Add Health Dataset Research Articles

Contraceptive Use and Its Associations With Social Determinants of Health Among Young Adults.

Nonuse and inconsistent use of contraception in young adults can result in undesired pregnancies. Prior research on contraceptive use has primarily focused on individual influences such as attitudes, beliefs, and skills related to contraceptive use, with little consideration of the social and economic determinants. This study explored domains from the Healthy People 2030 Social Determinants of Health (SDH) framework and examined their cross-sectional associations with young adults' self-reported contraceptive use (ie, any use in the past year and consistent use of contraception). Eleven measures related to 5 SDH domains were extracted from the Add Health dataset collected from 2001 to 2002 (Wave III). The final sample included 11,172 youth with a mean age of 21.88 years (SD: 1.84) and was 53.8% female. Measures associated with the past year's use of contraceptives included non-poverty status, employment, housing instability, high school graduation, enrollment in higher education, English language spoken at home, experience of civic participation, and access to primary care. For consistent use of contraception, associated measures were participants' non-poverty status, employment, high school graduation, enrollment in higher education, English language spoken at home, experience of civic participation, incarceration, and access to health care. Interventions that address social determinants of health can potentially promote contraceptive use.

Social determinants of health and diabetes: using a nationally representative sample to determine which social determinant of health model best predicts diabetes risk

ObjectivesSocial determinants of health (SDOH) research demonstrates poverty, access to healthcare, discrimination, and environmental factors influence health outcomes. Several models are commonly used to assess SDOH, yet there is limited understanding of how these models differ regarding their ability to predict the influence of social determinants on diabetes risk. This study compares the utility of four SDOH models for predicting diabetes disparities.Study designWe utilized The National Longitudinal Study of Adolescent to Adulthood (Add Health) to compare SDOH models and their ability to predict risk of diabetes and obesity.MethodsPrevious literature has identified the World Health Organization (WHO), Healthy People, County Health Rankings, and Kaiser Family Foundation as the conventional SDOH models. We used these models to operationalize SDOH using the Add Health dataset. Add Health data were used to perform logistic regressions for HbA1c and linear regressions for body mass index (BMI).ResultsThe Kaiser model accounted for the largest proportion of variance (19%) in BMI. Race/ethnicity was a consistent factor predicting BMI across models. Regarding HbA1c, the Kaiser model also accounted for the largest proportion of variance (17%). Race/ethnicity and wealth was a consistent factor predicting HbA1c across models.ConclusionPolicy and practice interventions should consider these factors when screening for and addressing the effects of SDOH on diabetes risk. Specific SDOH models can be constructed for diabetes based on which determinants have the largest predictive value.

Social order statistics models for ranking data with analysis of preferences in social networks

In the National Longitudinal Study of Adolescent to Adult Health (Add Health) in the United States during the 1994–95 school year, sampled high school students were asked to rank 17 romantic relationship activities in ideal chronological order and to nominate up to five male and five female best friends within the school. It is natural to see that the students’ rank-order progression of social, romantic and sexual relationships may be influenced strongly by their peers or friends. So far, traditional ranking models do not account for such social network dependency. In this article we introduce a new class of models, called social order statistics (SOS) models, to learn ranking data in social networks. The new models combine the order statistics models and spatial autoregressive models to account for social dependencies among the individuals. A flexible formulation of weight matrices in the spatial model is adopted to provide diverse network effects among the individuals for different items. Efficient and scalable MCMC algorithms are developed to perform Bayesian inference in a parallel manner for large networks with even a few thousand nodes. Analysis of the Add Health dataset reveals that social network effects are different for students’ preferences toward different activities in a relationship. In particular, students’ preferences on romantic and sexual events generally have stronger peer effects than those on social events, and opinions of close friends of the same gender tend to have a larger impact on the preferences on sexual events.

Sibling Training Hypothesis: Alcohol Use Behaviors Among Three Adolescent Siblings

ABSTRACT The sibling training hypothesis posits that younger siblings will come to resemble older siblings but has not been tested across three adolescent sibling dyads. In paired t-tests, sibling training was assessed across three sibling dyads (N = 102 families, 306 adolescents, Add Health data set, Waves 1 and 2). Evidence for drinking training appeared in the oldest/middle sibling dyad but not in dyads involving the youngest sibling. In latent congruence modeling, significant predictors of sibling drinking similarity included drinking by friends in the oldest/middle and oldest/youngest dyads. Research beyond the study of a single sibling dyad in a family is needed to better understand the generalizability of social influences, how they come about as well as consequences for adolescents and their families.

Relationship among interpersonal stressors in adolescence, social support buffers, and obesity in adulthood 14 years later.

Early life stress may be linked to adult obesity. Alternatively, social support can buffer the effects of stress. The study's objective was to examine in a nationally representative sample, whether adolescent interpersonal and financial stress predict later obesity in adulthood and whether social support and social cohesion might buffer this effect. The sample includes 6,504 participants across four waves in the Add Health dataset. Researchers created a structural equation model whereby latent measures of interpersonal stress and financial stress during adolescence were used to predict obesity (Body Mass Index, BMI ≥ 30) at Wave 4, 14 years later. Latent measures of social support and social cohesion were added to observe whether they buffered the effect of stress on adult BMI. Covariates included race/ethnicity, gender, self-rated health, smoking status, and age. In an initial model of interpersonal and financial stressors and covariates predicting BMI, interpersonal stress (p < .001) but not financial stress (p > .05) predicted adult BMI. In the full model including stressors, buffers and covariates, social support (p < .001) and social cohesion (p = .038) negatively predicted adult BMI, and they covaried with interpersonal stress in opposing directions, buffering the effects of interpersonal stress on adult BMI. Stressful interpersonal life events in adolescence such as having family members in jail or being the victim of a crime are linked to adult obesity 14 years later. However, these stress effects are buffered by the effects of social support and social cohesion, which are linked to lower adult BMI. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Adverse childhood experiences: Examining latent classes and associations with physical, psychological, and risk-related outcomes in adulthood

BackgroundAdverse childhood experiences (ACEs) are a public health crisis, affecting nearly half of children in the United States. Long-term effects of ACEs on psychological well-being, engagement in risk behaviors, and physical health have been observed. Moreover, many individuals exposed to ACEs are also affected by an accumulation of stressors due to broader structural inequities. ObjectiveThe current study examined heterogeneity in patterns of ACEs, explored how these patterns varied based on race/ethnicity, biological sex, and socioeconomic status, and assessed how ACE patterns were associated with physical health, mental health, and risk-related outcomes in adulthood. ParticipantsDrawing on the Add Health dataset, survey data from Waves I, III, IV (n = 12,288) were analyzed. Mean age of participants was 28.3 (SD = 1.9), more than half were female (54.4%), and a little less than half identified as youth of color (46.7%). MethodMultigroup latent class analysis explored heterogeneity in ACE exposure and variations based on structural inequities. Latent class regression assessed associations between ACE classes and outcomes. ResultsA four-class solution was identified. Class sizes and latent structures differed by biological sex. Among males and females, the low adversity class had more positive physical health, mental health, and risk-related outcomes compared to all classes, while the childhood maltreatment and high adversity/community violence classes engaged in more risk-related behaviors. Very small to medium effects were observed. ConclusionFindings highlight the importance of examining heterogeneity in ACE exposure, and how patterns of ACEs may differentially affect outcomes in adulthood.

50.11 Reported Discrimination Predicts Biomarkers of Adverse Medical Outcomes Using Machine Learning Methodology

Within the past decade, there has been increased attention to the relationship between stress, its physical manifestations, and complex psychosocial, interpersonal interactions. One such stressor that has been shown to significantly impact is the experience of perceived discrimination for members of African American (AA) communities. Racial trauma (or race-based stress response) is defined as the traumatic response to negative race-related experiences that are collectively characterized as racism, including acts of prejudice, discrimination, or violence against a subordinate racial group. Consistently, it has been shown that African Americans are among the hardest hit by this sociopolitical phenomenon, racism, in the United States. For African Americans, the experience of racism has been linked to earlier onset (and greater prevalence) of disability, reduced life expectancy, and various cardiometabolic, neurological, or mental health disorders than any other racial or ethnic minority group in the United States. In this study, we are interested in utilizing machine-based learning methodologies to investigate if such intrinsic factors, such as perceived prejudice or discrimination in childhood/adolescence, are predictive of poorer health outcomes in adulthood. Utilizing the National Longitudinal Study of Adolescent to Adult Health (Add Health) dataset, we utilized a recursive feature elimination regression analysis to investigate the relationship between reported discrimination and potential biomarkers of adverse health outcomes in adulthood. We found that reported discrimination was predictive of elevated levels of high-sensitivity C-reactive protein (hsCRP) (R2 = 0.2579), increased BMI, in adulthood. Our findings are proof of concept that machine learning methodology can be applied to large datasets to understand the potential health consequences of racial disparities in a hypothesis-free manner.

A two-stage Generalized Method of Moments model for feedback with time-dependent covariates.

Correlated observations in longitudinal studies are often due to repeated measures on the subjects. Additionally, correlation may be realized due to the association between responses at a particular time and the predictors at earlier times. There are also feedback effects (relation between responses in the present and the covariates at a later time), though these are not always relevant and are often ignored. All these cases of correlation must be accounted for as they can have different effects on the regression coefficients. Several authors have provided models that reflect the direct and delayed impact of covariates on the response, utilizing valid moment conditions to estimate the relevant regression coefficients. However, there are applications when one cannot ignore the effect of the responses on future covariates. A two-stage model to account for the feedback, modeling the direct as well as the delayed effects of the covariates on future responses and vice versa is presented. The use of the two-stage model is demonstrated by revisiting child morbidity and its impact on future values of body mass index using Philippines health data. Also, obesity status and its feedback effects on physical activity and depression levels using the Add Health dataset are analyzed.

Specification and Estimation of Network Formation and Network Interaction Models with the Exponential Probability Distribution

In this paper, we model network formation and network interactions under a unified framework. The key feature of our model is to allow individuals to respond to incentives that stem from interaction benefits of certain activities when they choose friends (network links), while capturing homophily in terms of unobserved characteristic variables in network formation and activities. There are two advantages of this modeling approach: first, one can evaluate whether incentives from certain interactions are important factors for friendship formation or not. Second, in addition to homophily effects in terms of unobserved characteristics, inclusion of incentive effects in the network formulation also corrects possible friendship selection bias on activity outcomes under network interactions. A theoretical foundation of this unified model is based on a sub-game perfect equilibrium of a two-stage game. A tractable Bayesian MCMC approach is proposed for the estimation of the model, and we demonstrate its finite sample performance in a simulation study. We apply the model to study empirically American high school students' friendship networks from the Add Health dataset. We consider two activity variables, GPA and smoking frequency, and find a significant incentive effect from GPA, but not from smoking, on friendship formation. These results suggest that the benefit of interactions in academic learning is an important factor for friendship formation, whereas the interaction benefit of smoking is not. On the other hand, from the perspective of network interactions, both GPA and smoking frequency are subject to significant positive interaction (peer) effects.

Racial Disparities in Arrests: A Race Specific Model Explaining Arrest Rates Across Black and White Young Adults

Three theories attempt to explain the racial disparities in arrest between White and Black Americans: Differential Involvement Hypothesis, Differential Selection and Processing Hypothesis, and Social Disorganization Theory. We tested these hypotheses simultaneously in a multiple-group longitudinal panel model with the ADD Health dataset (Black n = 2459, Whiten = 7403). After controlling for contextual and behavioral factors, we still found Black young adults were arrested seven times more often than their White counterparts. To maintain cultural competence, it is imperative for clinicians to be aware of these disparities when working with families of different races in order to adjust treatment accordingly, but advocacy for greater systemic change may be more important for some communities than therapy alone.

Specification and estimation of network formation and network interaction models with the exponential probability distribution

We model network formation and interactions under a unified framework by considering that individuals anticipate the effect of network structure on the utility of network interactions when choosing links. There are two advantages of this modeling approach: first, we can evaluate whether network interactions drive friendship formation or not. Second, we can control for the friendship selection bias on estimated interaction effects. We provide microfoundations of this statistical model based on the subgame perfect equilibrium of a two‐stage game and propose a Bayesian MCMC approach for estimating the model. We apply the model to study American high school students' friendship networks using the Add Health dataset. From two interaction variables, GPA and smoking frequency, we find that the utility of interactions in academic learning is important for friendship formation, whereas the utility of interactions in smoking is not. However, both GPA and smoking frequency are subject to significant peer effects.

Identifying Different Ways People Change: A Latent Basis Growth Mixture Model Example Identifying Nonlinear Trajectories of Binge Drinking

Our field of couple and family therapy focuses on change, different ways people change, and how therapists can facilitate change. Change can be modeled as an average trajectory-growth curve model-or as multiple classes of trajectories-growth mixture model. The field of MFT has not yet fully embraced implementing more advanced longitudinal modeling procedures to study what we care about most, change across time. To support our field moving in this direction, we provide a step-by-step description and example in Mplus software. Our example analysis used N=5,958 participants from the Add Health dataset, to identify unique classifications of trajectories of binge drinking. We discuss how these analytical methods provide increased options to advance family science and clinical research.

Gender and Race Informed Pathways From Childhood Sexual Abuse to Sexually Transmitted Infections: A Moderated Mediation Analysis Using Nationally Representative Data

PurposeAlthough data suggest child sexual abuse is linked with increased risk of contracting asexually transmitted infection (STI), the mechanisms through which these experiences are connected remain understudied. Moreover, there is a need to explore how race/ethnicity and gender influence these processes. MethodsThe present study examined the mediational pathways from child sexual abuse to risky sexual behavior to STIs and further evaluated the role of depressive symptomatology and nonmedical prescription drug use on the relationship between child sexual abuse and risky sexual behavior. In addition, race and gender were examined as moderators to account for potential different effects of these mechanisms on females and males and on different racial and ethnic groups. A nationally representative sample of 4,181 youth from the Add Health dataset was used. ResultsResults from a moderated mediation model indicated risky sexual behavior partially mediated the pathway from child sexual abuse to STI contraction and depressive symptomatology and nonmedical prescription drug use partially mediated pathway from child sexual abuse to risky sexual behavior. Race and gender moderated the relationship between risky sexual behavior and STI contraction. ConclusionsFindings underscore the need for STI prevention efforts among adolescents to focus on risk factors beyond risky sexual behaviors, such as childhood sexual abuse and mental health screening that includes depressive symptomatology and nonmedical prescription drug use. In addition, findings emphasize the need to further examine the different effects on different racial/ethnic and gender subgroups, particularly black women.

The Role of Poverty and Income in the Differential Etiology of Violence: An Empirical Test

ABSTRACTPoverty has been linked with criminal behavior in theory and empirical research. The authors test the differential etiology of violence thesis using a sample of teenagers from the Add Health data set. Employing a three-pronged test and conservative models, the authors find that income is associated with variability in violent offending, even controlling for nonviolent offending. The family incomes of nonviolent-only offenders were comparable to those of nonoffenders. The findings call into question general theories of criminality that do not distinguish between violent and other forms of offending and support the proposal that poverty is differentially associated with violent crime.

Multi-contextual influences on adolescent pregnancy and sexually transmitted infections in the United States

RationaleRates of adolescent pregnancy and sexually transmitted infections (STIs) in the United States remain high. Norms and beliefs about sex and the use of contraception have been identified as potential contributors to these. ObjectiveThe current study examined multi-contextual links between norms and beliefs about sex and contraception, and adolescent pregnancies and STIs. MethodHierarchical linear modeling (HLM) was used to test the study questions in the historical data from Add Health dataset (Wave I, 1994–1995, and Wave II, 1995–1996). Measures of beliefs about sex and contraception were administered to adolescents and their caregivers; pregnancies and STIs were reported by adolescents (N = 13,568; level 1). School-reported data (N = 132; level 2) included sexuality education and prevalence of pregnancies at schools. Community data included the density of family planning providers in each county. ResultsAdolescents negative beliefs about contraception were associated with a higher likelihood of pregnancies and STIs via their association with contraception use. The opposite effect was found for negative beliefs about sex. Parental disapproval of contraception was associated with a decreased likelihood of STIs, but with an increased likelihood of pregnancies. A greater number of pregnancies at school was associated with a greater likelihood of STIs. Unexpectedly, the number of county-level family planning providers was associated with a higher likelihood of STIs. ConclusionAdolescent individual beliefs emerged as the most salient predictors of both pregnancies and STIs.

On the prevalence of racial discrimination in the United States.

Boutwell, Nedelec, Winegard, Shackelford, Beaver, Vaughn, Barnes, & Wright (2017) published an article in this journal that interprets data from the Add Health dataset as showing that only one-quarter of individuals in the United States experience discrimination. In Study 1, we attempted to replicate Boutwell et al.’s findings using a more direct measure of discrimination. Using data from the Pew Research Center, we examined a large sample of American respondents (N = 3,716) and explored the prevalence of discrimination experiences among various racial groups. Our findings stand in contrast to Boutwell et al.’s estimates, revealing that between 50% and 75% of Black, Hispanic, and Asian respondents (depending on the group and analytic approach) reported discriminatory treatment. In Study 2, we explored whether question framing affected how participants responded to Boutwell’s question about experiencing less respect and courtesy. Regardless of question framing, non-White participants reported more experiences than White participants. Further, there was an interaction of participant race and question framing such that when participants were asked about experiences of less respect or courtesy broadly, there were no differences between non-White participants and White participants, but when they were asked about experiences that were specifically race-based, non-White participants reported more experiences than White participants. The current research provides a counterweight to the claim that discrimination is not a prevalent feature of the lives of minority groups and the serious implications this claim poses for research and public policy.

Long-Term Trends of Participation in Physical Activity During Adolescence With Educational Ambition and Attainment.

Insufficient physical activity is a significant concern because a growing body of research demonstrates that physical activity during adolescence has numerous benefits on physical health, mental health, and educational achievement. A less-studied area of physical activity research is how physical activity participation in adolescence relates to educational and career aspirations and attainment. Using the Add Health dataset consisting of over 15,000 participants, this study addressed several research questions relating to the long-term benefits of habitual physical activity in adolescence. Physical activity levels in adolescence and associations with activity participation in later adolescence and young adulthood were explored as were relationships with physical activity participation in adolescence and educational ambition and attainment in young adulthood. Moderator variables including adolescent sex, race, and age were also examined. Results indicated several significant relationships between physical activity in adolescence and future physical activity and educational attainment outcomes. Given short- and long-term benefits of physical activity, school personnel and parents should ensure children are meeting the required activity guidelines for children and find ways to keep children active as they approach adolescence.

Self-medication Hypothesis and Family Socialization Theory: Examining Independent and Common Mechanisms Responsible for Binge Drinking.

For many, binge drinking behaviors start early and become a persistent pattern of use throughout the lifespan. In an effort to strengthen understanding of etiology, this study considered the mechanisms from the self-medication hypothesis and family socialization theory. The goal was to identify whether emotional distress is a potential shared mechanism that accounts for the development of binge drinking in different developmental periods. This study used the National Longitudinal Study of Adolescent to Adult Health (Add Health) dataset to examine binge drinking across time for n=9,421 participants ranging in age from 11 to 18 (M=15.39, SD=1.62) at Wave I and ranging from 24 to 32 (M=28.09, SD=1.61) at Wave IV of the study. Using an autoregressive cross-lagged model, I examined how parent-child closeness, depressive symptoms, and binge drinking were related over three developmental periods. In examining cross-sectional and longitudinal relations, depressive symptoms were significantly related to binge drinking more often than parent-child closeness; however, results indicated the self-medication model may primarily account for concurrent drinking behaviors rather than long-term. The family socialization theory was indicated to account for some variability above and beyond the self-medication hypothesis. No indirect association between binge drinking and the parent-child relationship was detected through depressive symptoms, failing to support a shared mechanism between the two theories. The results provide support for a multifaceted assessment process for substance using clients, and support the use of Multisystemic Family Therapy, Multidimensional Family Therapy, and perhaps Attachment-Based Family Therapy.

Genetic and environmental effects on the development of depressive symptoms from adolescence to adulthood in a nationally representative sample

BackgroundPrevious research has shown substantial heritability for depressive symptoms, yet, there are few genetically-informed studies which focused on developmental changes. The current study sought to model prototypical developmental trajectories of depressive symptoms from adolescence to adulthood and to elucidate genetic and environmental contributions to these changes. MethodsThe Add Health data set, a nationally representative sample of adolescents, was used. For the genetically-informed analyses, a subsample of N = 531 same-sex monozygotic and dizygotic male and female twin pairs was selected. Longitudinal development was modeled separately for two waves in early adolescence and for four waves from middle adolescence to young adulthood using a latent growth model (LGM). Both models were extended to twin models to estimate the effects of heritability and the environment. ResultsThe rates of depressive symptoms peaked in mid-adolescence and then sharply declined as individuals moved from adolescence to young adulthood, leveling off in the twenties. The effects of the shared environment were substantial among early adolescents, but negligible for middle-to-late adolescents. An opposite pattern was found for heritability. The largest proportion of developmental changes was driven by nonshared environmental effects. LimitationsThe study only used same-sex twins as there exist mixed findings regarding the possibility of qualitative or quantitative genetic effects. ConclusionsThe salience of unique experiences and, to a lesser extent, heritable factors in affecting developmental changes in depressive symptoms underscore the need for targeting environments that place individuals with genetic predisposition at double the risk for the development of depression.

The Long-Term Impact of Natural Mentoring Relationships: A Counterfactual Analysis.

Previous research suggests that youth's natural mentoring relationships are associated with better academic, vocational, and psychosocial functioning. However, little is known about the extent to which the impact of mentoring endures beyond adolescence and early adulthood. Furthermore, most natural mentoring research is confounded by selection bias. In this study, we examined the long-term impact of mentoring using the nationally representative, longitudinal Add Health dataset. We conducted counterfactual analysis, a more stringent test of causality than regression-based approaches. Compared to their unmentored counterparts, adults (ages 33-42) who had a natural mentor during adolescence or emerging adulthood reported higher educational attainment, more time spent volunteering, and more close friends, after controlling for a range of confounding factors. However, outcomes differed when mentors were classified as "strong ties" (e.g., grandparents, friends) or "weak ties" (e.g., teachers, coaches, employers). Having a strong-tie mentor was associated with having more close friends and a lower income. In contrast, having a weak-tie mentor was associated with higher educational attainment, higher income, and more time spent volunteering. These findings suggest that natural mentoring relationships can exert lasting influence on young people's developmental trajectories, providing strong rationale for efforts to expand their availability and scope.