Introduction

Adolescent pregnancy rates are closely observed by national policy makers and the public; recent reported rates of teen births are 39.1 per 1,000, which is a 9 % decline from previous years (Hamilton and Ventura 2012). Two issues are of particular concern: teenage pregnancy is associated with higher medical risks and psychosocial complications (Klein 2005), and solid evidence suggests that depression in pregnancy can have a negative impact on both newborn and mother (Marcus 2009; Bonari et al. 2004; Rahman et al. 2004). The latter is particularly relevant for teens, and several studies have concluded that depressive symptoms in pregnancy are about twice as common in adolescents than in adults (Barnet et al. 1996; Figueiredo et al. 2007), with one study indicating that as many as 68 % of pregnant adolescents attending school report depressive symptoms (Logsdon et al. 2004). Despite this high rate, it is well known that screening is not done universally and that pregnant adolescents face higher barriers for both depression screening and treatment (Logsdon et al. 2009; Koniak-Griffin et al. 1996), due to poor recognition by both patients and providers.

Research on the effect of age on depression in pregnancy has yielded contradictory results (Lancaster et al. 2010). Moreover, research on risk factors for depression in adolescent pregnancy has involved relatively small samples of about 50 to 150 participants (Hamilton and Ventura 2012; Marcus 2009; Koniak-Griffin et al. 1996). None of these studies has found significant associations between depression scores and sociodemographic factors such as age, school status, and employment and cohabitation status. On the other hand, limited social support and life stress have consistently emerged as significant predictors for depressive symptoms in pregnant adolescents (Marcus 2009; Koniak-Griffin et al. 1996; Stephenson et al. 1999; Logsdon 2004). Insecure attachment style and early parental loss have also been suggested to explain depression in teenage pregnancy as compared to adults (Figueiredo et al. 2006).

The aim of the present study was to examine potential risk factors for depressive symptoms in adolescent pregnancy in a subsample of adolescents in their late teenage years. The risk factors of interest were sociodemographic characteristics and obstetrical factors.

Methods

Participants and procedure

Five hundred nine adolescent women between 4 and 38 weeks of pregnancy were recruited from a university hospital OB-GYN clinic (N = 186) or one of six maternal health-care centers in Eastern Iowa (N = 323). All pregnant women attending regular appointments were asked to participate in the study. After completing a signed consent approved by the University of Iowa IRB, women completed the Beck Depression Inventory (BDI; Beck et al. 1961) and a demographic questionnaire. Demographic information included age, ethnicity, years of education, marital status, employment, income, and a number of items regarding obstetrical history. Information regarding psychotropic use was also obtained.

The BDI is a commonly used 21-item self-report inventory used to measure the severity of depression. Each question consists of four statements describing increasing intensities of symptoms of depression. Items are rated on a scale of 0–3, reflecting how participants have felt over the last week. Total scores range from 0 to 48 with higher scores reflecting more severe depressive symptomatology. Cronbach’s alpha in the current sample was 0.89.

Statistics

First, we performed correlation and ANOVA analysis on the demographic and obstetrical variables in relation to BDI total scores. Second, given the good ethnic diversity in the sample, we compared BDI total scores between Caucasian and minority (non-Caucasian) ethnic groups. Third, we explored the frequency and type of psychiatric medication intake among participants for whom we had data on medication history and compared total BDI scores by antidepressant use.

Results

Demographics and obstetrical history

Participants’ ages were 18–20 (mean = 19.09, SD = 0.59). The mean number of years of education was 11.60 (SD = 1.12). About half of the sample was unemployed (58.6 %), single (49.5), and living with a partner (58.3 %), and the majority self-identified as Caucasian (78 %). A large number of the participants (79 %) reported annual family incomes of \$20,000 or less (Table 1).

Table 1 Demographic and obstetric characteristics
Full size table

Notably, many of the adolescents had previous pregnancies (mean = 1.44, SD = 0.90) and a history of miscarriages (mean = 0.72; SD = 0.96); about a half had already had a full-term birth (mean = 0.46; SD = 0.60). Premature births (mean = 0.10; SD = 0.35), abortions (mean = 0.16; SD = 0.39), and stillbirths (mean = 0.03; SD = 0.18) were rare (Table 1).

BDI scores

The mean total BDI score for the sample was 10.93 (SD = 7.38). The distribution of total BDI scores (see Fig. 1) showed a normal distribution of scores around the mean and a separate tail for scores above 20. Some authors have suggested that a score of 20 should be used as the threshold for clinical depression in adolescent pregnancy (Steer et al. 1990). In our sample, 11.6 % of adolescents scored a BDI above 20. Also, using the standard BDI cutoff scores for severity, 55.9 % of subjects scored in the mild depression range (BDI 10–14), 32.3 % were moderate (BDI 15–24), and 11.8 % severe (BDI > 25).

Fig. 1
figure 1

Distribution of BDI scores in the sample

Full size image

Analysis of associations between risk factors and total BDI score

Correlation analysis was performed for interval variables in the entire sample. Those factors were age, week of pregnancy, years of education, income, number of children, and number of previous pregnancies, full-term births, miscarriages, premature births, and stillbirths. None of these factors showed any significant correlation with total BDI scores.

ANOVA analysis of differences between groups defined by a risk factor was conducted on categorical variables (predictors) with total BDI score as a dependent variable. These categories included marital status, living with partner or not, ethnicity, head of household status, annual income less than \$20,000, employment and primiparity. None of these factors were associated with total BDI scores.

Total BDI score by ethnicity status

ANOVA analysis compared total BDI scores between Caucasian and minority (non-Caucasian) ethnic status in the sample. While the minority group had a minimally higher score than the Caucasian group (11.37 vs. 10.75), the results were not statistically significant (F = 0.63; p = 0.43).

Total BDI score by medication history

The frequency of all psychiatric medication uses among participants was 4.3 % (Table 2). Participants were further divided into two groups depending on whether they were taking antidepressants or not. ANOVA analysis compared total BDI scores between the groups and found out significantly higher mean total BDI scores for participants who were taking antidepressants (BDI 14.58 vs. 10.79; F = 4.89; p = 0.028).

Table 2 Psychiatric medication use
Full size table

Conclusions

This is the first study on risk factors for depression in pregnant adolescents using a relatively large population sample. Participants had diverse ethnic backgrounds (Koniak-Griffin et al. 1996). A broad range of demographic variables were included in the analyses: age, week of pregnancy, years of education, income, number of children, marital status, living with partner or not, ethnicity, head of household status, annual income less than \$20,000, and employment. Despite the large sample size and good variance, none of these demographic factors reached a statistically significant association with depression severity scores in adolescent pregnancy. This is consistent with previous smaller studies which did not identify any specific demographic risk factors other than limited social support (Hamilton and Ventura 2012; Marcus 2009; Koniak-Griffin et al. 1996).

Beyond other studies, we examined a number of obstetrical risk factors: primiparity status, number of previous pregnancies, full-term births, miscarriages, premature births, and stillbirths. Similar to the results above, none of these factors showed any significant association with total BDI scores.

To our knowledge, reports on medication intake in relation to antepartum depression in adolescents have not been published. Our results demonstrate that antidepressant use in pregnant adolescents is not frequent, and only 3.7 % of the participants were on antidepressant medications, which is comparable to data on antidepressant use in the general pregnant population (Ramos et al. 2007; Ververs et al. 2006).

Pregnant adolescents from our sample were predominantly single and of lower socioeconomic status, and the majority had been previously pregnant, with a fairly high number of miscarriages. The presence of mixed demographic background allowed us to compare BDI scores between Caucasian and non-Caucasian participants; ethnicity was not associated with higher scores. Our data did not include information on social support, although this is the only factor shown to be significantly associated with depression severity scores in pregnant adolescents (Marcus 2009; Koniak-Griffin et al. 1996; Stephenson et al. 1999; Logsdon 2004). This suggests that demographic information is not a reasonable proxy measure of social support, so clinicians may be required to obtain additional history on this particular factor.

Our report suffers from several limitations. First, it includes adolescents in their late teenage years only. This is due to the design of the initial study from which data for this report were taken, where recruitment was performed at adult clinical centers only. Therefore, the interpretation of the results may not necessarily apply to early teen adolescents. However, other previous reports, which had included early and late teens, also did not elicit positive demographic or obstetrical risk factors for depression in adolescent pregnancy. Future studies can specifically be conducted in specialized teen pregnancy centers to encompass all adolescents as a defined group. Another important limitation of this report is the cross-sectional design and the unavailability of data on previous history of depression or the progression of depressive symptoms in follow-up visits. Other studies are under way to specifically address these questions.

In summary, we failed to find an association between BDI scores and standard demographic and obstetrical factors in pregnant adolescents in late teen years. It is likely that the nature of adolescence as a developmental stage in combination with pregnancy render a high risk for depression, regardless of the associated fixed demographic and obstetric characteristics. This suggests that, without question, universal screening for depression in pregnant adolescents should be implemented in all clinical settings.