The extramurally-funded projects listed here examine the mechanisms involved in self-regulation and how multidisciplinary factors, at multiple levels, interact to predict substance abuse and other risky behaviors that result from regulatory failure.

Early Truancy Prevention Project

PRINCIPAL Investigator: Philip Cook
Funding Source: Institute of Education Sciences (R305A1205261)

The Early Truancy Prevention Project (ETPP) is an innovative intervention project designed to prevent adolescent problem behaviors by improving student attendance in the early grades. In contrast to most truancy programs that rely heavily on social workers or truancy officers, the ETPP recognizes elementary teachers as important change agents in truancy prevention. There is substantial evidence that teacher-child relationships, teacher attitudes, home-school relations and student engagement are important predictors of student outcomes. Therefore, the ETPP targets these mediating mechanisms through three intervention components:

  1. universal teacher home visiting to promote positive home-school relations;
  2. early identification and assessment of “target” students at-risk for truancy;
  3. teacher-led individualized attendance interventions for target students.

Based on pilot data, a “typology” of proximal causes of elementary truancy has been developed by the research team to guide attendance assessment and individualized intervention efforts by teachers. Initial findings from this pilot study will support the expansion of early truancy prevention programs in the United States. The C-StARR provides expertise in measurement selection that promotes synergy with the PAC project (see below) and allows for the exploration of whether self-regulation skills moderate or mediate the effects of the intervention.

Vulnerability to Drug Abuse: Pathways to Recovery

PRINCIPAL Investigator: Jane Costello
Funding Source: National Institute on Drug Abuse (R01DA011301)

The Great Smoky Mountains Study (GSMS) is a prospective, longitudinal study of factors underlying the development of substance use disorders (SUD). The study has been ongoing since 1992, and has produced over 10,000 interviews of 1420 subjects. Three cohorts (49% girls), aged 9, 11, and 13 at intake in 1993, were originally selected from the population of some 20,000 children in 11 counties in western NC using a household equal probability approach.

The study includes 350 American Indian youth, oversampled because they are an understudied group known to be at high risk for psychiatric problems including SUD. Data collection includes substance use disorders, comorbid psychiatric disorders, risk and protective factors, and measures of self-regulation. In addition, finger-prick blood spot samples were obtained at each assessment and these samples have been used to assay sex steroids (testosterone, estrodial, and androstendione), HPA axis (cortisol and cortisol reactivity), immune (EBV antibodies) and inflammation markers (C-reactive protein). DNA has also been extracted from these dried blood samples for genome-wide genotyping (completed in 2010). A sub-study of 160 subjects used fMRI to examine self-regulation in a gambling context, comparing youth with persistent SUD to those with adolescent limited SUD and a control group with low levels of SUD throughout development.

The C-StARR supports integration of the computer-based self-regulatory measures from GSMS with those collected by De Bellis in N-CANDA, a longitudinal study involving adolescent imaging (see below).

National Consortium on Alcohol and NeuroDevelopment in Adolescence: Duke

Principal Investigator: Michael De Bellis
Funding Source: National Institute on Alcohol Abuse and Alcoholism (U01AA021682)

The National Consortium on Alcohol and Neurodevelopment in Adolescence (N-CANDA): Duke is one of five Research Components for the conduct of a multisite longitudinal study to address factors underlying the development of alcohol and substance use, misuse, and disorders (SUD) in adolescents. This study is being completed as part of a consortium of five universities, including the University of Pittsburgh, University of San Diego, SRI International (Menlo Park, CA) and Oregon Health and Science University. Fifty percent of the sample will be representative of the community while the other 50% will be high in risk factors for alcohol and substance disorders but have not yet been diagnosed with problem substance use behaviors. At each of the four yearly waves, subjects will receive detailed assessments that measure substance use, misuse, and disorders, comorbid psychiatric disorders, pubertal maturation, risk and protective factors, sleep measures, executive function, and measures of self-regulation. Subjects will also receive a detailed MRI brain scan that includes, anatomical MRI, resting state conductivity, diffusion tensor imaging and an fMRI Reward Modulated Anti-saccade Paradigm to measure decision making and reward systems.

The resources of the C-StARR enable collaboration between the GSMS and the N-CANDA: Duke studies to complete the first longitudinal study of DNA methylation markers associated with SUD trajectories.

Community Prevention of Child Maltreatment

Principal Investigator: Ken Dodge
Funding Source: National Institute of Child Health and Human Development (R01HD0699813)

This project follows a representative sample of 4,780 families giving birth, from birth through age 5, to test impact of an early intervention and to identify processes in child development. A randomized controlled trial tests the efficacy of the Durham Connects (DC) brief postpartum nurse home-visiting program. In DC, a nurse screens for risk factors, including substance abuse, uses motivational interviewing to guide parents toward treatment, and implements short-term interventions to enhance parenting skills and self-regulation. Data collection follows families’ receipt of a variety of community services (e.g., substance abuse treatment, economic supports, health care, parent groups) for five years. This study tests the efficacy of random assignment to DC on parent mental health and infant outcomes and a model of how parent risk factors (including substance abuse) alter parenting behavior and child development of self-regulation. Hospital record findings at age 12 months (Dodge et al., 2012) indicate that infants in families randomly assigned to DC spend 60% fewer overnights in the hospital, saving large costs and possibly reducing risk. Over the long run, children of substance-abusing parents are at risk for substance-use disorders, and so this project will inform the developmental psychopathology of substance use and the prevention of substance abuse.

The C-StARR provides support in designing measurement of hormonal processes in self-regulation, in collecting DNA, and for follow-up studies examining the impact of the intervention on brain processes in self-regulation.

Parenting, Adolescent Self-Regulation and Risk-Taking Across Cultures

Principal Investigator: Jennifer Lansford
Funding Source: National Institute of Child Health and Human Development (R01HD0548054)

The Parenting Across Cultures (PAC) project aims to understand how risk-taking develops across adolescence as a function of biological maturation (puberty and age) and socialization (parenting and culture). Project staff members have assessed families annually through interviews with mothers, fathers, and children about the parent-child relationship; the child’s adjustment; attitudes and beliefs; and cultural values. During the next five years (when children will be ages 12-16), interviews will be conducted annually with mothers, fathers, and adolescents to assess their relationships and adolescents’ risk-taking. Adolescents’ reward-seeking, self-regulation, social information processing, and risk-taking are assessed via a computerized battery administered at ages 12, 14, and 16. These data will be used to test hypotheses that risk-taking behavior during adolescence is influenced by biological, parenting, and cultural forces; that these influences are mediated, at least in part, by psychological constructs of reward-seeking, self-regulation, and social information processing; and that these influences will be moderated by cultural normativeness of parenting behaviors and culturally-shaped opportunity for risk-taking.

The C-StARR provides data management support and statistical expertise, in addition to enhancing the measurement of biomarkers via the related pilot project “Neural Mechanisms in Adolescent Self-Regulation Across Cultures.”

Triangle Census Research Network

Principal Investigator: Jerome Reiter
Funding Source: National Science Foundation (SES11318975)

The goals of the Triangle Census Research Network (TCRN) are to develop broadly-applicable methodologies that transform and improve data dissemination practice. It focuses on three interrelated areas:

  1. disseminating public use data with high quality and acceptable disclosure risks
  2. handling missing data and correcting faulty data in large complex surveys, and c) integrating information from multiple data sources
  3. integrating information from multiple data sources

As examples, the TCRN is developing flexible, nonparametric Bayesian models built specifically for high dimensional data with longitudinal and multi-level aspects, such as several of the longitudinal studies led by C-StARR faculty members. We use these models to fill in missing values or simulate replacement values for confidential data. The TCRN is improving current practice for handling missing and faulty data by developing frameworks for simultaneous imputation of missing data and editing of faulty data by integrating paradigms from statistics and operations research. The TCRN is improving methodology for integrating information across sources by developing methods that data owners and secondary analysts can use to properly account for uncertainty in inferences in imperfect record linkage settings, as well as to pass on that uncertainty in public use data products via multiply-imputed datasets.

The C-StARR facilitates access to rich epidemiological and educational data sources, allowing the TCRN to implement and enhance their methodological developments across other projects in the research base.

Behavioral and Genomic Mechanisms Linking Childhood Violence Exposure to Health: The E-Risk Twin Study

Principal Investigators: Avshalom Caspi, Terrie Moffitt
Funding Source: National Institute of Child Health and Human Development (R01HD077482)

The E-Risk Twin Study tests whether young people who are exposed to violence in childhood and adolescence will, by age 18, show compromised neuropsychological functioning, telomere erosion, increased inflammation, and/or epigenomic changes. The study follows a 1994-95 birth cohort of 2,232 twins born in Great Britain. The children have been studied repeatedly since birth. We are now collecting new data and updating our biobank at age 18 years. With these new data, we are in a unique position to discover how childhood and adolescent exposure to violence becomes biologically embedded by the time exposed youth reach adulthood. A key need is for violence exposure to be disentangled from associated risk factors, including poverty, parents’ mental health, and genetic liability. Analyses with the E-Risk cohort will isolate these effects of violence exposure by (a) comparing twins who are discordant for violence exposure on their psychobiological outcomes, (b) studying within-individual changes in psychobiological outcomes, using each child as his or her own control, and ultimately (c) identifying protective factors that mitigate adverse effects.

The C-StARR provides research ideas through interaction with the network of C-StARR scientists. It also provides support for postdoctoral researchers to analyze and write up life-course data from the New Zealand and British longitudinal cohort studies on the etiology, course, and consequences of self-regulation.

The Resilience Project

Principal Investigator: Timothy Strauman
Funding Source: The Duke Endowment (Grant Agreement No. 13-EDU-MS-01-SP)

One of the primary responsibilities of colleges and universities is to prepare people not only with the knowledge and skills that they will need to be successful but also with the psychological resources that are needed to handle life demands, manage stress, make wise life choices, remain healthy, and otherwise have a high quality of life. But surprisingly, relatively little is known about how best to promote resilience among young adults. Although the physical and psychological health consequences of the lack of resilience have been well documented, and universities make significant efforts to keep their students healthy and thriving, there is a dearth of knowledge regarding how to systematically create the conditions that foster resilience across the domains of students’ experiences at college. In response to these challenges, and in partnership with the Duke Endowment’s ongoing interest in the well-being of students at Duke, a three-year program of research draws on a range of concepts and methodologies across behavioral science to create and test a framework for building resilience within university students. The research is a substantial first step in identifying and developing resources that universities and colleges can use to enhance the developing resilience of their students. The ultimate goal of the research is to disseminate the knowledge, methods, and interventions that emerge to as broad a range of colleges and universities as possible, as well as to psychological scientists and practitioners more broadly.

The C-StARR contributes staff support and provides expertise in measurement selection and development, data collection and management, and statistical analyses.

Substance Abuse and Treatment Gaps in Asian, Pacific Islands, and Multiple-Race Individuals

Principal Investigator: Li-Tzy Wu
Funding Source: National Institute on Minority Health and Health Disparities (R01MD007658)

Asian Americans, Pacific Islanders, and multiple-race individuals are among the most understudied and the least known minority populations, especially for substance problems and treatment needs. The lack of population-based data on the patterns and magnitude of substance-related disorders and treatment needs make it almost impossible to develop evidence-based programs and policies to address substance-related burdens for these underserved populations. These groups are underserved, especially for behavioral health. Compared with whites, Asian Americans and Pacific Islanders as a group receive poor quality of health care and underutilize health care. Multiple factors may contribute to poor health or underutilization of health care among these groups, including poverty among various subgroups, poor English proficiency, absence of culturally competent interventions for substance problems, limited financial resources, immigration-related stress, and culture-related differences in perception of health care. Capitalizing on datasets from surveys of substance use disorders and the electronic medical record data, this project will examine substance abuse and treatment gaps for these groups, including onsets of substance use; the patterns and correlates of polysubstance use and substance use disorders; patterns and correlates of substance abuse treatment use; psychiatric comorbid disorders and their associated treatment use; and pathways of psychiatric treatment and clinical outcomes. The findings will identify at-risk subgroups to facilitate targeting primary prevention programs. They also will specify subgroups showing elevated odds of drug use disorders or other mental disorders, so they can be targeted for focused interventions to reduce serious consequences. The field will understand better the unmet needs for and barriers to treatment use among Asian Americans, Pacific Islanders, and multiple-race individuals.

The C-StARR contributes staff support and provides expertise in measurement selection and development, data collection and management, and statistical analyses.