Educational goals and achievements are likely to be particularly important for adolescents, as school is their major social and occupational context . However, most studies that linked maternal education with birth outcomes used maternal educational attainment in adulthood or at the child’s birth [8–12, 14]. Few studies have explored whether academic performance and educational aspiration during adolescence are associated with birth outcomes among adolescent mothers. In this study, data were collected prospectively, which allowed us to assess academic performance and educational aspiration prior to the teen pregnancy.
Multivariable analysis indicated that higher aspiration for college education was related to higher birthweight among non-Black adolescents. Higher educational aspiration could indicate a greater future orientation and greater social support  which would be associated with healthier pregnancies through health-protective behaviors, such as good nutrition and physical activity . On the other hand, girls who do not have high aspiration for their educational futures may be reflecting a family or community not generally supportive of the girl or her goals, which might translate to reduced social support during pregnancy . It also might indicate their life circumstances, which could include poor health or a community that does not have many people going to college [46, 47]. This may be part of a fatalistic attitude towards pregnancy and pregnancy outcomes, which might lead to poorer self-care.
We found that grade skipping was associated with higher birthweight and gestational age among non-Black girls, as well as higher birthweight among Black girls. Grade skipping is one way to accomplish the concept of appropriate developmental placement [27, 48]. It allows intellectually precocious students to experience more developmentally appropriate content by skipping over what they have already known or can easily learn . It has been suggested to have merit as the intellectually talented students potentially gain intellectual development, interpersonal maturity and time when they advance more quickly through the educational system . It is possible that adolescents who were accelerated had greater social support  and personal resilience and thus have healthier lifestyles and behaviors during pregnancy which lead to better birth outcomes.
No statistically significant relationship was found between adolescents’ GPA and either of the birth outcomes. However, we observed that among non-Black adolescents, those whose GPA was below average had higher birthweight and gestational age than those with GPA above average and the magnitude of the coefficients was larger than that for grade skipping. It is possible that this finding is (partly) due to the imperfect measurement of adolescents’ GPA; adolescents may have mis-reported their letter grades in Wave I in-home interview. Nonetheless, this was an unexpected finding. We tested the interaction between GPA and grade skipping, as well as interaction between GPA and grade repeating, to determine whether a low GPA represented more challenging courses relative to age, but neither was statistically significant (p > 0.05).
Past studies suggested that the effect of maternal education on birth outcomes could be accounted for by the health-related risk behaviors, particularly maternal smoking and alcohol use during pregnancy and late entry into prenatal care [11, 14, 15, 19]. However, in our study the addition of smoking during pregnancy and prenatal care visit did not change the regression coefficients of the education variables, thus there was little evidence of a mediating effect of these factors. Among Black girls, we were unable to examine the mediating effect of alcohol use during pregnancy due to small cells (i.e. only four girls ever used alcohol during pregnancy). Among non-Black girls, the addition of alcohol use during pregnancy did not change the regression coefficients of the education variables (See Additional file 1: Table S1). There was no relationship between education variables and potential mediator variables. It would be interesting to explore alternative pathways that might explain the effect of adolescent educational aspiration and grade skipping on birth outcomes.
Surprisingly, we found that smoking during pregnancy was positively related to Black adolescent mothers’ birthweight and gestational age. This finding is inconsistent with extensive literature documenting increased risk of adverse birth outcomes with smoking during pregnancy . It is possible that this unexpected result is due to the very small number of Black adolescent mothers who reported any smoking during pregnancy (n = 14), which may lead to spurious results. It may also be due to sample selection biases, since our sample was limited to mothers who experienced a pregnancy as an adolescent and had a singleton live birth, or reporting biases. Given these potential issues with the ascertainment of smoking during pregnancy, results should be interpreted with caution.
In the post-hoc analysis, we found that lower GPA was related with a higher risk of teen pregnancy in both Black and non-Black girls, and that higher educational aspiration was related with a lower risk of teen pregnancy in non-Black girls. Our findings were consistent with existing literature which suggests that better grades in school and/or higher educational aspirations are associated with postponed sexual activities [52–55]. On the other hand, Tucker et al. did not find a significant association between parental educational expectations and teen pregnancy using data from Add Health . However, the authors used parents’ educational expectations  while we examined adolescents’ own educational aspiration. Among non-Black girls, those who ever repeated a grade were more likely to experience teen pregnancy. This is consistent with evidence from existing studies that students who are over-age for grade experience a higher level of emotional stress and more behavioral problems, including substance abuse and early sexual debut [57–59]. We found that grade skipping was related with a higher risk of having teen pregnancy among non-Black girls, which is an unexpected finding. There were 24 non-Black girls who ever skipped a grade and had teen pregnancy. With this relatively small cell size, the possibility that this finding is due to chance cannot be ruled out. These results suggest that adolescent girls’ academic performance and educational aspiration influence both the likelihood of getting pregnant as a teenager and birth outcomes of teenage pregnancy.
Our findings suggest that interventions that affect adolescents’ educational aspirations and grade acceleration might have positive effects on birth outcomes should the adolescent become pregnant. It has been found that both personal and social attributes are important factors that influence educational aspiration and grade acceleration in adolescent students [47, 60, 61]. Adolescents’ liking and enjoyment of school as well as a sense of affinity increases the probability of putting further education into the plan. A friendly atmosphere at school and perceived support from parents are also influential with regard to adolescents’ educational motivation . Wells et al.  found that children with parents who more often discuss school programs, school activities, and things studied in class were more likely to have grade acceleration. Thus, the development of a supportive family and school environment is important in stimulating educational aspiration and promoting academic performance, which also has the benefit of being positively associated with birth outcomes among adolescent mothers. In addition, efforts should be made to identify and place students in appropriate learning environment/grades [27, 50].
Our study used a large, nationally-representative dataset. Measurement of exposures was made prior to outcome data, which allows for the temporal inference between academic performance and educational expectation and the birth outcomes. Despite these strengths, our study has several limitations. First, we used adolescents’ self-reported letter grades rather than school transcript, which would have provided a more accurate and reliable measure of adolescents’ academic performance. The measurement of grade retention was also subject to reporting bias. Second, we reduced our sample size by requiring complete covariate data, which could bias our study results. Analyses comparing included (n = 763) versus excluded (n = 215) adolescents indicated no significant difference in birthweight (p = 0.71), gestational age (p = 0.62), race (p = 0.10), age at pregnancy (p = 0.16), parental education (p = 0.10), GPA (p = 0.41), grade skipping (p = 0.20), adolescents’ educational aspiration (p = 0.15), smoking during pregnancy (p = 0.47), and prenatal care visit (p = 0.17). However, there were significant differences in baseline age (15.33 ± 0.14 vs. 15.91 ± 0.20 in included and excluded sample, respectively; p < 0.001), baseline BMI (22.07 ± 0.19 vs. 22.94 ± 0.44 in included and excluded sample, respectively; p = 0.03), and percent of grade repeating (20.18% vs. 38.35% in included and excluded sample, respectively; p < 0.001). Third, we relied on maternal self-reports of both birthweight and gestational age, which may be subject to recall biases. However, past studies comparing birth records to maternal self-report have found maternal reports of birth outcomes are largely a valid method of obtaining data .