Full Summary — Using Science to Inform Practice and Policy: A Coordinated Approach to Research Priority Setting

Overview

December 11, 2017

The National Institute on Drug Abuse (NIDA), in partnership with ECRI Institute, convened the meeting as part of the Cutting-Edge Science Meeting Series to End the Opioid Crisis. Participants included leaders and subject matter experts from agencies and offices within the U.S. Department of Health and Human Services (HHS), representatives from the Department of Veterans Affairs and Department of Defense, and national organizations.

The goals of the meeting were to (1) inform the NIDA’s strategic thinking and planning, as well as that of other participating organizations, (2) identify and foster potential collaborations among the public- and private-sector participants, and (3) continue to build the inventory of top priorities that will benefit from an enhanced science base.

The following high-priority action items identified by participants include:

Strengthen the connection between research and practice:
- Note that a learning health care system can shorten the time it takes to generate evidence and translate evidence-based practices into the real world
- Adapt interventions to local needs and context
- Recognize that research should inform clinical practice
Launch a pilot demonstration project to improve treatment for OUD within a state by focusing on a state’s system of OUD care
Explore, maximize, and expedite the use of all available data sources and study designs (develop a complete inventory of relevant research results, including those from outside HHS):
- Use clinical trials, as well as observational and secondary data sources, to develop evidence
- Consider barriers to using the data, including confidentiality issues and the need for a feedback loop to primary care providers
Engage “citizen scientists” to help develop outcome and quality measures:
- Need standardized metrics that patients and families can use to report on patient experience and treatment outcomes
Determine criteria to indicate the most appropriate treatment setting, e.g., inpatient versus outpatient treatment:
- Collect longer-term outcomes data to understand the utility of different treatments, durations, settings, and modalities
Establish which treatment interventions are effective for patients with mild OUD or subclinical opioid misuse
Conduct research to address cost and sustainability and to develop effective public messaging

Background

Welcome and Introductory Remarks

NIDA Director Nora Volkow, M.D., welcomed participants to the meeting and acknowledged Francis S. Collins, M.D., Ph.D., Director of the National Institutes of Health (NIH), for his support toward this effort along with Jeffrey Lerner, Ph.D., President and Chief Executive Officer of ECRI Institute, and Carlos Blanco, M.D., for their work in planning the meeting.

Introductory remarks by Dr. Collins, Acting HHS Secretary Eric Hargan, and Richard Baum, Acting Director of the Office of National Drug Control Policy, highlighted the severity of the opioid overdose (OD) epidemic—with more than 42,000 opioid-related deaths in 2016.

Meeting Goals

The desired outcome for the meeting was a list of research gaps and potential solutions that organizations at the table can implement to end the opioid OD epidemic and improve treatment and prevention of opioid misuse and opioid use disorder (OUD). A longer-term goal is to create a sustainable research–practice feedback loop that involves public and private organizations, i.e., a learning health care system.

To advance these goals, a selection of participating organizations was asked to submit three 1-page summaries that identified their top priority issues that currently do not have adequate data to guide evidence-based practice. Twelve of these papers were grouped into four topic areas presented in panels at the meeting:

Panel 1—Comparative Effectiveness/Implementation of Medication-Assisted Treatment (MAT) for Opioid Use Disorders (OUDs)
Panel 2—Monitoring Quality of Care and Outcomes/Risk Stratification and Screening
Panel 3—Overdose Prevention and Reversal
Panel 4—Treatment of Pain

Panel 1: Comparative Effectiveness/Implementation of Medication-Assisted Treatment (MAT) for Opioid Use Disorders (OUDs)

Summary

Panel 1 presenters focused on the effectiveness of MAT for OUD and potential areas for improvement, specifically outlining current knowledge and areas for future research. They covered variation in treatment outcomes, the comparative effectiveness of different types of MAT, the use of this approach in primary care, and telehealth as a delivery mechanism for MAT. Participants highlighted the need for research on inpatient versus outpatient treatment, optimal treatment duration, research on care delivery models for special populations, and ways to scale up primary care-based best practices for MAT delivery. Open discussion touched upon research gaps; the need to address policy and regulatory barriers that affect MAT access, including the use of telehealth for MAT; data challenges (e.g., privacy regulations); and the need for information on evidence-based treatment among patients and families.

Suggested Steps for Advancing the Research/Practice Agenda

Conduct a large-scale intervention trial for improving addiction treatment that involves coordinated care across emergency departments (EDs) and primary care settings
Study the costs (collect information on relative cost differentials for implementation) of MAT to systems and patients
Take advantage of knowledge gained by observational studies and natural experiments occurring across the country, such as Medicaid Section 1115 demonstration waivers, and develop an inventory of studies in the field along with the funding investments
- Monitoring outcomes of SAMHSA STR grants are one such key opportunity to assess the details of novel service provision approaches and associated outcomes
Involve regulators in efforts to expand MAT access, build on current efforts (including telehealth), and address MAT provision capacity issues (e.g., address the root causes of why providers with waivers are not offering treatment at full capacity); rigorously evaluate telehealth implementation of treatment programs
Obtain and analyze data from health plans in addition to Medicare and Medicaid
Develop real-time metrics to monitor interventions at multiple levels (e.g., state, pharmacy, and criminal justice) and the point of care
Consider developing a research-based public health campaign to disseminate messages that will change the behavior of patients, families, community members, and practitioners toward the medical model of OUD
Determine effective models of OUD treatment for special populations (e.g., injection drug users, pregnant women, people with chronic diseases/ comorbidities, and adolescents)
Utilize implementation science to study the best approaches to scaling up MAT in primary care settings
Facilitate linkages to treatment and community-based services for high-risk, justice-involved populations

Key Issues and Research Strategies

Uptake of MAT: Study (1) cost-effective implementation methods, (2) optimal induction strategies, and (3) strategies to improve adherence

Varying treatment costs: Conduct comparison studies of residential treatment versus intensive outpatient programs

Long-term outcomes: Conduct prospective studies with sufficient follow-up times

Treatment access: Research improved telehealth models

Workforce development: Investigate evidence-based technical assistance strategies

Different settings and populations: Conduct culturally and geographically relevant studies

Panel 2 Discussion: Monitoring Quality of Care and Outcomes/Risk Stratification and Screening

Summary

Panel 2 presenters focused on how the field can improve the early identification of OUD, which is essential for earlier referral and patient engagement in treatment. Clinicians need better screening tools if they are to effectively detect OUD, as this will result in more expeditiously linking individuals to treatment, enhancing patient engagement, and improving identification of risk factors. There is a need for research on genetic and behavioral biomarkers that can be used to target analgesic approaches for pain management and on making information on associations with behavioral factors into actionable clinical guidance. Although evidence supports the effectiveness of current interventions for moderate to severe OUD, research is needed on interventions for screen-detected hazardous use.

Suggested Steps for Advancing the Research/Practice Agenda

Expand prevention of and screening for adverse childhood experiences, as these are risk factors for later OUD
Implement stepped care (akin to treating prediabetes) in early intervention for opioid misuse
Support research demonstrating the utility of screening tests in capturing hazardous use
Conduct studies on the best way to treat individuals with less severe opioid misuse and the Intervention and Referral to Treatment components of Screening, Brief Intervention, and Referral to Treatment (SBIRT) so that clinicians know what they should do in case of a positive screen result
Address issues around reimbursement for SBIRT
Coordinate different data sets that could contribute to research on long-term outcomes of screening, intervention, and treatment
Focus on quality improvement—i.e., meeting quality metrics and improvements connected with payment—which is a top-of-mind issue for clinicians and represents an immediate opportunity
Support “citizen science” (e.g., reviews of treatment programs) to develop better ways for patients and families to find high-quality treatment
Coordinate and improve HHS resources to help locate evidence-based treatment programs
Address ways for clinicians to identify, treat, and be adequately reimbursed for providing care for polysubstance use, which is the norm among individuals with OUD
Support qualitative research on relapse and its social determinants

Key Issues and Research Strategies

Quality care: Validate outcome metrics, and engage patients in reporting

Early identification: Develop enhanced screening/assessment tools

Biomarkers of risk and resilience: Conduct large well-characterized cohort studies

Comorbid conditions: Determine the impact of comorbid factors on pain and OUD as well as mental illness, HIV/HCV, pain, and post-traumatic stress disorder

Early OUD: Develop effective treatments for less severe OUD

SBIRT: Consider that there is evidence for screening but not for the effectiveness of interventions

Access to data: Determine ways to address the challenges from 42 C.F.R. Part 2 and other privacy restrictions, in addition to a lack of standardized reporting

Panel 3 Discussion: Overdose Prevention and Reversal

Summary

Panel 3 presenters focused on prevention of opioid ODs, treatment of OD with carfentanil and other synthetic opioids, and initiation of OUD treatment in the ED after OD and naloxone reversal. Research is needed on approaches to reduce opioid use in veterans who already are taking opioids, while minimizing their risks for pain, suicide, or conversion to street drugs. Approaches to tapering opioids and the potential adverse consequences should be studied in this population. A randomized trial should determine whether higher doses or a different dosing regimen of naloxone is needed when carfentanil and/or other substances are involved in OD. Preliminary research suggests that clinicians can successfully implement buprenorphine induction in people who experienced OD reversal in the ED, but larger studies are needed. Data from the Substance Abuse and Mental Health Services Administration’s (SAMHSA’s) State Targeted Response to the Opioid Crisis grantees may provide information on this important research gap.

Suggested Steps for Advancing the Research/Practice Agenda

Collect data on the proportion of fatal ODs that are preceded by ED visits for nonfatal ODs
Identify promising innovations in opioid OD reversal involving emergency medical services (EMS) from SAMHSA’s first responder grants
Use the Centers for Medicare and Medicaid Services’ (CMS’s) mechanisms (e.g., public health advisories) to determine whether multiple administrations of naloxone are needed during EMS visits
Collect real-time data on the location of ODs, perhaps using cellular telephone applications and identifying how survivors are linked to OUD treatment
Examine state data on nonfatal ODs based on ED information
Increase public awareness and pressure to improve ED protocols for nonfatal OD patients (e.g., patients should not be leaving the ED without a treatment plan)
Educate hospital attorneys on the Health Insurance Portability and Accountability Act (HIPAA) rules regarding communicating with family members about OD reversals
Partner with state Medicaid programs to address the opioid crisis and conduct research
Conduct research on ED buprenorphine implementation and its qualitative components
Engage primary care providers in training to provide MAT for OD survivors referred from EDs
Facilitate the development of clinical guidance on linking ED patients who have experienced an OD with OUD treatment (use lessons learned from the American College of Emergency Physicians’ pilot in Colorado)
Remove reimbursement barriers/establish reimbursement incentives to promote linkage to care following ED treatment of OD
Expedite surveillance reporting of opioid overdose deaths

Key Issues and Research Strategies

Synthetics and drug combinations: Determine optimal naloxone dosing and delivery; Identify alternative medications/relevant drug mixtures; Disseminate devices (autoinjectors) for OD when patients are alone

OD characterization: Study patient characteristics, coprescribing, and suicides

Prevention of OD: Improve MAT retention, nonpharmacological pain treatment, and comorbid treatment

ED aftercare: Improve identification and reporting of ODs in the ED; Improve ED aftercare models

Post-OD populations: Develop protocols for post-OD treatment populations

Panel 4 Discussion: Treatment of Pain

Summary

Panel 4 presenters focused on evidence-based strategies for opioid prescribing and pain management, treatment of pain in current clinical practice, and management of chronic pain in special populations. The CDC Guideline for Prescribing Opioids for Chronic Pain (2016) offers primary care providers 12 recommendations to help them determine when to initiate or continue opioids in adult patients who do not have active cancer or require palliative care. Research is needed to identify effective approaches to supporting the provision of guideline-concordant care and to assessing the impact of the guidelines. The field needs a stronger evidence base on the long-term effectiveness of nonpharmacological pain management options (e.g., acupuncture, cognitive behavioral therapy, and pain management coaching), as this knowledge could help stem the overprescribing of opioids for patients with chronic pain. It is crucial to identify best practices in chronic pain management for special populations to ensure safe and effective pain treatment. Additional research on successful chronic pain management could determine which therapies work best and reduce the risk of OUD.

Suggested Steps for Advancing the Research/Practice Agenda

Conduct research on pain management for pregnant women, children, and adolescents
Develop sufficient evidence that may enable the CMS to consider coverage of services (particularly alternatives to opioids and multimodal strategies)
Consider the large studies on nonpharmacological pain treatments and other relevant research sponsored by the Patient-Centered Outcomes Research Institute and the outcomes of expanded access to alternative pain treatments and opioid restriction in the Oregon Medicaid program
Include all socioeconomic classes (e.g., not everyone has coverage for physical therapy, paid time off, child care, and transportation) and the social determinants of health in research on nonpharmacological pain treatments and primary care models that address such barriers (e.g., colocating physical therapy and integrating behavioral health and primary care at health centers)
Consider applying effective strategies that have changed prescribing practices in other areas (e.g., antibiotics) to opioid prescribing
Improve accessibility of safe disposal for unused medications
Determine strategies to “de-implement” or “de-intensify” practices related to existing interventions that are inappropriate, ineffective, or potentially harmful
Collect detailed information on the prevalence, sources, and definitions of different types of pain in learning health care systems
Study the effectiveness of opioid prescribing management approaches that mitigate risk (risk assessment tools, opioid prescribing agreements and registries, urine drug monitoring, and use of prescription drug monitoring programs)
Utilize implementation science to study the best approaches to scaling up effective and safe chronic pain management services in primary care and other settings
Research best practices for tapering opioids for people who have been treated with opioids long-term for chronic pain
Study how quality metrics and patient satisfaction data can serve as disincentives to providers’ adherence to safe opioid prescribing guidelines and prescription of alternative pain management approaches

Key Issues and Research Strategies

Health outcomes: Develop strategies to enhance pain guideline-based care

Nonopioid treatment: Use performance metrics to assess treatment adherence and outcomes

Special population needs: Develop population-specific pain treatment; Measure the outcomes of pain for patients tapered off opioids

Next Steps

Dr. Volkow summarized next steps based on the paper submissions and the meeting discussion. She highlighted urgent issues to advance. High-priority areas for action include:

Strengthen the connection between research and practice:
- Note that a learning health care system can shorten the time it takes to generate evidence and translate evidence-based practices into the real world
- Adapt interventions to local needs and context
- Recognize that research should inform clinical practice
Launch a pilot demonstration project to improve treatment for OUD within a state by focusing on a state’s system of OUD care
Explore, maximize, and expedite the use of all available data sources and study designs (develop a complete inventory of relevant research results, including those from outside HHS):
- Use clinical trials, as well as observational and secondary data sources, to develop evidence
- Consider barriers to using the data, including confidentiality issues and the need for a feedback loop to primary care providers
Engage “citizen scientists” to help develop outcome and quality measures:
Need standardized metrics that patients and families can use to report on patient experience and treatment outcomes
Determine criteria to indicate the most appropriate treatment setting, e.g., inpatient versus outpatient treatment:
Collect longer-term outcomes data to understand the utility of different treatments, durations, settings, and modalities
Establish which treatment interventions are effective for patients with mild OUD or subclinical opioid misuse
Conduct research to address cost and sustainability and to develop effective public messaging

Implementation Demonstration Pilot

The group discussed the possibility of launching a pilot program to improve treatment for OUD within a state by focusing on a state’s system of OUD care; many participants were enthusiastic about this idea. Key points made include:

Such a pilot can be conducted in more than one state simultaneously.
The pilot should incorporate the concept of a learning health care system.
Policy and practice need to be addressed.
Questions that are critical to states or regions must be answered.
Conducting this pilot in one integrated health system would be helpful.
Urban and rural areas and diverse populations need to be included.
Existing evidence-based interventions must be implemented.
The low levels of MAT provision in the criminal justice system need to be addressed.

Next Steps and Adjournment

The inventory of high-priority issues—more than 30 papers from participating organizations that identify research gaps—was distributed to participants. This unique inventory may continue to grow and should inform all agencies’ and organizations’ strategic thinking and plans on the issues, as well as the NIDA’s.

A summary of this meeting will be distributed to participants and placed on the NIH website for public viewing.