National Partnership for Women & Families

Monthly Women's Health Research Review

Researchers examine relationship between adolescent pregnancy, social determinants of health

Summary of "Social determinants of health and adolescent pregnancy: An analysis from the national longitudinal study of adolescent to adult health," Maness et al., Journal of Adolescent Health, June 2016.

"Although rates of adolescent pregnancy in the United States are at historic lows, marked variations exist across racial, ethnic, and geographic populations," according to Sarah Maness of the University of Oklahoma's Department of Health and Exercise Science and colleagues. The researchers noted U.S. rates of adolescent pregnancy and birth are "substantially higher than in other developed countries," and within the United States, "black and Hispanic adolescents have more than twice the pregnancy rates compared with white adolescents."

The researchers noted that "adolescent pregnancy prevention efforts in the United States [historically] have been based on individual and interpersonal levels of behavior change," resulting in government funded programs that "tend not to take into account the larger social factors -- or social determinants of health [SDoH] -- that may play a critical role in impacting pregnancy." The researchers defined SDoH "as the differences in social conditions that lead to health inequities." According to the researchers, assessing the relationship between SDoH and teenage pregnancy rates "may identify and alter factors contributing to adolescent pregnancy that are not feasible with individual or interpersonal behavior change interventions."

For the study, the researchers assessed whether there was an "association between adolescent pregnancy and each element of the SDoH," the strength and direction of any association and whether that association remains "after controlling for additional factors."

Methods

For the study, researchers used data from the National Longitudinal Study of Adolescent to Adult Health, or Add Health, which followed several thousand students at 80 high schools in the United States. Add Health began the study in 1994-1995, when participants were in grades 7-12, and followed up through 2009 with in-home surveys of students and students' parents. Maness and colleagues examined data from a subgroup of 9,204 study participants who had "reported information regarding pregnancy history and answered all questions representing variables in the present study."

According to the researchers, the Add Health data also included information on "whether a participant experienced a pregnancy before the age of 18 years."

The researchers looked at Add Health survey data on the SDoH "areas of education, economic stability, social and community context, health and health care, and neighborhood and built environment." The researchers examined these factors via proxy measures "to answer ... whether adolescent pregnancy is associated with a number of SDoH." The proxy measures included:

  • Poverty and employment status to evaluate the SDoH of economic security;
  • "[H]igh school graduation, school policies that support health promotion, school environment, and enrollment in higher education" to evaluate the SDoH of education;
  • "[F]amily structure, social cohesion, perceptions of discrimination and equity, civic participation, and incarceration" to evaluate the SDoH of social and community context;
  • Health care access and primary care access to evaluate the SDoH of health and health care; and
  • "[C]rime and violence and environmental conditions" to evaluate the SDoH of neighborhood and built environment.

Results

The researchers had a final sample of 9,204 study participants, most of whom were white (57.2 percent) and female (53.5 percent). Participants reported a mean age of 15.8 years old at the beginning of the study and by the end of the study, 403 participants reported experiencing a pregnancy before age 18.

In bivariate testing, the researchers found that "[n]ine of 17 SDoH variables showed a statistically significant relationship with adolescent pregnancy." Those variables included "three items from the parental survey, including receiving public assistance ... being unemployed but looking for employment ...and reporting trash as a problem in the neighborhood environment." The other related variables included factors "in the areas of education and social and community context including receiving a high school diploma ... enrollment in higher education ... and living in a single parent household." Additional factors included regularly participating in community service, not receiving needed medical care within the last year and being involved in acts of violence within the last year.

When the researchers conducted another analysis controlling for age, race and sex, they found a "statistically significant association of 6 of the 17 SDoH predictors with adolescent pregnancy." The researchers found variables that were negatively associated with teenage pregnancy under "the SDoH areas of social and community context and education," as well as a variable with positive association under neighborhood and built environment. According to the researchers, "None of the measures from the SDoH areas of economic stability and health and health care had statistically significant results" in the second analysis.

In terms of social and community context, the researchers found that study participants who reported living in a two-parent home and those who reported participating regularly in community service before age 18 both "had .72 lower odds of reporting an adolescent pregnancy." The researchers also found that "[i]n comparison with those who strongly did not feel like a part of school, participants who gave responses in all other categories had lower odds of reporting an adolescent pregnancy."

In terms of neighborhood and built environment, the researchers found that participants whose parents reported litter in the neighborhood as a "big problem" were less likely than those whose parents said it was "no problem at all" to report an adolescent pregnancy.

In terms of education, the researchers found that participants who reported graduating from high school and those who reported enrolling in higher education were less likely than those who did not graduate from high school or enroll in higher education to report an adolescent pregnancy.

Factors that did not have any statistical significance in the second analysis included prejudice among peers, involvement in the criminal justice system, reporting trash as a "small problem," measures of crime and violence and "[s]chool policies promoting health or feeling safe in school."

Discussion

"Overall, our findings support the relationship between adolescent pregnancy and several measures of SDoH," the researchers wrote, noting that the "findings provide support for specific areas, particularly in relation to education and social and community context, in which to focus resources and interventions in adolescent pregnancy."

According to the researchers, the majority of their findings align with prior research, particularly on how "areas related to education, community involvement, and family structure ... reduce adolescent pregnancy." The researchers noted that other areas, such as poverty and incarceration, that were not statistically significant in the study "have shown mixed findings in prior research," underscoring a need for further studies.

The researchers highlighted the unexpected negative relationship between reporting neighborhood litter as a big problem and adolescent pregnancy, noting that prior research has "linked aspects of neighborhood to adolescent sexual behavior, although not specifically to pregnancy." The researchers added that because the perception of litter is subjective, varying by person and location, "the measure of trash in one's neighborhood may not adequately measure aspects of environmental conditions related to adolescent pregnancy."

The researchers also addressed how the "measures of the SDoH area of health and health care did not produce significant results." According to the researchers, this finding could reflect a lack of specificity in the data methods used in the Add Health survey. For example, respondents were asked about whether they received a physical within the past year, but were not asked about whether they received sexuality education.

In addition, the researchers noted that their finding that economic security did not have a statistically significant relationship with adolescent pregnancy countered prior research, which has largely "relat[ed] poverty to adolescent pregnancy." The researchers noted that the difference could stem from discrepancies in how their study measured economic security compared with how other studies have measured it, and they recommended further research "into the use of measures and potential differences in study populations."

According to the researchers, the study's finding that there was not a statistically significant relationship between incarceration and adolescent pregnancy aligned with prior research, although at least one study has found a positive relationship between the incarceration of a family member and adolescent pregnancy.

The researchers wrote that their findings showing a negative association between adolescent pregnancy and feeling like part of one's school or participating regularly in community service "reflect elements of the past research related to positive youth development and adolescent pregnancy prevention." They noted, "This research gives to previously established associations with adolescent pregnancy, enhances the literature in these areas, and lends support to further exploring the pathways that link these areas to individual behavior leading to adolescent pregnancy."

The researchers stated that the "study can inform future adolescent health research and indicate areas of SDoH with associations to adolescent pregnancy that can be considered for federal funding priorities," particularly "areas of education and social and community context." They called for further research into the study's findings and into "the pathways between SDoH to individual behavior," as well as for "consensus on the definition of SDoH." According to the researchers, "Results of this research can also provide support for the development and tailoring of existing interventions based on evidence of the relationship between specific areas of SDoH and adolescent pregnancy."