Patrick J. Curran - US grants

The proposed research project is a first submission of an R01 application by a young investigator. The project draws directly upon developmental theories of substance use and conduct problems to identify important research questions that are currently difficult to evaluate empirically and would greatly benefit from the development of new quantitative methodologies. The proposal is organized around three major goals. First, a new class of innovative longitudinal models of individual differences in change over time will be developed with the explicit purpose of allowing powerful, flexible and dynamic empirical tests of developmental trajectories of substance use. These new models will allow for the estimation of a variety of multivariate latent curve models that are more robust to non-normality and model misspecification, that incorporate two- and three-way interactions among latent growth factors in the prediction of later problem behaviors, and that simultaneously examine characteristics of multiple developmental trajectories before and after important life transitions. Second, the validity and utility of these new quantitative methods will be closely evaluated with particular emphasis on future applications in studies of substance use over time. Evaluation of these new models will primarily be accomplished through the use of comprehensive computer simulations that will focus on finite sampling characteristics that are commonly encountered in applied longitudinal research of substance use and abuse including small sample size, multivariate non-normality, and model misspecification. Third, these new models will be applied to a high quality existing longitudinal data set to examine the course, causes and consequences of co-occurring developmental trajectories of conduct problems and substance use in a large sample of adolescents. Data will be drawn from an existing longitudinal study of over 1000 adolescents ranging in age from 9 to 18 assessed once a year for four years. Application of the newly developed quantitative, methods to this data will allow for a detailed evaluation of a number of theoretically derived research hypotheses including questions about reciprocal developmental relations between substance use and conduct problems over time, interactions between trajectories of conduct problems and substance use in the prediction of later problem behavior, and the effects of school and pubertal transitions on these developmental trajectories. The empirical applications will also serve to demonstrate and disseminate the newly developed models to a wide audience of applied substance use researchers. The unifying goal of the proposed project is to focus extended effort on explicitly linking the fields of quantitative methodology and applied developmental psychopathology to create and disseminate a new class of innovative statistical models that are optimally suited for studying individual differences in developmental trajectories of substance use over time.

DESCRIPTION (provided by applicant): In our first project period we focused on the systematic study of longitudinal latent curve models (LCMs) applied to continuous scale scores with a particular emphasis on challenges that commonly arise in the empirical study of substance use. Despite the many advantages of these LCMs, one limitation is that methods are currently not well developed to fit LCMs to repeated measures that are ordinally scaled, particularly in the presence of missing data. A second limitation stems from the potential violation of a strict set of required assumptions governing the structure of the measurement model of continuous or ordinally observed scale scores over time. To address these limitations, we have drawn upon our findings from the initial project period and have designed the revision of our proposed continuation project around the systematic study of measurement models in latent curve analysis. Our proposed project is organized around four specific aims. In Aim 1 we propose to study existing challenges and identify optimal strategies for fitting LCMs to ordinal manifest scale scores assessed overtime both with complete and missing data. In Aim 2 we plan to study the incorporation of latent factors with continuously scaled indicators in LCMs to allow for tests of measurement invariance and the inclusion of formal measurement models. In Aim 3 we propose extending the findings of Aim 2 to include the incorporation of latent factors with ordinally scaled indicators in LCMs. Finally, in Aim 4 we plan to study the implications of item scaling and measurement invariance across all prior aims with respect to the estimation of statistical power and optimal study design. These project goals will be pursued through the integrated use of analytical review and organization, computer simulation studies, and the analysis of data drawn from an existing longitudinal study of the parental alcoholism effects on the development of drug use in a large sample of adolescent. Taken together, we believe the proposed study has the potential for making significant unique contributions to the field of quantitative methodology and to the rigorous empirical study of developmental trajectories of substance use and abuse.

DESCRIPTION (provided by applicant): In this application, we evaluate an internalizing pathway to substance use disorders (SUDs) and consider the etiology of Negative Affect SUDs, a potential phenotype of SUDs identifiable by comorbidity with disorders involving negative affect (i.e., depression and anxiety). We conduct an integrative (secondary) analysis (refined in our prior project period) involving the simultaneous analysis of three nationally prominent longitudinal studies of children of alcoholic parents and matched controls that collectively span the first four decades of life (ages 2 through 38). Understanding developmental pathways leading to SUDs is critical in efforts to design and implement effective intervention and treatment programs for youth. Although behavioral indicators of risk for adolescent substance involvement appear as early as 2-5 years of age, few theories about developmental pathways to SUDs consider risk processes that begin in the preschool years. Moreover, few studies are capable of evaluating theories about these developmental pathways. Longitudinal studies that span the first four decades of life, when SUDs emerge, peak and begin to decline, are rare. Available cohort- or population-based studies often yield too few cases to disentangle different pathways for SUDs. Thus, long-term, longitudinal studies of high-risk populations, such as children of alcoholic parents, are invaluable for articulating developmental pathways to SUDs. Using this method, we pursue five specific aims: (1) to define and test an internalizing pathway to SUDs and to evaluate whether evidence supports heterotypic continuity (i.e., different developmental expressions of a single underlying trait) of Negative Affect SUDs over time, (2) to examine the developmentally-varying unique, mediated and interactive effects of internalizing and externalizing processes as predictors of substance involvement and Negative Affect SUDs across the first four decades of life, (3) to examine gender, parental depressive alcoholism, and contextual factors as important moderators of progression along an internalizing pathway toward Negative Affect SUDs as a function of developmental timing, (4) to examine whether the predictive utility of an internalizing model differs over various substances used in isolation (i.e., alcohol, tobacco, marijuana, stimulant and depressant drug) or in combination with one another, and (5) to develop methods for integrative analysis, to utilize these methods to test Aims 1-4, and to disseminate these methods to other applied researchers. PUBLIC HEALTH RELEVANCE: Our application tests an internalizing pathway to substance use disorder over the first four decades of life. Studies of such early emerging but persistent pathways are rare but critical to efforts to design and implement effective intervention and treatment programs for youth.

DESCRIPTION (provided by applicant): Increasing the connection between treatment research and treatment as it is practiced in community settings (i.e., ecological validity) is essential for translating recent empirically supported approaches to individuals who would benefit most from them. A recent series of federally supported initiatives has focused on enhancing ecological validity by (a) translating findings from federally funded efficacy trials to treatment effectiveness work in real-world settings (particularly with the genesis of the National Drug Abuse Clinical Trials Network) and (b) advancing quantitative methods specifically geared toward addressing the complexities that arise in (and have stifled) treatment research in real-world contexts. The investigation proposed in this R01 application combines work from an ongoing Clinical Trials Network (CTN) effectiveness trial on the most widely used treatment for comorbid substance abuse and PTSD available for women, Seeking Safety, with continuing development, refinement and extension of statistical approaches to handle data from therapy groups when therapy group membership itself changes over time by design. Although progress has been made in developing methods to handle longitudinal data from open-enrollment (also known as "rolling") therapy groups, the most common milieu for treatment delivery in community-based substance abuse programs, several salient critical issues to the stakeholders in substance abuse treatment research remain unresolved and are of primary methodological interest in this investigation. This includes modeling the impact of group membership turnover on (a) the long-term recovery management process, (b) the identification of the active components in ecologically valid treatment trials, and (c) (sub) optimal patient mix in treatment groups and measurement of treatment group (in) stability. The ultimate goal of the proposed secondary data analytic investigation is to place more powerful analytic and methodological tools in the hands of substance abuse and alcoholism treatment researchers for (a) analyzing existing data from open- enrollment groups in a more defensible manner, maximizing the likelihood that incorrect inferences are avoided (i.e., advocating for a treatment that may be ineffective in reality) and that the full complexities of open-enrollment group trials are captured analytically, and/or (b) submitting new applications for trials in all areas of behavioral treatment research with protocols that more closely resemble treatment in community settings. In addition to the practical benefits of the methodological development in this project (e.g., more accurate inferences concerning treatment efficacy, better reflection of the open-enrollment process in data analytic models), this application represents an opportunity for further dissemination, and a more nuanced examination of, what is to date, the largest randomized clinical trial for Seeking Safety. The proposed project seeks to develop more accurate methods to analyze data from open-enrollment substance abuse treatment trials, which present special analytic challenges (i.e., changes in group membership over time). If we are successful in continuing the development and refinement of new modeling tools to address analytic complexities in open-enrollment trials, substance abuse treatment researchers will be positioned to develop federally funded open-enrollment treatment trials that better resemble treatment as it is practiced in community settings. These developments may also facilitate building a "community-friendly" treatment research portfolio for funding agencies that supports substance abuse and alcoholism treatment research.