Funded Projects

Monitoring and reporting substance use and substance use disorder is difficult to obtain in real-time using conventional methods. However, social media captures large amounts of data about substance use that are reported by diverse groups of people. Analysis of these data can provide population- or subpopulation-level insights, at low cost and in near real-time. This project aims to convert large amounts of social media data on substance use into actionable knowledge using advanced natural language processing and artificial intelligence approaches. The researchers will publicly release the aggregated statistics through a dedicated dashboard and provide user-friendly, open-source tools to determine trends and analyze disparities.

Addressing the current opioid overdose crisis requires tracking risky opioid use in a timely manner so that public health agencies can plan accordingly and supply life-saving resources. American Indian Tribes often lack such tools, even though American Indians and Alaska Natives have the highest rates of opioid overdose fatalities. This project will adapt commercialized monitoring technologies for use in Tribal communities, in consultation with affected Tribes. Through a partnership with a Tribal Fire Department and a software company providing data analytics for public safety agencies, this research will build a near real-time opioid overdose dashboard for use within Tribal boundaries. The findings may also improve data collection and outbreak monitoring for other substances, including methamphetamine and cocaine.

To respond quickly and effectively to the constantly changing dynamics of the opioid crisis, public health agencies and organizations need timely state and local data to make critical decisions about where to allocate resources and target responses. This project creates the Rapid Actionable Data for Opioid Response in Kentucky system, a near real-time statewide surveillance system. This resource will combine data from multiple state agencies to provide actionable and timely information to support opioid overdose prevention, harm reduction, evidence-based treatment, and recovery services. The project will also develop user-driven reporting and visualization tools (mobile and web-based apps) that provide immediate access to near real-time community or state level data, reports, and visual analytics.

This project creates the HEAL Data2Action Modeling and Economic Resource Center that will conduct research as well as support the HEAL Data2Action Innovation Projects with expertise and consultation about simulation modeling and economic evaluation methods. The consultation service will advise on how to select and use various research methods, including economic evaluation, simulation modeling, advanced statistical analysis, behavioral economics, treatment program organization research, and cost analysis. The center will use a dynamic simulation model of opioid use disorder to enhance data-driven decision making. The center will also provide online training resources, tools, and other resources to assess a variety of economic aspects related to the HEAL Data2Action Innovation Projects.

Effective overdose prevention requires timely, location-specific data to efficiently direct resources and interventions as well as to inform healthcare policy. However, rarely is such information available, resulting in responses that are frequently too late. This project will partner with local government agencies from Los Angeles County, California, to rapidly acquire and analyze information in near real-time from multiple sources to identify overdose hotspots and determine overdose metrics. This information will be used to develop a publicly available Rapid Overdose Surveillance Los Angeles online dashboard with built-in analytics. The dashboard will help determine the scope and specifics of overdose and opioid use in Los Angeles County to help target response and guide mobile addiction treatment and harm reduction services.

This project will use a data-to-action framework to guide implementation of opioid use disorder treatment and harm reduction interventions based on rapid data analysis. It will leverage existing data systems such as Emergency Medical Services (EMS) reports, mobile integrated health case management data, and medical examiner data for near real-time data analyses, visualization, and action planning. This research will collect a range of data (opioid treatment, use of acute care services, morbidity, mortality, incarceration, housing support, and cost benefits) from a sub-acute stabilization center for people at high risk for opioid overdose, including those who have recently overdosed and are referred and transported by EMS teams.

This project creates the HEAL Data2Action Data Infrastructure Support Center that will conduct research as well as support the HEAL Data2Action Innovation Projects with analytics and support for navigating electronic health records and other “big data” sources, as well as logistical processes related to privacy and ethics. The center will help plan for data collection and assessing outcomes, as well as provide data visualization efforts for various stakeholders. Key support mechanisms include courses, seminars, conferences, trainings, and on-demand technical assistance.

Guidelines published by the Centers for Disease Control and Prevention in 2016 recommended gradual reductions in opioid medication doses (opioid tapering) for people with chronic pain and substance use disorder and recommended that those patients with pain and opioid use disorder should be switched to opioid use disorder medications. Despite wide implementation, little is known about the consequences of opioid tapering among patients with co-occurring chronic pain and substance use disorder. This project will use various databases (Cerner Real-World DataTM, American Hospital Association data, and U.S. Census data) to create a representative electronic health records database. This database will be used to determine the relationship between opioid tapering, multidisciplinary pain treatment, and medications for opioid use disorder – as well as monitor outcomes for patients with chronic pain and co-occurring substance use disorder.

Chronic pain affects millions of Americans and remains poorly understood and challenging to manage. Researchers do not fully understand brain processes involved in chronic pain, which can vary considerably from person to person. This project will analyze brain function using magnetic resonance imaging (MRI) in individuals with and without chronic pain. The research will also directly determine the degree of pain-related brain imaging changes by using a large database of brain imaging data.

It is unclear if long-term use of opioids by head and neck cancer patients affects risk for depression, which is higher in this population compared to people without cancer. This knowledge could inform interventions such as increased opioid prescription safety or alternative pain management approaches and could thus help reduce the risk for depression-related outcomes. This project will use data from a national cancer database linked to Medicare claims and a Veterans Administration database to determine whether people with head and neck cancer that take opioid medications for more than 90 days have increased risk for new-onset or worsening depression or suicide death.

Fifty million Americans experience chronic pain, including about 25 million who report pain that substantially interferes with daily activities and reduces quality of life. Mental health problems, including depression, anxiety, and post-traumatic stress disorder, increase risk for severe chronic pain. This project will use genetic data, information about observable characteristics (phenotypic data), and neuroimaging data from three large databases to identify psychosocial factors that predict chronic pain, assess differences across diverse U.S. populations, and determine whether risk profiles predict post-surgical chronic pain.

Sickle cell disease is an inherited blood disorder affecting about 100,000 Americans and more than 20 million people worldwide. It is caused by a mutation in the gene for beta-globin that results in the characteristic sickled shape of red blood cells, life-long severe pain, and shortened lifespan. Sickle cell disease pain episodes are usually unanticipated, making it hard for people with the condition to manage their pain and putting them at risk for increased use of opioids and poor health outcomes. This project will use existing real-time health data to identify factors that can predict onset, severity, and worsening of daily pain in children with sickle cell disease.

Black/African American and Hispanic/Latino individuals suffer a disproportionate burden of co-occurring substance use and mental illness, in part due to reduced access to culturally responsive quality healthcare, compared to other racial/ethnic groups. In addition, Black/African American and Hispanic/Latino individuals are not well represented in clinical trials that could help reduce these health disparities. This research aims to improve the recruitment of Black/African American and Hispanic/Latino individuals to participate in clinical research related to co-occurring substance use and mental illness. The project will conduct community engagement and community-based participatory research, establishing a bidirectional partnership between researchers and community members.