A novel application of Bayesian methods for modeling substance use trajectories


Principal Investigator: SIERRA BAINTER
Abstract: DESCRIPTION (provided by applicant): Substance use is one of the most commonly occurring health risk behaviors in adolescence and has been unambiguously linked to a variety of negative physical, biological, and psychological outcomes (e.g., Hingson &Kenkel, 2004;USDHHS, 2007). These serious public health issues have impelled substantial growth in the theoretical conceptualization of pathways to substance use, especially factors that exist in childhood and adolescence (e.g., Hussong et al., 2011;Zucker, Heitzeg, &Nigg, 2011). However, researchers currently disagree as to whether patterns of onset and escalation are best captured through the identification of discrete groups or "types" of individuals or through modeling individual variability across continuous trajectories of use. Growth mixture models are widely used to identify qualitatively different subgroups, yet key limitations of these models directly undermine the extent to which competing theories of substance use and abuse can be validly tested and compared. It has been extensively demonstrated that maximum likelihood (ML), the current gold-standard method for estimating growth mixture models, is unable to reliably reproduce the true population structure. Mixture models are highly sensitive to even slight model misspecifications or improper restrictions, and if data are non-normally distributed spurious classes will be detected (Bauer &Curran, 2003a;2004). If latent classes truly exist, it is difficult to correctly determine the number and form of latent trajectory classes using existing fit statistics (Tofighi &Enders, 2008). Compared to ML, there is the potential for more parameters and more complex models to be identified in a Bayesian analysis (Muthen &Asparouhov, in press), and Bayesian estimation is more stable and has more power in small samples (Asparouhov &Muthen, 2010). If a model with too many classes is estimated (Rousseau &Mengersen, 2011), Bayesian estimation will reliably "empty" the unneeded classes whereas ML estimation becomes unstable. In a fully Bayesian latent class analysis, the distribution of the number of classes can be estimated and examined as an unknown parameter (Richardson &Greene, 1997), whereas no information of this kind is available using ML estimation. However, Bayesian estimation still needs to be rigorously studied in the context of adolescent substance use. The three core aims of my project are to (1) compare Bayesian and ML estimation of growth mixture models when the population is truly homogenous;(2) compare Bayesian and ML estimation when multiple trajectory classes do truly exist;and (3) apply novel Bayesian methods to existing adolescent substance use data. My proposed project will fully integrate advanced quantitative methods with substantive theory so that researchers can reliably and validly test developmental theories of adolescent substance use.
Funding Period: 2013-08-01 - 2016-07-31
more information: NIH RePORT

Detail Information

Research Grants30

  1. Center for the Neurobiology of Addiction Treatment
    Steven R Childers; Fiscal Year: 2013
    ..These projects utilize several animal models and technologies uniquely developed by the Center investigators over the previous years of its funding history. ..
  2. Substance Abuse Research - Medications Development Center
    JOY MARIE SCHMITZ; Fiscal Year: 2013
    ..abstract_text> ..
  3. Implementation of Evidence-Based Preventive Parenting Programs
    Anne Marie Mauricio; Fiscal Year: 2013
    ..In Aim 3, we examine changes in fidelity, adaptation, and quality over time and the influence of provider characteristics on these changes. ..
  4. Effectiveness of Recovery High Schools as Continuing Care
    Ken C Winters; Fiscal Year: 2013
  5. CDART - Center for Drug Abuse Research Translation
    Michael T Bardo; Fiscal Year: 2013
    ..The long-range goal is to improve the design and implementation of targeted anti-drug preventive interventions. ..
  6. Integrative Risk Reduction and Treatment for Teen Substance Use Problems and PTSD
    Carla Kmett Danielson; Fiscal Year: 2013
    ..Demonstrating the efficacy of this promising risk reduction and treatment approach could provide a valuable clinical tool for community-based therapists. ..
  7. Behavioral Economic Analysis of Medical Marijuana Use in HIV+ Patients
    Mark K Greenwald; Fiscal Year: 2013
    ..These findings will be disseminated to patients, providers and society to promote public health and informed policy debate. ..
  8. Multi-Court Trial of NBP to Prevent Substance Abuse and Mental Health Disorder
    Irwin N Sandler; Fiscal Year: 2013
    ..5 million U.S. children whose parents divorce each year. ..
  9. Drug Abuse, Incarceration &Health Disparities in HIV/AIDS: A Longitudinal Study
    Linda A Teplin; Fiscal Year: 2013
    ..This study responds to the initiatives of NIDA, NIAAA, and other NIH institutes to reduce health disparities in HIV/AIDS in minority populations. ..
  10. Evaluating existing and innovative models for adolescent substance use data
    James McGinley; Fiscal Year: 2013
    ..Thus, the proposed project will fully integrate advanced quantitative methods with substantive theory so that researchers can reliable and validly test intricate developmental theories of adolescent substance use. ..
    KYLE MATTHEW KAMPMAN; Fiscal Year: 2013
    ..By starting with a large number of candidate medications and sequentially testing as described, we hope to more rapidly identify effective medications that justify the next level of development: Multi-site trials. ..
  12. Integrative Data Analysis of HIV Prevention Trials
    JENNIFER LYNN WALSH; Fiscal Year: 2013
  13. Monitoring the Future: Drug Use and Lifestyles of American Youth
    Lloyd D Johnston; Fiscal Year: 2013
    ..Results will continue to elucidate drug use from adolescence through middle adulthood-including the introduction of new drugs-with major implications for the policy, research, treatment, and prevention agendas. ..
  14. Monitoring the Future: A Cohort-Sequential Panel Study of Drug Use, Ages 19-55
    Lloyd D Johnston; Fiscal Year: 2013
    ..Results will continue to elucidate drug use from adolescence through middle adulthood-including the introduction of new drugs-with major implications for the policy, research, prevention, and treatment agendas. ..
  15. Risk Genes and Environment Interactions in NTDs
    MARGARET ELIZABETH ROSS; Fiscal Year: 2013
    ..Progress made for this disorder can provide useful analytical tools for identifying molecular network interactions relevant to later-onset complex genetic disorders, like schizophrenia and autism. ..
  16. First Longitudinal Study of Missed Treatment Opportunities Using DOD and VA Data
    Mary Jo Larson; Fiscal Year: 2013
    ..The findings will provide operationally actionable data useful to quality improvement programs in the MHS and VA on urgent issues requiring clinical and policy attention. ..