NRSA Fellowship for Emily Hankosky: Big Data to Treatment: Repurposing pharmacotherapies for psychostimulant use disorder

Grants and Contracts Details


Misuse of psychostimulants, such as cocaine and amphetamines, is a serious public health problem, contributing to one of the leading causes of preventable death and disability in the United States. As of 2013, there were an estimated 1.3 million Americans meeting diagnostic criteria for a psychostimulant use disorder (PSUD; SAMHSA, 2014). Currently, however, there are no FDA-approved pharmacotherapies to treat PSUDs and national estimates indicate that less than 20% of individuals with a PSUD receive treatment (TEDS, 2015; SAMSHA, 2014). As a result, treatments for this disorder are limited to current psychosocial interventions, which are only modestly effective (Minozzi et al., 2016). The goal of this proposal is to address this unmet clinical need by identifying pharmacotherapies that can be repurposed for PSUDs. Drug repurposing is a timeand cost-effective approach of identifying new therapeutic indications for existing pharmacotherapies (Ashburn and Thor, 2004). Monitoring health outcomes is one approach to determine novel indications for existing therapeutics, as was the case when the antidepressant Wellbutrin was identified as a smoking-cessation agent. Recently, this approach has been successfully applied to health claims data to identify a potential new pharmacotherapy for age-related macular degeneration (Brilliant et al., 2016). Using our local state and nationwide health claims databases, we will evaluate the potential of FDA-approved pharmacotherapies to be repurposed for PSUDs. Specifically, we will apply big data analytics to administrative health claims data to elucidate associations between prescribed pharmacotherapies and health outcomes in individuals with PSUDs. We will identify pharmacotherapies with potential to treat PSUD through two aims using hypothesis-driven and data-driven approaches. Our hypothesis-driven approach will be informed by current theories on the underlying causes of PSUDs that implicate self-medication of comorbid disorders, drug-induced cognitive deficits, and withdrawal as factors contributing to PSUDs (Quello et al., 2005; Negus and Henningfield, 2014; Sofuoglu et al., 2016). Using the statewide Kentucky Medicaid and national Truven Marketscan databases, we will accomplish the following Specific Aims: Aim 1: Discover pharmacotherapies that are associated with improved health outcomes in persons with a PSUD. Using a hypothesis-driven approach, we identified three medications with putative therapeutic potential based on current knowledge of PSUDs and we will evaluate the association between these pharmacotherapies and health outcomes in persons with a PSUD. Consistent with current theories of factors that contribute to PSUDs, we will test specific hypotheses that pharmacotherapies intended to treat comorbid psychiatric disorders, enhance cognition, or act as agonist replacements are associated with increased incidence of remission. Specifically, we will evaluate time to remission diagnosis in enrollees with a PSUD after matching for age, sex, comorbid psychiatric disorders, polysubstance use, cognitive/behavioral interventions, and duration of PSUD. In addition to remission diagnoses, we will model healthcare utilization, healthcare expenditures, emergency department visits, and mortality as outcomes in terms of our predictor variables. Aim 2: Generate testable hypotheses regarding pharmacotherapies to be repurposed using data mining. Using a data-driven approach, we will identify relationships between pharmacotherapies and health outcomes with data-mining algorithms. Using a combination of supervised and unsupervised data-mining algorithms, we will discover patterns and associations between health outcomes in individuals with PSUD and prescribed medications to facilitate knowledge discovery of pharmacotherapies that may be repurposed for PSUDs. Impact. Achievement of the proposed Aims will reduce the burden of PSUDs through identification of pharmacotherapeutic alternatives and/or adjuncts to psychosocial interventions, thereby improving clinical practice with the potential to transform the treatment of PSUDs. The discovery of novel pharmacotherapies will advance our understanding of mechanisms underlying PSUDs and increase the diversity of biological targets pursued in drug discovery efforts. Furthermore, success of this proposal could augment the pipeline of drug discovery and development to include early identification of biological targets through the application of big data analytics to administrative health claims data. Importantly, consistent with the mission of NIH’s Big Data to Knowledge initiative, completion of these Aims will provide Dr. Hankosky with the multidisciplinary training experiences necessary to develop an independent research program combining basic science with big data informatics to advance the development of pharmacotherapies for PSUDs.
Effective start/end date8/1/187/7/19


  • National Institute on Drug Abuse: $65,136.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.