, Mudaliar N, et al: Offspring of teenage mothers: congenital malformations, low birth weights and other findings. J Reprod Med 1982; 27:89 4. Parikh CR, McCall D, Engelman C, et al: Congenital renal agenesis: case-control analysis of birth characteristics. Am J Kidney Dis 2002; 39:689 5. Siega-Riz AM, Herring AH, Olshan AF, et al: The joint effects of maternal prepregnancy body mass index and age on the risk of gastroschisis. Paediatr Perinat Epidemiol 2009; 23:51 6. Olausson PO, Cnattingius S, Haglund B: Teenage pregnancies and risk of late fetal death and infant mortality. Br J Obstet Gynaecol 1999; 106:116 7. Menacker F, Martin JA, MacDorman MF, et al: Births to 10-14 year-old mothers, 1990-2002: trends and health outcomes. Natl Vital Stat Rep 2004; 53:1 8. Lawlor DA, Shaw M: Too much too young? Teenage pregnancy is not a public health problem. Int J Epidemiol 2002; 31:552 9. Stevens-Simon C, McAnarney ER: Skeletal maturity and growth of adolescent mothers: relationship to pregnancy outcome. J Adolesc Health 1993; 14:428 10. Chabra S, Gleason CA, Seidel K, et al: Rising prevalence of gastroschisis in Washington State. J Toxicol Environ Health A 2011; 74:336 11. Carmody D, Doyle A, Firth RG, et al: Teenage pregnancy in type 1 diabetes mellitus. Pediatr Diab 2010; 11:111 12. Phipps MG, Rosengard C, Weitzen S, et al: Age group differences among pregnant adolescents: sexual behavior, health habits and contraceptive use. J Pediatr Adolesc Gynecol 2008; 21:9 13. Harrison PA, Sidebottom AC: Alcohol and drug use before and during pregnancy: an examination of use patterns and predictors of cessation. Matern Child Health J 2009; 13:386 14. Dott M, Rasmussen SA, Hogue CJ, et al: Association between pregnancy intention and reproductive-health related behaviors before and after pregnancy recognition, National Birth Defects Prevention Study, 1997-2002. Matern Child Health J 2010; 14:373 15. Than LC, Honein MA, Watkins ML, et al: Intent to become pregnant as a predictor of exposures during pregnancy: is there a relation? J Reprod Med 2005; 50:389 16. Naimi TS, Lipscomb LE, Brewer RD, et al: Binge drinking in the preconception period and the risk of unintended pregnancy: implications for women and their children. Pediatrics 2003; 111:1136 17. Weisman CS, Hillemeier MM, Chase GA, et al: Preconceptional health: risks of adverse pregnancy outcomes by reproductive life stage in the Central Pennsylvania Women’s Health Study (CePAWHS). Women's Health Issues 2006; 16:216 18. D'Angelo D, Williams L, Morrow B, et al: Preconception and interconception health status of women who recently gave birth to a live-born infant: Pregnancy Risk Assessment Monitoring System (PRAMS), United States, 26 reporting areas, 2004. MMWR Surveill Summ 2007; 56:1 19. Charron-Prochownik D, Sereika SM, Falsetti D, et al: Knowledge, attitudes and behaviors related to sexuality and family planning in adolescent women with and without diabetes. Pediatr Diab 2006; 7:267 20. Phipps MG, Nunes AP: Assessing pregnancy intention and associated risks in pregnant adolescents. Matern Child Health J 2012; 16:1820 21. Rubin V, East PL: Adolescents' pregnancy intentions: relations to life situations and caretaking behaviors prenatally and 2 years postpartum. J Adolesc Health 1999; 24:313 22. Bloom KC, Hall DS: Pregnancy wantedness in adolescents presenting for pregnancy testing. MCN Am J Matern Child Nurs 1999; 24:296
23. Rosenberg KD, Gelow JM, Sandoval AP: Pregnancy intendedness and the use of periconceptional folic acid. Pediatrics 2003; 111:1142 24. Cheng D, Schwarz EB, Douglas E, et al: Unintended pregnancy and associated maternal preconception, prenatal and postpartum behaviors. Contraception 2009; 79:194 25. Bartz D, Shew M, Ofner S, et al: Pregnancy intentions and contraceptive behaviors among adolescent women: a coital event level analysis. J Adolesc Health 2007; 41:271 26. Prinstein MJ, Boergers J, Spirito A: Adolescents' and their friends' health-risk behavior: factors that alter or add to peer influence. J Pediatr Psychol 2001; 26:287 27. Santelli J, Carter M, Orr M, et al: Trends in sexual risk behaviors, by nonsexual risk behavior involvement, U.S. high school students, 1991-2007. J Adolesc Health 2009; 44:372 28. Cogswell ME, Bitsko RH, Anderka M, et al: Control selection and participation in an ongoing, population-based, case-control study of birth defects: the National Birth Defects Prevention Study. Am J Epidemiol 2009; 170:975 29. Yoon PW, Rasmussen SA, Lynberg MC, et al: The National Birth Defects Prevention Study. Public Health Rep 2001; 116(Suppl 1):32 30. Cutler GJ, Flood A, Hannan P, et al: Multiple sociodemographic and socioenvironmental characteristics are correlated with major patterns of dietary intake in adolescents. J Am Diet Assoc 2011; 111:230 31. Fryar CD, Merino MC, Hirsch R, et al: Smoking, alcohol use, and illicit drug use reported by adolescents aged 12e17 years: United States, 1991e2004. National health statistics reports; no 15. Hyattsville, MD: National Center for Health Statistics. 2009. 32. Ogden CL, Carroll MD, Curtin LR, et al: Prevalence of overweight and obesity in the United States, 1999-2004. JAMA 2006; 295:1549 33. Bonsergent E, Agrinier N, Thilly N, et al: Overweight and obesity prevention for adolescents: a cluster randomized controlled trial in a school setting. Am J Prev Med 2013; 44:30 34. Case AP, Royle M, Scheuerle AE, et al: Birth defects, causal attributions, and ethnicity in the National Birth Defects Prevention Study. J Genet Couns 2014; 23:860 35. Cardozo ER, Dune TJ, Neff LM, et al: Knowledge of obesity and its impact on reproductive health outcomes among urban women. J Commun Health 2013; 38:261 36. Quinn GP, Murphy D, Wang H, et al: Healthy adolescent girls' perceptions of cancer-related infertility. Having cancer doesn't change wanting a baby: healthy adolescent girls' perceptions of cancer-related infertility. J Adolesc Health 2013; 52:164 37. Picciano MF, Dwyer JT, Radimer KL, et al: Dietary supplement use among infants, children, and adolescents in the United States, 1999-2002. Arch Pediatr Adolesc Med 2007; 161:978 38. Leslie LK, James S, Monn A, et al: Health-risk behaviors in young adolescents in the child welfare system. J Adolesc Health 2010; 47:26 39. McCrory C, McNally S: The effect of pregnancy intention on maternal prenatal behaviours and parent and child health: results of an Irish cohort study. Paediatr Perinat Epidemiol 2013; 27:208 40. Abma JC, Martinez GM, Copen CE: Teenagers in the United States: Sexual Activity, Contraceptive Use, and Childbearing, National Survey of Family Growth 2006-2008. Washington, DC, National Center for Health Statistics, Vital Health Statistics, 2010 41. Montgomery KS: Creating consistency and control out of chaos: a qualitative view of planned pregnancy during adolescence. J Perinat Educ 2000; 9:7
Texas Department of State Health Services, Austin, TX The University of Texas School of Public Health, Austin Regional Campus & Michael and Susan Dell Center for Healthy Living, Austin, TX
a b s t r a c t Study Objectives: We sought to determine whether selected periconceptional health behaviors that influence risk for birth defects differ between older and younger adolescents and whether pregnancy intention predicts more positive preconception health behaviors among teens. Design and Participants: We analyzed interview responses from 954 adolescent control group participants from the National Birth Defects Prevention Study who delivered live infants during 1997-2007. Main Outcome Measures: Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were calculated for factors of interest by age categories (13-15, 16-17, and 18 years, relative to 19 years). To construct a composite periconceptional behavior index, we summed the following healthy behaviors: nonsmoker, nondrinker, folic acid supplementation, and eating 5 or more servings of fruits and vegetables per day. Results: Analyses indicated that women in the youngest group (13-15 years of age) were more likely to be Hispanic (aOR 2.83, 95% CI 1.405.70) and less likely to engage in some unhealthy pregnancy-related behaviors compared with 19-year-olds, such as smoking (aOR 0.45, 95% CI 0.20-0.99) and being overweight or obese (aOR 0.32, 95% CI 0.16-0.61). However, they were also less likely to have taken periconceptional folic acid (aOR 0.44, 95% CI 0.21-0.90). About one-third of teen mothers indicated that their pregnancies had been intended. Among 18- and 19-year-olds, this predicted a higher mean value for the composite periconceptional behavior index (2.30 versus 1.94, P # .01). Conclusions: Teen mothers are not a homogeneous group. Each age subgroup presents varied demographic and behavioral factors that put them at varying levels of risk for birth defects. Furthermore, caregivers should not assume that teens do not plan pregnancies or that they need not be informed of the importance of periconceptional health. Key Words: Adolescent, Preconception care, Health behavior, Tobacco use, Adolescent behavior, Drinking behavior, Fruit, Vegetables, Folic acid, Congenital anomalies, Health surveys
Introduction
Although birth rates for US teens continue to fall, about 400,000 babies are delivered each year to mothers aged 20 years or younger; about 14% of those are born to mothers younger than 17 years.1 Several types of birth defects, including gastroschisis and renal agenesis, disproportionally affect the offspring of younger mothers,2e5 and there is some evidence to suggest that births to very young teens disproportionately contribute to this burden,6,7 underscoring the possibility that they may have different risk factors compared with older adolescents. Findings on the contribution of physiological immaturity to poorer birth outcomes among offspring born to adolescents have been inconsistent,8,9 suggesting that differing periconceptional health and behaviors play a part.10 However, while it is clear from studies of within-group differences of nonpregnant Funding for this study was provided in part by the Texas Center for Birth Defects Research and Prevention, at the Texas Department of State Health Services, through cooperative agreement No. U50/CCU613232 from the Centers for Disease Control and Prevention (CDC). No manifest or potential conflicts of interest are noted. * Address correspondence to: Amy P. Case, PhD, Birth Defects Epidemiology & Surveillance Branch, Texas Department of State Health Services, P.O. Box 149347, Austin, TX 78714-9347, USA; Phone: (512) 565-1687; fax: (512) 458-7330 E-mail address: [email protected] (A.P. Case).
teens that each age group presents a different behavioral and physiological profile, most etiological birth defect studies treat teen mothers as homogeneous regardless of their age. Therefore, it is important for future studies of birth defects risk factors to better understand whether and how relevant periconceptional health and behaviors vary within this age group. Several patterns about the health behaviors of interest have been observed in the literature that inform the basis for this study: First, these behaviors tend to differ between nonpregnant older and younger teens11,12; second, younger and older teen mothers and their adult counterparts exhibit different behaviors during pregnancy13; and, third, in general, mothers report several different periconceptional behaviors depending on whether the pregnancy was planned.14,15 However, virtually all studies of preconception health treat teens as a single stratum, and only occasionally are younger teens represented in these studies.16e18 Although there are notable exceptions,12,19 scant attention has been directed to preconception behaviors and exposures among the youngest mothers and whether they differ from older teens. Finally, the extent to which pregnancy intention exerts a similar influence among teens has not been explored.
1083-3188/$ - see front matter Ó 2015 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpag.2014.09.004
264
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Although it is often assumed that pregnancies among teen mothers are universally unintended, research indicates that between 7% and 35% of these conceptions were planned or wanted.20e22 Although the proportion varies by age group, even among the youngest teens, planned pregnancy is not unknown.20 Pregnancy intention is a particularly important characteristic when considering preconception behaviors,15,23,24 because the intent to become pregnant may act as a strong motivator to change unhealthy behaviors to improve the prospects of that pregnancy. Although teen mothers have exhibited improved self-care behaviors during and after pregnancy,21 it is not clear whether adolescents, whose motives, attitudes, and awareness toward pregnancy planning may be particularly complex,25 tend to adopt healthier behaviors before an intended conception occurs. Many health promotion strategies rely on tailoring to specific characteristics of a population, including attitudes and behavioral intentions. Further, health behaviors tend to cluster in the same individuals26; conversely, unhealthy behaviors co-vary considerably and may cluster in different patterns between adolescent age groups.27 Identifying how risk and protective factors cluster will further inform the development of health promotion strategies. The primary aim of this study was to determine whether periconceptional health conditions and behaviors, particularly those relevant to birth defects prevention, vary among teen mothers by age subgroups. An additional aim was to investigate whether selected behaviors vary according to whether teen mothers indicated having planned their pregnancies and whether those patterns also vary for older and younger teens.
Subjects and Methods Study Population
We conducted a nested cross-sectional analysis of responses given by control participants in the National Birth Defects Prevention Study (NBDPS). The NBDPS is a multicenter case-control study with 3 major components: (1) collection and categorization of information regarding the index infant and maternal demographics obtained from state birth defects registries, (2) a 1-hour computer-assisted telephone interview of the mother, and (3) collection of DNA from maternal, paternal, and child buccal cell samples. The interview, which must be concluded within 2 years after the index delivery, includes questions regarding periconceptional health, such as dietary habits, exposures during pregnancy, occupation, and health status around the periconceptional period. Control mothers gave birth to liveborn infants selected from the same base population as cases but their children had no major malformations and they had an estimated date of delivery occurring in the same year as cases. They are selected from either birth certificates or birth hospitals by using a stratified random sampling scheme.28 Responses from control mothers who were younger than 20 years and gave birth to a child without a birth defect covered by the NBDPS between 1997
and 2007 form the basis of the present analysis. The NBDPS has been described in greater detail elsewhere.28,29 The NBDPS was approved by the Office of Management and Budget and the appropriate institutional review boards at the Centers for Disease Control and Prevention and each participating site. Data Analyses
Variables of interest were behaviors or characteristics that are known risk factors for birth defects, with particular attention to behaviors that, based on a review of the literature, could be expected to vary among the age groups of interest (eg, alcohol use, high blood pressure). A priori power calculations were performed for the selected variables, which indicated that maternal smoking, environmental tobacco exposure, drinking, high blood pressure, obesity, folic acid use, and maternal nativity were within the capacity of this analysis to identify meaningful associations with the age groups of interest. In contrast, medications other than oral contraceptives and prepregnancy diabetes were eliminated from further analyses due to insufficient power. Responses regarding the 3 months before and 1 month after conception were included for behavioral variables (eg, folic acid supplementation, binge drinking); dietary recall, however, refers to average consumption during the 1 year before pregnancy. (A detailed description of variables is given in Table 1.) We calculated c2 values or 2-sided Fisher's exact estimates in cases where cell sizes were less than 5 for our risk factors of interest and for the 4 age categories (13-15, 16-17, 18, and 19 years). Three separate binary logistic regression models were run for each of the 3 teen subcategories against 19-year-olds (ie, 13- to 15-year-olds, 16- to 17-yearolds, and 18-year-olds versus 19-year-olds) for all maternal risk factors. Both crude and adjusted multivariate models were assessed adjusting for all other factors considered significant in at least 1 of the crude models. All statistical calculations were performed with SAS software, version 9.3 (SAS Institute Inc, 2011). Recognizing that adolescent risk behaviors often occur together, we constructed a composite periconceptional behavior index (CPBI) by computing mean behaviors by age group for the following reported behavioral characteristics, which are known to play a role in preventing certain birth defects. Cumulative responses to 4 known risk/protective factors for birth defects were grouped by age and number of positive responses (smoking and drinking were inverted so that the positive corollary could be compared to the protective factors). The 4 behaviors were (a) not smoking (3 months before to 1 month after conception), (b) not drinking any alcohol (3 months before to 1 month after conception), (c) consuming at least 5 servings of fruits or vegetables per day in the year before conception (proxy for a balanced diet),30 and (d) taking folic acid supplements (3 months before to 1 month after conception). We used the CPBI in 2 analyses used to determine if composite behavior patterns differed by age groups: (a) cumulative percent reporting each level of the CPBI (potential range 0-4,
Table 1 Description of Selected Periconceptional Health and Behavioral Variables Variable
Responses Used in Analysis of This Variable*
Were you ever told by a doctor that you had diabetes (including gestational diabetes), sometimes called sugar diabetes or diabetes mellitus? If Yes: What type of diabetes did you have?
High blood pressure during pregnancy
Did you have pregnancy-related high blood pressure when you were pregnant with (this pregnancy)?
Pregnancy Intention
At the time that you became pregnant (with this pregnancy), did you (a) want to become pregnant then; (b) want to wait until later, or (c) not want to become pregnant at all? 1. Did you drink any wine, beer, mixed drinks, or shots of liquor three months prior to conception to the end of pregnancy? If Yes: Which months/days did you drink alcoholic beverages? During which months did you drink any alcoholic beverages? 2. In the (third/second/first month before pregnancy, first/second/third month of pregnancy, second/third trimester), on average, how many days/month did you drink alcoholic beverages? On those days that you drank alcoholic beverages, on average, how many drinks did you have per day? What was the greatest number of drinks you had on 1 occasion? From 3 months before you became pregnant to the end of your pregnancy, did you smoke cigarettes? If Yes, during which months did you smoke? Did anyone in your household smoke cigarettes in your home between 3 months before you became pregnant to the end of your pregnancy? During which months did someone smoke in your home? From 3 months before you became pregnant to the end of your pregnancy, did you take any of the following single vitamins or minerals? (W)hat your usual diet was like (in the year) before you were pregnant? For seasonal foods, fruits, and vegetables, you can average over the 6 months prior to (this pregnancy). For foods that you ate less than once a month, you can report as never or none. (Interviewer reads list of food items, and asks, “How often, on average, did you use.”)
Was it: Gestational (diagnosed during pregnancy only) Insulin-dependent diabetes, also called type 1 or juvenile; noneinsulin-dependent diabetes, also called type 2 or adult onset (diagnosed before pregnancy only) Yes (diagnosed during pregnancy only) No Intendedy Unintended (Mistimed, ambivalent, or unwanted)y Any: Yes, B3-P1 Binge: $4 drinks on any occasion, B3-P1
Alcohol use
Smoking Environmental tobacco smoke (home) Folic acid $5 Fruits/vegetables per day
Yes: B3-P1 No Yes: B3-P1 Yes to folic acid: B3-P1
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Question Summary
Diabetes
Average 5 servings/day of any combination of fruit or vegetable items (including orange and tomato juices)
* B signifies month(s) prior to conception (eg, B1); P signifies month(s) after conception (eg, P3). y The coded variable actually uses responses to several questions to classify pregnancy intention, including questions about contraceptive use at the time of conception. See Dott et al, “Association between pregnancy intention and reproductive-health related behaviors before and after pregnancy recognition, National Birth Defects Prevention Study, 1997-2002” (2010) for more detail.
265
266
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Table 2 Selected Demographic and Periconceptional Characteristics by Age at Conception, NBDPS Control Mothers (1997-2007)* Younger than 20 Years at Conception (N 5 954) Demographic and Periconceptional Characteristics
All Teens 13-15 n
Totalz Maternal ethnicity NH white NH black Hispanic Other Nativity Foreign born US born Pregnancy intended Yes No Body mass index, kg/m2 Underweight (!18.5) Normal Overweight-obese (25þ) Diabetes Prepregnancy Yes No Gestational Yes No High blood pressure during pregnancy Yes No Alcohol usejj Any None Binge drinkingjj Binge None Cigarette smokingjj Any None Environmental tobacco smoke exposurejj Any None Folic acid supplementationjj Any None $5 Fruits/vegetables per day{ Yes No
Py
Age at Conception, y 16-17 %
n
18 %
308
n
19 %
247
n
%
954
110
289
333 163 401 57
26 22 55 7
23.6 20.0 50.0 6.4
91 49 148 20
29.5 15.9 48.1 6.5
90 44 98 15
36.4 17.8 39.7 6.1
126 48 100 15
43.6 16.6 34.6 5.2
.05
173 747
13 91
12.5 87.5
53 243
17.9 82.1
45 198
18.5 81.5
62 215
22.4 77.6
.31
306 648
24 86
21.8 78.2
90 218
29.2 70.8
90 157
36.4 63.6
102 187
35.3 64.7
.84
92 546 258
13 76 17
12.3 71.7 16.0
30 182 79
10.3 62.5 27.2
23 147 63
9.9 63.1 27.0
26 141 99
9.8 53.0 37.2
.02
4 950
1 109
0.9 99.1
1 307
0.3 99.7
2 245
0.8 99.2
0 289
0.0 100.0
.26x
31 923
1 109
0.9 99.1
11 297
3.6 96.4
9 238
3.6 96.4
10 279
3.5 96.5
.55x
94 860
14 96
12.7 87.3
31 277
10.1 89.9
23 224
9.3 90.7
26 263
9.0 91.0
.97
247 675
18 87
17.1 82.9
62 234
20.9 79.1
72 172
29.5 70.5
95 182
34.3 65.7
.12
122 695
8 91
8.1 91.9
32 235
12.0 88.0
37 177
17.3 82.7
45 192
19.0 81.0
.53
242 681
15 90
14.3 85.7
72 225
24.2 75.8
73 171
29.9 70.1
82 195
29.6 70.4
.86
389 533
48 56
46.2 53.8
104 193
35.0 65.0
113 130
46.5 53.5
124 154
44.6 55.4
.69
233 709
14 93
13.1 86.9
73 230
24.1 75.9
64 182
26.0 74.0
82 204
28.7 71.3
.66
204 750
24 86
21.8 78.2
63 245
20.5 79.6
45 202
18.2 81.8
72 217
24.9 75.1
.05
* Excludes Iowa, the Centers for Disease Control and Prevention, and New York for the years 2004-2007. y Mantel-Haenszel c2 or Fisher's exact test if indicated calculated for all age groups by respective demographic characteristics and risk factors. z Totals for some categories may vary due to missing values. x Two-sided Fisher's exact test. jj Maternal exposure 3 months before conception through the first month of pregnancy (see Table 1). { Participants questioned on how often they consumed a particular food during the year before becoming pregnant with index birth (see Table 1).
with 4 representing all behaviors) were calculated by age groups and (b) mean number of behaviors for each group was calculated for both intended and unintended pregnancies. A Student t statistic was computed to compare the mean values. Results
Participants for the current analysis were selected from the larger NBDPS database based on the following criteria: (a) live birth between 1997 and 2007, (b) no birth defect noted, and (c) the mother was younger than 20 years when she gave birth. Mothers from Iowa, Atlanta, and New York for the years 2004-2007 were excluded because teens were not included in data collection for these states. Two
additional records were excluded because of invalid data. Thus, responses from 954 teen control mothers (!20 years of age at conception) who gave birth between 1997 and 2007 were included in this analysis; 289 (30%) were age 19 at conception and formed the referent group for univariate and multivariate analyses. Several measures that could potentially introduce bias into our results were first assessed. Of teen mothers who were contacted to participate, 58.2% completed the interview, compared with 65.7% overall. Because of concern that recall bias might disproportionally impact responses from 1 or more age groups, we assessed the mean time to interview and found no differences among groups (data not shown). Likewise, we surmised that the month that pregnancy was recognized
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
267
Table 3 Multivariable Analysis of Selected Demographic and Periconceptional Characteristics by Age at Conception, NBDPS Control Mothers (1997-2007)* Younger than 20 Years at Conception (N 5 954) Age at Conception (!20), yy
Demographic and Periconceptional Characteristics*
13-15 OR (95% CI)
aOR (95% CI)z
x
Total Maternal ethnicity NH white NH Bback Hispanic Other Maternal nativity Foreign born US born Pregnancy intended Yes No Body mass index, kg/m2 Underweight (!18.5) Normal Overweight-obese (25þ) Diabetes Prepregnancy Yes No Gestational Yes No High blood pressure during pregnancy Yes No Alcohol use (B3-P1) Any None Binge drinking (B3-P1) Binge None Cigarette smoking (B3-P1) Any None Environmental tobacco smoking exposure (B3-P1) Any None Folic acid supplementation (B3-P1) Any None $5 Fruits/vegetables per day Yes No
16-17 n
OR (95% CI)
18
aOR (95% CI)z
110
n
OR (95% CI)
aOR (95% CI)z
308
n 247
1.00 2.22 (1.15-4.29) 2.66 (1.56-4.55) 2.26 (0.84-6.10)
1.00 1.84 (0.84-4.02) 2.83 (1.40-5.70) 1.27 (0.31-5.29)
25 21 50 3
1.00 1.41 (0.87-2.29) 2.05 (1.41-2.97) 1.85 (0.90-3.80)
1.00 1.41 (0.81-2.47) 2.58 (1.56-4.29) 1.95 (0.85-4.46)
87 48 129 18
1.00 1.28 (0.79-2.10) 1.37 (0.93-2.02) 1.40 (0.65-3.01)
1.00 1.37 (0.78-2.41) 1.55 (0.92-2.59) 1.42 (0.59-3.39)
90 43 82 15
0.50 (0.26-0.94) 1.00
0.25 (0.10-0.59) 1.00
11 88
0.76 (0.50-1.14) 1.00
0.42 (0.24-0.74) 1.00
42 240
0.79 (0.51-1.21) 1.00
0.64 (0.35-1.15) 1.00
33 197
0.51 (0.31-0.85) 1.00
0.52 (0.28-0.97) 1.00
21 78
0.76 (0.54-1.07) 1.00
0.72 (0.48-1.09) 1.00
81 201
1.05 (0.74-1.50) 1.00
0.98 (0.64-1.49) 1.00
81 149
0.93 (0.45-1.91) 1.00 0.32 (0.18-0.57)
1.01 (0.42-2.41) 1.00 0.32 (0.16-0.61)
12 70 17
0.89 (0.51-1.58) 1.00 0.62 (0.43-0.89)
1.05 (0.54-2.02) 1.00 0.61 (0.40-0.92)
29 175 78
0.85 (0.46-1.56) 1.00 0.61 (0.41-0.90)
0.81 (0.41-1.60) 1.00 0.61 (0.39-0.94)
23 145 62
N/A 1.00
N/A
N/A 1.00
N/A
N/A 1.00
N/A
0.26 (0.03-2.02) 1.00
1.03 (0.43-2.47) 1.00
1.48 (0.74-2.94) 1.00
N/A
0.40 (0.22-0.70) 1.00
0.97 (0.31-3.05) 1.00
0.37 (0.17-0.83) 1.00
1.05 (0.42-2.64) 1.00
1.13 (0.65-1.96) 1.00
N/A
1.04 (0.58-1.87) 1.00
N/A
18 81
0.51 (0.35-0.74) 1.00
0.54 (0.24-1.21) 1.00
61 221
0.80 (0.55-1.16) 1.00
0.69 (0.31-1.51) 1.00
72 158
0.45 (0.11-1.77) 1.00
8 85
0.58 (0.35-0.95) 1.00
0.87 (0.35-2.16) 1.00
30 223
0.89 (0.55-1.44) 1.00
1.22 (0.51-2.91) 1.00
37 163
0.40 (0.22-0.73) 1.00
0.45 (0.20-0.99) 1.00
14 85
0.76 (0.53-1.10) 1.00
1.04 (0.63-1.71) 1.00
69 213
1.01 (0.70-1.48) 1.00
1.02 (0.61-1.71) 1.00
72 158
1.06 (0.68-1.67) 1.00
1.59 (0.89-2.84) 1.00
44 55
0.67 (0.48-0.94) 1.00
0.88 (0.58-1.34) 1.00
102 180
1.08 (0.76-1.52) 1.00
1.08 (0.71-1.66) 1.00
109 121
0.37 (0.20-0.69) 1.00
0.44 (0.21-0.90) 1.00
14 85
0.79 (0.55-1.14) 1.00
0.90 (0.57-1.40) 1.00
70 212
0.87 (0.60-1.28) 1.00
0.97 (0.62-1.53) 1.00
63 167
0.84 (0.50-1.42) 1.00
N/A
0.77 (0.53-1.14) 1.00
N/A
0.67 (0.44-1.02) 1.00
N/A
N/A, Estimates not presented due to insufficient sample size * Excludes Iowa, the Centers for Disease Control and Prevention, and New York for the years 2004-2007. y Referent group: 19-year-olds. z Totals for some categories may vary due to missing values. All n values presented are based on adjusted estimates. See Table 2 for crude n values. x Adjusted for maternal race, nativity, body mass index, alcohol use (B3-P1), binge drinking (B3-P1), cigarette smoking (B3-P1), environmental tobacco smoking (B3-P1), folic acid supplementation (B3-P1), and/or pregnancy intention.
might vary by age and therefore affect behavior change in the periconceptional period (eg, that older teens might have better access to pregnancy confirmation tests and therefore be aware much earlier of the pregnancy, and thus be more likely to change their health during the early weeks of pregnancy). However, no differences by age groups were observed, and nearly one-half of interviews were missing a response to this question. Finally, age group distribution among the 8 study sites indicated that the only significant difference was that Texas had somewhat greater proportions of 13- to 15-year-olds and 15- to 16-year-olds (data not shown).
Demographic Characteristics
In our bivariate analysis (Table 2), compared with women who were 19 at the time of conception, younger teen mothers were significantly more likely to be Hispanic but less likely to be born outside of the United States. This association persisted in our multivariable models after adjusting for potential confounders, with both 13- to 15-year-olds and 16- to 17year-olds more likely to be Hispanic (adjusted odds ratio [aOR] 2.83 and 95% confidence interval [CI] 1.56-4.29, aOR 2.58 and 95% CI 1.40-5.70, respectively) and less likely to be foreign born (aOR 0.42 and 0.25) (Table 3).
268
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
Table 4 Cumulative Percent Reporting 0-4 Selected Healthy Periconceptional Behaviors Cumulative Behaviors* Reported
$1y $2 $3 All 4
Mother's Age, y 13-15
16-17
18
19
96.6 78.9 22.4 5.9
92.6 71.2 28.8 4.3
91.8 67.2 30.3 3.6
89.5 68.3 33.1 4.9
* No smoking, no alcohol, consumed $5 servings fruits/vegetables per day, taking folic acid supplements. y Mantel-Haenszel c2 indicated P ! .05.
Health and Health Behaviors
In our multivariable analysis (Table 3), 13- to 15-yearolds presented a somewhat healthier picture than the oldest (referent) group in regard to cigarette smoking (aOR 0.45, 95% CI 0.20-0.99) and being either overweight or obese (aOR 0.32, 95% CI 0.16-0.61). However, the youngest teens (13- to 15-year-olds) were far less likely to have taken any folic acid supplements during the periconceptional period (aOR 0.44, 95% CI 0.21-0.90). Pregnancy Intention
Overall, about one-third of teens indicated that their pregnancies had been planned, but the youngest teens were about half as likely to indicate this compared with 19-yearolds (aOR 0.52) (Table 3). CPBI
Mean healthy behaviors for all teens on the CPBI was 2.15 (range 0-4) and did not significantly differ by age group. Reporting at least 1 positive periconceptional behavior was associated with a negative age gradient (ie, the younger the mothers, the more likely they were to report at least 1 positive behavior); otherwise, no difference in cumulative mean behaviors were indicated (Table 4). On the other hand, women who reported having intended pregnancy had a higher mean CPBI compared with their peers who did not (2.30 versus 1.94, P # .01) (Table 5), suggesting that older teens are significantly more likely to adopt positive behaviors when they are planning a pregnancy, but this pattern is not observed among younger teens. Discussion
This study used a population-based sample of women who had recently given birth and had become pregnant while still in their teens, to better understand how known risk factors for birth defects vary by age subgroups and Table 5 Mean Healthy Periconceptional Behaviors by Pregnancy Intention Mother's Age, y 13-15 16-17 18 19 All
Intended
Unintended
t
2.27 2.07 2.31 2.23 2.30
1.99 1.93 1.71 1.80 1.94
0.25 0.30 !0.01 !0.01 !0.01
pregnancy intention. Most studies of teen health behavior stratify either by age or by sex but not by both. Similarly, most studies of women's periconceptional behaviors do not differentiate among adolescent ages. To our knowledge, this is the first study to stratify periconceptional behaviors by age subgroups among teens using a nationally representative sample. The National Health and Nutrition Examination Surveys (NHANES) report of adolescent behaviors for 1999-2004 provides useful validation of our results. About 15% of the 13- to 15-year-olds in our study reported having smoked during the periconceptional period, which is consistent with the 14% of 14- to 15-year-old girls who indicated that they had smoked in the 30 days before responding to NHANES.31 Likewise, 24% of 16- to 17-year-old control subjects in the NBDPS had indicated smoking, compared with 25% among the NHANES sample (18- to 19-year-olds were not included). Reported binge drinking in the 2 studies were 8% in the 13- to 15-year-old group (NBDPS) and 9% for the 14- to 15-year-old girls in NHANES. However, in the 16- to 17-year-old group, rates among NHANES respondents were higher those than in the current study (19% and 12%, respectively). Similar alcohol consumption patterns between older and younger teen moms have been observed elsewhere.13 An important difference between NHANES and the current study (as well as the analysis by Harrison and Sidebottom) is that the NHANES respondents are not reporting behaviors during the periconceptional period. This suggests at least 2 possible implications. First, the lower rates of binge drinking among NBDPS participants may reflect behavior change associated with planning a pregnancy, taken with our observation that older teens were more likely to adopt certain good periconceptional behaviors. Second, NBDPS participants, having been pregnant, may be much more aware of the potential risks to the fetus of smoking and alcohol consumption and therefore may be more likely to give a more socially desirable response. Our finding of prepregnancy BMI in the overweight or obese range among teens (29%) was similar to that reported in NHANES (30-32% of females aged 12-19).32 However, in the present study, the youngest group (ages 13-15) reported prepregnancy mean BMI approximately one-half of their older counterparts. While cognitive approaches to healthy weight maintenance (ie, education) show little success in general,33 concerns for the health of future children may add sufficiently meaningful information to motivate some girls to maintain a healthy weight throughout adolescence. Moreover, knowledge of the effect of maternal prepregnancy weight on reproductive health is quite low,34,35 so advising adolescents that overweight and obesity could impair the health of their future children may provide completely new information. Although current attitudes and behaviors among adolescents regarding future childbearing is rarely discussed, much less studied, there is some evidence that teens can and do consider their desires for childbearing in light of their health.36 Folic acid intake among teens has been consistently lower than that of their older counterparts, although most studies do not distinguish between girls at risk for
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
pregnancy or who have been pregnant and those who are unlikely to become pregnant. Rosenberg at al23 observed that only 9% of teen respondents to the Oregon Pregnancy Risk Assessment Monitoring System (PRAMS) reported having supplemented with periconceptional folic acid, whereas NHANES reported that only 14% of adolescents aged 14-18 (including boys) reported taking multivitamins.37 These findings stand in contrast to our findings that nearly 25% of control teen mothers reported taking folic acid around the time of conception (22.6% among those with unintended pregnancies and 29.2% of those with intended pregnancies, data not shown). In addition to our sample containing women who were planning a pregnancy, it is possible that this discrepancy reflects dynamics similar to the questions about alcohol and tobacco, discussed earlier. Indeed, teens who have been pregnant may be much more likely to know that taking folic acid is beneficial and thus their responses may be more sensitive to social desirability bias. Using data from the National Vital Statistics system,7 researchers reported that 4.1% of 18- to 19-year-olds had experienced pregnancy-associated hypertension, whereas our subjects, including younger teens, reported more than 9%. Compared with the same study, we found much higher rates of gestational diabetes (3.2% for all teens, compared with 0.9%). These differences likely reflect methodological advantages conferred by using interview data in the NBDPS; thus, this study adds important data on pregnancy conditions among teens, previously unavailable from birth certificate data. The CPBI provides a picture of how periconceptional risk factors cluster. It is noteworthy that younger teens showed a slightly better periconceptional risk pattern compared with their older teen counterparts, primarily because they were less likely to have smoked or consumed alcohol in the weeks around conception. Most prior studies of adolescent risk behaviors have looked only at substance (ie, tobacco, alcohol, illicit drugs) exposure (although often in combination with sexual risk-taking), so their results are not directly comparable.38 However, in general, our results are consistent with those of other researchers in that potentially harmful preconceptual and periconceptional behaviors tend to cluster. For example, using a exploratory factor analysis, Santelli et al27 also observed that similar measures of nonsexual risk behavior are highly correlated (and are correlated with sexual risk behaviors associated with higher chance of pregnancy); they also found that risk behaviors were less common among younger teens. Likewise, in a study of pregnancies in Ireland, all 4 periconceptional behaviors included in our CPBI varied in a similar direction,39 although they examined aORs for each behavior individually. Further, as mentioned earlier, data for female subjects only by age are not readily available. Overall, nearly one-third of the adolescents in this study indicated that the index pregnancy was intended (ie, neither mistimed nor unwanted). This is considerably higher than has been observed in the past; previous studies of pregnancy intent have observed proportions of intended pregnancies among teens from 14%20 to 21%.25 The main contributor to this difference is found among the youngest
269
teensd22% of 13- to 15-year-old NBDPS control mothers indicated that the pregnancy was intended, compared to 5% of 12- to 15-year-olds interviewed by Phipps and Nunes.20 The National Survey of Family Growth found that only 9% of 15- to 17-year-olds and 19% of 18- to 19-year-olds reported that they would feel “pleased/very pleased” to become pregnant.40 However, measuring pregnancy intention, especially among teens, is particularly complex.41 This complexity, in addition to varying age groupings and question wording, makes meaningful comparison among studies difficult. At least 3 methodological limitations should be taken into account in interpreting these findings. First, because we used controls from a case-control study designed to look at risk factors for birth defects, these subjects by definition did not have offspring with any of the anomalies of interest, and therefore they may not represent the population in regard to known risk factors. Second, as with any study based on self-report of health behaviors, the answers may be affected by recall and social desirability bias. However, there is no evidence that these biases have a differential impact among groups within the study. Finally, we were unable to adequately assess whether month of recognition of pregnancy varied by age group, because nearly half of the interviewees did not provide a response to this question. We were looking at behaviors in the first 2 months after conception, so a high correlation between age and early pregnancy recognition could have affected multivariate analyses. We have demonstrated that several important periconceptional behaviors and characteristics, which have implications for the health of teen mothers' offspring and for the design of birth defects prevention programs, differ by age even among teens. Despite difficulties in measuring pregnancy intention, it is clear that a significant proportion of teen mothers, even very young ones, did plan their pregnancies. Our finding that healthy periconceptional behaviors were significantly higher among older teens who reported having planned their pregnancy compared with those who did not is comparable to findings for mothers of all ages.14 Thus, our results suggest that these young women are interested in learning about behaviors that can reduce the risk of birth defects and therefore highlight an opportunity for preventive interventions among this group. In addition, given that the younger teens were more likely to not drink or smoke, caregivers should consider that young women in this age group may be receptive to information affirming those choices and educating them about the positive benefits for future pregnancies. As with other age groups, knowledge about the need to begin healthy behaviors before a positive pregnancy test is essential. Acknowledgments
The authors thank the participating families, staff, and scientists from all NBDPS sites. References 1. Hoyert DL, Mathews TJ, Menacker F, et al: Annual summary of vital statistics: 2004. Pediatrics 2006; 117:168
270
A.P. Case et al. / J Pediatr Adolesc Gynecol 28 (2015) 263e270
2. Salihu HM, Pierre-Louis BJ, Druschel CM, et al: Omphalocele and gastroschisis in the State of New York,1992-1999. Birth Defects Res A Clin Mol Teratol 2003; 67:630 3. Grazi RV, Redheendran R, Mudaliar N, et al: Offspring of teenage mothers: congenital malformations, low birth weights and other findings. J Reprod Med 1982; 27:89 4. Parikh CR, McCall D, Engelman C, et al: Congenital renal agenesis: case-control analysis of birth characteristics. Am J Kidney Dis 2002; 39:689 5. Siega-Riz AM, Herring AH, Olshan AF, et al: The joint effects of maternal prepregnancy body mass index and age on the risk of gastroschisis. Paediatr Perinat Epidemiol 2009; 23:51 6. Olausson PO, Cnattingius S, Haglund B: Teenage pregnancies and risk of late fetal death and infant mortality. Br J Obstet Gynaecol 1999; 106:116 7. Menacker F, Martin JA, MacDorman MF, et al: Births to 10-14 year-old mothers, 1990-2002: trends and health outcomes. Natl Vital Stat Rep 2004; 53:1 8. Lawlor DA, Shaw M: Too much too young? Teenage pregnancy is not a public health problem. Int J Epidemiol 2002; 31:552 9. Stevens-Simon C, McAnarney ER: Skeletal maturity and growth of adolescent mothers: relationship to pregnancy outcome. J Adolesc Health 1993; 14:428 10. Chabra S, Gleason CA, Seidel K, et al: Rising prevalence of gastroschisis in Washington State. J Toxicol Environ Health A 2011; 74:336 11. Carmody D, Doyle A, Firth RG, et al: Teenage pregnancy in type 1 diabetes mellitus. Pediatr Diab 2010; 11:111 12. Phipps MG, Rosengard C, Weitzen S, et al: Age group differences among pregnant adolescents: sexual behavior, health habits and contraceptive use. J Pediatr Adolesc Gynecol 2008; 21:9 13. Harrison PA, Sidebottom AC: Alcohol and drug use before and during pregnancy: an examination of use patterns and predictors of cessation. Matern Child Health J 2009; 13:386 14. Dott M, Rasmussen SA, Hogue CJ, et al: Association between pregnancy intention and reproductive-health related behaviors before and after pregnancy recognition, National Birth Defects Prevention Study, 1997-2002. Matern Child Health J 2010; 14:373 15. Than LC, Honein MA, Watkins ML, et al: Intent to become pregnant as a predictor of exposures during pregnancy: is there a relation? J Reprod Med 2005; 50:389 16. Naimi TS, Lipscomb LE, Brewer RD, et al: Binge drinking in the preconception period and the risk of unintended pregnancy: implications for women and their children. Pediatrics 2003; 111:1136 17. Weisman CS, Hillemeier MM, Chase GA, et al: Preconceptional health: risks of adverse pregnancy outcomes by reproductive life stage in the Central Pennsylvania Women’s Health Study (CePAWHS). Women's Health Issues 2006; 16:216 18. D'Angelo D, Williams L, Morrow B, et al: Preconception and interconception health status of women who recently gave birth to a live-born infant: Pregnancy Risk Assessment Monitoring System (PRAMS), United States, 26 reporting areas, 2004. MMWR Surveill Summ 2007; 56:1 19. Charron-Prochownik D, Sereika SM, Falsetti D, et al: Knowledge, attitudes and behaviors related to sexuality and family planning in adolescent women with and without diabetes. Pediatr Diab 2006; 7:267 20. Phipps MG, Nunes AP: Assessing pregnancy intention and associated risks in pregnant adolescents. Matern Child Health J 2012; 16:1820 21. Rubin V, East PL: Adolescents' pregnancy intentions: relations to life situations and caretaking behaviors prenatally and 2 years postpartum. J Adolesc Health 1999; 24:313 22. Bloom KC, Hall DS: Pregnancy wantedness in adolescents presenting for pregnancy testing. MCN Am J Matern Child Nurs 1999; 24:296
23. Rosenberg KD, Gelow JM, Sandoval AP: Pregnancy intendedness and the use of periconceptional folic acid. Pediatrics 2003; 111:1142 24. Cheng D, Schwarz EB, Douglas E, et al: Unintended pregnancy and associated maternal preconception, prenatal and postpartum behaviors. Contraception 2009; 79:194 25. Bartz D, Shew M, Ofner S, et al: Pregnancy intentions and contraceptive behaviors among adolescent women: a coital event level analysis. J Adolesc Health 2007; 41:271 26. Prinstein MJ, Boergers J, Spirito A: Adolescents' and their friends' health-risk behavior: factors that alter or add to peer influence. J Pediatr Psychol 2001; 26:287 27. Santelli J, Carter M, Orr M, et al: Trends in sexual risk behaviors, by nonsexual risk behavior involvement, U.S. high school students, 1991-2007. J Adolesc Health 2009; 44:372 28. Cogswell ME, Bitsko RH, Anderka M, et al: Control selection and participation in an ongoing, population-based, case-control study of birth defects: the National Birth Defects Prevention Study. Am J Epidemiol 2009; 170:975 29. Yoon PW, Rasmussen SA, Lynberg MC, et al: The National Birth Defects Prevention Study. Public Health Rep 2001; 116(Suppl 1):32 30. Cutler GJ, Flood A, Hannan P, et al: Multiple sociodemographic and socioenvironmental characteristics are correlated with major patterns of dietary intake in adolescents. J Am Diet Assoc 2011; 111:230 31. Fryar CD, Merino MC, Hirsch R, et al: Smoking, alcohol use, and illicit drug use reported by adolescents aged 12e17 years: United States, 1991e2004. National health statistics reports; no 15. Hyattsville, MD: National Center for Health Statistics. 2009. 32. Ogden CL, Carroll MD, Curtin LR, et al: Prevalence of overweight and obesity in the United States, 1999-2004. JAMA 2006; 295:1549 33. Bonsergent E, Agrinier N, Thilly N, et al: Overweight and obesity prevention for adolescents: a cluster randomized controlled trial in a school setting. Am J Prev Med 2013; 44:30 34. Case AP, Royle M, Scheuerle AE, et al: Birth defects, causal attributions, and ethnicity in the National Birth Defects Prevention Study. J Genet Couns 2014; 23:860 35. Cardozo ER, Dune TJ, Neff LM, et al: Knowledge of obesity and its impact on reproductive health outcomes among urban women. J Commun Health 2013; 38:261 36. Quinn GP, Murphy D, Wang H, et al: Healthy adolescent girls' perceptions of cancer-related infertility. Having cancer doesn't change wanting a baby: healthy adolescent girls' perceptions of cancer-related infertility. J Adolesc Health 2013; 52:164 37. Picciano MF, Dwyer JT, Radimer KL, et al: Dietary supplement use among infants, children, and adolescents in the United States, 1999-2002. Arch Pediatr Adolesc Med 2007; 161:978 38. Leslie LK, James S, Monn A, et al: Health-risk behaviors in young adolescents in the child welfare system. J Adolesc Health 2010; 47:26 39. McCrory C, McNally S: The effect of pregnancy intention on maternal prenatal behaviours and parent and child health: results of an Irish cohort study. Paediatr Perinat Epidemiol 2013; 27:208 40. Abma JC, Martinez GM, Copen CE: Teenagers in the United States: Sexual Activity, Contraceptive Use, and Childbearing, National Survey of Family Growth 2006-2008. Washington, DC, National Center for Health Statistics, Vital Health Statistics, 2010 41. Montgomery KS: Creating consistency and control out of chaos: a qualitative view of planned pregnancy during adolescence. J Perinat Educ 2000; 9:7
