Disentangling unique etiologies of prescription opioid misuse and heroin use: Analysis of longitudinal, twin, and genomic data

PROJECT SUMMARY/ABSTRACT Broad/Long Term Objectives: In line with NIDA?s positioning of the United States? opioid crisis as a priority area of research, the broad goal of this project is to elucidate the nature of the association between prescription opioid misuse (POM) and heroin use.

Integrating longitudinal and genetic methods, this study aims to identify the contribution (or lack thereof) of POM to entry into heroin use and to delineate potentially unique etiologies, courses, and correlates of POM and heroin use.

Specific Aims: The aims of the proposed project are to identify distinct opioid use trajectories and their correlates; estimate heritability of and genetic correlations between POM, heroin use, and other forms of drug use in genomic data; and utilize multivariate twin modeling to test whether POM is more etiologically similar to other prescription misuse behavior than to heroin use.

Research Design and Method: Longitudinal panel data from Monitoring the Future will be used to identify broad opioid use trajectories, trajectory subgroups, and movement between subgroups (e.g., POM to heroin use) over time via growth curve modeling, growth mixture modeling, and latent transition analysis.

Genomic data from the National Epidemiologic Survey on Alcohol and Related Conditions-III will be used estimate heritability of POM and heroin use, and genetic correlations between 1) POM and heroin use, 2) POM and other prescription misuse, and 3) heroin use and other illicit drug use via genome-wide complex trait analysis.

Australian Twin Registry Cohorts II and III will be used to test a multivariate twin model, which will incorporate several indicators of prescription misuse and illicit use to determine the extent to which specific and common genetic liability influence each phenotype.

Significance: This project will advance understanding of the unique etiological mechanisms underlying risk for POM and heroin use, thereby better isolating factors contributing to the United States? opioid crisis and informing multipronged public health approaches to mitigating the impact of opioids at the population level.

Training Plan and Environment: The training plan is designed to provide the applicant with a rich program of training in advanced longitudinal and statistical genetic modeling, and their application to addiction science.

Training will take place in the Department of Psychological Sciences at the University of Missouri, which has a highly reputed addiction training program funded by an NIAAA institutional training grant (T32; PI: Kenneth Sher); over 25% of the faculty in the department conduct substance use research.

Fellowship training faculty were drawn from both clinical (Dr. Wendy Slutske [sponsor], Dr. Ian Gizer [co-sponsor]) and quantitative psychology (Dr.

Phillip Wood [consultant]) training areas, and bring expertise in addiction research, longitudinal methods, and statistical genetics. External consultants (Drs. Arpana Agrawal, Elliot Nelson, and Rachel Winograd) will provide expertise on opioid use and use disorders.

All external consultants are faculty members of nearby Missouri institutions and are University of Missouri affiliates, supporting a synergistic training environment.