As we grapple with the COVID-19 pandemic, another persistent public health crisis—the opioid epidemic—continues to affect millions of people in the United States. Our difficulty in addressing the opioid epidemic stems from its complexity. Opioid use and abuse are influenced by factors at multiple levels within and outside of the health system such as a patient’s pain intensity and mental health, the clinician’s prescribing practices and workflow, and societal level factors such as access to care.
Combating a crisis that contributes to the death of an average of 136 people every day not only requires individual-level interventions, but also demands an informed approach driven by of-the-moment data and research. Data that deepens our understanding of the many factors that fuel the opioid epidemic are essential. These data must flow from all corners of the health system—and producing, collecting, and managing the data requires improvements in data infrastructure.
If you are following the discussions at this week’s Health Datapalooza, it is clear that the robustness of data infrastructure has a foundational impact on the ability of researchers to put data to use to inform policy and practice. To fuel patient-centered research on opioid use disorder that generates actionable clinical strategies, data infrastructure solutions are urgently needed.
Fortunately, several cross-agency projects are already underway in an effort to address this need. Under the Office of the Secretary Patient-Centered Outcomes Research Trust Fund (OS-PCORTF), the Assistant Secretary for Planning and Evaluation (ASPE) supports and coordinates a range of cross-agency projects aimed at building data infrastructure for patient-centered outcomes research (PCOR). Several of the projects are targeting the opioid crisis directly by improving the quality and timeliness of opioids-related outcomes data, increasing collection of patient-reported information on opioid use, and building linkages to address co-morbid mental health conditions that can correlate with opioid use and worsen patient outcomes.
Improving timeliness and detail of outcomes data
In order to have an impact on policy and practice, data must be both timely and of sufficient detail for researchers to answer pressing questions about the opioid epidemic as it shifts and changes, including the emergence of new drugs. Unfortunately, information on clinical outcomes such as opioid-related overdose and mortality may be slower to come and less detailed than researchers would need. Currently the US codes all death records in the United States to ICD-10 codes. While ICD-10 can be used to differentiate between classes of drugs (synthetic opioids), it does not allow for detailed analysis of specific drugs (e.g. fentanyl). This makes it difficult for researchers to track the specific drug(s) involved in overdose deaths.
A project led by the NCHS seeks to address this by strengthening mortality data infrastructure for outcomes research on deaths associated with opioid overdose. Specifically, the project is redesigning and enhancing the automated cause of death coding system to identify specific drugs involved in deaths and to expand the Vital Statistics Rapid Release (VSRR) to make a broader array of geographic and demographic data available. In addition, this project will also increase the timeliness of the data by allowing faster coding of drugs. This work will help researchers study population health trends and risk factors for opioid-related mortality, and it will improve the quality, availability, and timeliness of data available to better address the opioid crisis.
Building capacity for collection of patient-reported outcomes
In addition to clinical outcomes, the collection and use of patient-reported outcomes (PROs) holds particular promise for addressing the opioid epidemic. Research has identified several overdose risk behaviors that only the patient can report such as mood swings, tobacco use, illegal drug use, and frequency of taking medication other than the way it was prescribed. Access to these data can help researchers better understand risk behaviors as well as the effectiveness of substance use disorder treatment.
While tracking PROs is challenging in even a routine clinical visit, it can be even more so in an emergency department (ED) environment. This is a particular challenge for opioid use disorder (OUD) given the ED is an important point of entry into the health care system for many OUD patients. A project led by a team of researchers at Yale School of Medicine in collaboration with the National Institute on Drug Abuse (NIDA) is exploring the feasibility of collecting PROs related to opioid use after an ED visit through a patient-friendly app. The goal is to standardize OUD measurement and help track and ultimately improve the quality of care delivered to OUD patients.
Building linkages to address co-morbidities
Co-occurring disorders are common among OUD patients. Mental health issues are a particularly important aspect to consider. Data show that nearly 15 percent of adults with serious mental illness misused opioids in 2018. While systems to collect this information already exist, the data are often not linked at the individual or patient level. For example, the clinical electronic health records (EHR) data and claims data that are generated from an ED visit will be missing a key piece of outcomes data if the patient dies of an overdose outside the hospital at a later date. Likewise, outpatient EHR data may document other mental health or substance use issues, but is not linked to ED data.
A project led by the CDC seeks to address this by linking hospital care and mortality data to identify co-occurring mental health disorders among opioid users. This will allow researchers to answer questions such as which opioid-using patients have a co-occurring mental health disorder and how do they differ from other opioid users without mental health illnesses? The linked data can also be used to help assess risk levels for repeat hospital visits among patients with co-occurring mental health illness. This project builds on previous OS-PCORTF-funded work aimed at other advancements in finding patterns in opioid use and treatment response among patients’ hospital encounters. Find more details about this work in this project spotlight.
Looking to the future
The opioid crisis is complex and making progress to alleviate it requires not only better data, but improvements across nearly every aspect associated with how the data is collected and managed – things like improved standardization of key metrics and indicators, more robust linkages across a range of data sources, and a continuous coordinated effort within and across federal and state institutions. Through the OS-PCORTF portfolio, HHS agencies have collaborated closely to improve the quality and timeliness of outcomes data, increase collection of patient-reported information, and build linkages to address co-morbid conditions that affect patient outcomes. This work is paving the way for patient-centered outcomes research to help address the opioid epidemic through improved data infrastructure. It’s simultaneously addressing both the vast and the minute aspects of the crisis. And that’s exactly what we need.
Learn more about other OS-PCORTF projects working to address the opioid crisis in a vignette on the topic here.