Catching Up with Pilot Grant Recipient: Rachel Epstein

Rachel Epstein

Rachel Epstein, MD, MScE, (she/her/hers) is an assistant professor of medicine and pediatrics at Boston Medical Center (BMC), a principal investigator at the Syndemics Lab, and an adult and pediatric infectious diseases physician working with patients living with or at risk for HCV, HIV, and other infectious diseases.

Rachel Epstein is a clinician-scientist with demonstrated experience in analyzing large datasets to inform hepatitis C (HCV), HIV, and substance use disorder interventions. A CHERISH pilot grantee in 2019, she worked closely with her co-investigators Benjamin Linas, CHERISH Population Data and Modeling Core director, and Shashi Kapadia, a CHERISH Research Affiliate and an infectious diseases physician at Weill Cornell Medicine, to analyze claims data from Truven Analytics Marketscan, and investigate the unintended or spillover effects restrictive Medicaid policies have on commercially insured individuals. Since the conclusion of her CHERISH pilot grant experience, Epstein continues to apply economic evaluation methods to her research.

“Health economics is a powerful tool to help reach program administrators and policymakers to point out what interventions are not only clinically effective, but are also feasible to fund with the resources available. Health economics studies can predict realistic strategies to eliminate infectious diseases and reduce negative health outcomes such as opioid overdoses and deaths. The pilot grant experience – the skills learned, the mentorship I received, and the analysis I completed – helped build the foundation for my NIH/NIDA career development award,” Epstein said.

In her latest NIH/NIDA career-development project, Epstein aims to evaluate the clinical effectiveness and cost-effectiveness of introducing interventions in perinatal care to reduce opioid-related mortality and HCV transmission.

What are some critical issues at the intersection of perinatal care, substance use, and HCV?

One important issue is the lack of HCV testing among infants and children. My research is among a growing body of literature demonstrating that less than half of HCV-exposed infants or adolescents diagnosed with opioid use disorder are tested for HCV. In addition, very few children have been treated despite the availability of direct-acting antivirals to cure HCV in children as young as 3 years old.

Pregnant and post-partum women are also populations who need improvements in testing, linkage to care, and treatment. My K-award research aims to determine which areas in the HCV care cascade for pregnant and post-partum women could be most helpful to direct resources into to improve HCV diagnosis and cure.

Who are the peers or mentors that have influenced you or guided you?

Dr. Benjamin Linas is my primary research mentor at Boston University Chobanian and Avedisian School of Medicine/Boston Medical Center and for my K01 career development award. He has been instrumental in helping to shape and grow my research career from the beginning with his guidance on projects and grants.

Dr. Sabrina Assoumou has also been a wonderful career mentor and role model for me as a successful fellow med-peds physician incorporating pediatric and adult infectious diseases research together, and an excellent senior author on an article we published together on adolescent HCV testing in JAMA.

So many more mentors and peers have been critical throughout my career: my fellowship program directors, Drs. Elizabeth Barnett and Tamar Barlam, as well as Drs. Stephen Pelton and Bob Horsburgh. My peers have also been key in giving feedback, moral support, and great research suggestions along the way.

Do you have advice for early career investigators looking to incorporate health economics or modeling into their research?

CHERISH has been an incredible resource for learning how to conduct cost-effectiveness analyses and understand the impact of policy on health outcomes. For me, starting out by participating in cost-effectiveness work others led, taking coursework in decision modeling, attending conferences through CHERISH, and then leading cost-effectiveness analyses has helped build a strong foundation in health economics research. I have applied this perspective and knowledge in my successful applications for a career development award and for the Patrick and Catherine Weldon Donaghue Foundation’s Greater Value Portfolio grant, and I am excited to continue learning and collaborating on new modeling and health economics projects through CHERISH and other avenues.

Addressing Statistical Issues Encountered in Conducting Prospective Economic Evaluations Alongside Clinical Trials

Summary of research
Ali Jalali

Ali Jalali, PhD, MA, is a CHERISH Research Affiliate, former pilot grant recipient, and an assistant professor of population health sciences at Weill Cornell Medicine.

Conducting an economic evaluation alongside a randomized clinical trial provides valuable information for clinicians, insurers, and policy stakeholders. Integrating an economic evaluation within a trial provides researchers the opportunity to collect patient-level data at multiple time points on both cost and effectiveness outcomes, and to measure the comparative impact of treatments on these outcomes while controlling for confounding factors, such as a participant’s medical history and socioeconomic characteristics.  

However, bias associated with missing data is a common problem encountered when collecting data from trial participants, and sampling uncertainty must also be accounted for when determining the likelihood that a treatment is cost-effective. The nonparametric bootstrap method is widely recommended to account for sampling uncertainty in economic evaluations, but there are multiple valid approaches that researchers can choose from to address missing data biases. There are also multiple ways of combining these methods together in a single analysis.

In a new study published in the 2022 theme issue Epidemiological Reviews titled “Clinical Trials: Current Status and Future Directions,” Assistant Professor Ali Jalali from Weill Cornell Medicine; Associate Professor Sean M. Murphy from Weill Cornell Medicine; Professor Rulla Tamimi from Weill Cornell Medicine; and Professor Sterling M. McPherson from Elson S. Floyd College of Medicine at Washington State University, outline potential pros and cons of approaches to address these statistical issues. The authors argue that the method a researcher chooses to implement to correct for missing data bias and how this method is combined with the nonparametric bootstrap are not independent decisions. This poses unique statistical and practical challenges for researchers conducting an economic evaluation alongside a clinical trial.

Using prior-published systematic reviews of economic evaluations and a targeted literature search, the authors identified and critically reviewed the statistical methods employed by 29 studies to address missingness, identified whether the nonparametric bootstrap was included in each analysis, and evaluated how each study combined these methods in order to highlight the advantages and shortcomings of each approach. Given the scope and complexity of potential statistical issues that may arise for researchers interested in including an economic evaluation in a clinical trial, the authors recommend establishing close and early collaboration between economists and clinical scientists in the clinical trial design process.

The study, “Econometric Issues in Prospective Economic Evaluations Alongside Clinical Trials: Combining the Nonparametric Bootstrap with Methods that Address Missing Data” was published in the Epidemiologic Reviews on September 14, 2022.

Engaging with People with Lived Experiences in OUD Modeling Research

the word opportunity is written in chalk lettering against a dark blue background

Drug overdose deaths continue to rise in the United States and Canada, with the risks for greater opioid use sharply increasing in Mexico. As modelers, we know simulation modeling is a valuable tool that can inform policies and interventions to prevent overdose deaths and reduce related harms including drug overdose, HIV, and hepatitis C virus infections. However, our research also has the potential to perpetuate discrimination against people who use drugs.

Rethink Our Approach to Modeling

In November 2021, modelers who attended the panel, “Engaging People with Lived Experience: Best Practices” co-hosted by the Center for Health Economics of Treatment Interventions for Substance Use Disorder, HCV, and HIV (CHERISH) and the University of California, San Diego, learned that models commonly make assumptions that do not reflect how people who use drugs live their lives. Oftentimes, models are also designed to address questions that are less relevant to people who use drugs. To prevent drug-related deaths, modelers and other researchers must collaborate with people with living experience to identify and implement effective solutions.

More recently we heard from experts working on the front lines of substance use research and advocacy about how to design and conduct research in collaboration with people with lived experience. In May 2022, Drug Policy Alliance, Urban Survivors Union, and the Network of Drug Researchers with Lived Experience co-hosted a four-part online series, “Beyond Participatory-Based Research,” to encourage researchers to push the boundaries of how research is typically conceptualized and think critically about the relationships we have with people who use drugs. The series also provided tools to authentically partner with people impacted by the drug war in every phase of the research process. 

Applications of community driven research to drug policy | Beyond Participatory-Based Research

Applications of Community Driven Research to Drug Policy

Applications of Community Driven Research to Drug Policy,” revealed how community-driven research can overcome power imbalances and structural issues in collaborations with academic researchers. This session was moderated by Jules Netherland, CHERISH Policy Advisory Board member and managing director for Research and Academic Engagement at Drug Policy Alliance, with Emily Einstein, Science and Policy Branch chief at NIDA presenting. Panelists included advocates from the Health in Justice Action Lab and the Network of Drug Researchers with Lived Experience and researchers from the University of North Carolina and the University of California, Los Angeles.

Innovative examples in community driven research | Beyond Participatory-Based Research

Innovative Examples in Community Driven Research

Innovative Examples in Community Driven Research,” highlighted researchers and organizations experienced in community-driven research and covered how best to partner with these organizations, develop the research capacity of all project participants, collectively identify community needs, develop research questions, and design and carry out research projects that directly benefit community needs by concurrently fostering public advocacy projects. The insights shared in this panel were also described in a recent commentary published in the International Journal of Drug Policy by Urban Survivors Union authors. This session was moderated by Aliza Cohen, research coordinator at Drug Policy Alliance. Panelists included Shaquita Borden, Mary Figgatt, Laura McTighe, and Louise Vincent.

Challenge the Academic Framework

These conversations during the past year have taught us as modelers to approach research not as an extraction of knowledge but as a collaborative process with people with living experience. Co-leading research projects with people with living experience can guide how the objective of the study is framed and help researchers break out of academic conceptual frames. Working with people affected by the drug war can also enhance the policy impact of research findings and identify critical real-world limitations.

Inspired by the “Beyond Participatory-Based Research” series, we provide the following suggestions to guide modelers on how to partner with people with lived experience.

Considerations for Conducting Community-driven Research

Be thoughtful from the start.

  • Consider why you want to engage with people with lived experience. Examining your reasons early on will guide research decisions and allow people with lived experience to meaningfully drive the project.
  • Consider potential harms that may arise from your research. Will this work contribute to or decrease the harms and stigma that people who use substances experience?
  • Understand and minimize the risks to the individuals you will engage with when designing and conducting your research. For example, depending on where and how research is conducted, participation may increase the risks of negative encounters with the criminal-legal system. Consult with individuals with living experience on potential risks you may not be aware of.

Let go of assumptions.

  • People with lived experience are the experts. They have the life experience and the knowledge as organizers and service providers to be trusted contributors to your research.
  • Let your research questions be co-created by people with lived experience. Don’t ask questions based on your own experience only. Be open to other perspectives and experiences.
  • People who use drugs are not a homogeneous group. Connecting with a variety of organizations for your research can be a foundational step to including diverse voices and perspectives that reflect unique experiences, thoughts, and beliefs.

See research as a collaborative process.

  • Collaborate with people with lived experience across the entire research project. Too often, the community is narrowly engaged in order to collect data but is left out of the analysis or the decision-making process of the research. People with living experience can provide valuable insights starting from the initial framing and conceptualization to evaluation and interpretation of the data, and publication and dissemination. 
  • Be flexible and inclusive of community priorities and needs. Individuals are more inclined to become involved if your project aligns with their priorities. Ask people with living experience what their community needs are and how your project may address or synergize with them.
  • Pay attention to word choice. Language adapts and evolves over time. Check that the language you use in substance use research does not further stigmatize the community and perpetuate discrimination.

Build effective partnerships.

  • Seek out organizations as partners. Organizations such as Drug Policy Alliance’s Department of Research and Academic Engagement can connect researchers to people with lived experience.
  • Compensate participants and think creatively about financial support. Compensation may look different for each participant or organization. Provide financial support in a format that is useful and minimizes administrative burdens.
  • Don’t rush the process. Building trust takes time and is the foundation of a good partnership. The most successful partnerships are built on clear communication and strong relationships of trust.

The article is written by Natasha Martin, CHERISH Research Affiliate and professor in the Department of Medicine at the University of California San Diego, and Bruce Schackman, CHERISH director and Saul P. Steinberg Distinguished Professor at Weill Cornell Medicine, with contributions from Caty Simon from Urban Survivors Union.

Prevalence of HIV Preexposure Prophylaxis Prescribing Among Persons with Commercial Insurance and Likely Injection Drug Use

The United States has made important strides to reduce the incidence and negative outcomes of HIV, including increasing the availability of effective treatment and prevention in the form of pre-exposure prophylaxis (PrEP). While the incidence of new HIV infections fell significantly from 130,000 in 1985, progress has stalled at approximately 39,000 new infections per year since 2013.

One challenge has been the transmission of HIV through shared injection equipment. PrEP has been shown to reduce HIV incidence by 74% among persons who inject drugs (PWID). However, the implementation and uptake of PrEP among PWID have been inadequate. To better understand trends in PrEP use among people who inject drugs, especially in the context of a worsening drug overdose epidemic, we investigated PrEP prescribing rates in a national database of commercially insured persons with an opioid or stimulant use disorder from January 1, 2010, and December 31, 2019.

Photo of Jake Morgan

Jake Morgan, PhD, CHERISH investigator and Research Assistant Professor at Boston University School of Public Health.

We used an established algorithm to include in our cohort individuals who were at least 16 years of age with a documented diagnosis of opioid or stimulant use disorder. We excluded those with evidence of HIV or hepatitis B to ensure any PrEP medication we found was not being prescribed for HIV treatment. We also used the algorithm to observe the association between injection drug use and PrEP uptake.

We found that injection drug use was associated with 1.5 times greater odds of having received PrEP compared to those with no evidence of injection drug use. However, the actual number of individuals receiving PrEP was abysmally low. Only 0.15% of individuals with evidence of injection use and 0.08% of individuals with no evidence of injection received at least one prescription during the study period. These findings suggest that this evidence-based prevention strategy is gravely underused.

"We need better ways of identifying those who would benefit from PrEP in claims data, but even an imperfect approach shows striking gaps."

This study also highlights urgent data challenges that need to be addressed to foster more accurate research on HIV and substance use disorder, particularly research conducted with observational data. For example, diagnosis codes for tracking opioid use disorder and injection behaviors are imperfect, and medical claims do not fully capture those in need of care because individuals do not seek care for these issues. Therefore, our study may have missed individuals with opioid use disorder or individuals who inject drugs who were misclassified as having no evidence of injecting.

These data challenges suggest that the individuals we included in our study were those with the most severe substance use issues including people who feel they need to seek care for their substance use, those who have providers who decide the use is worthy of documentation, and individuals who seek treatment for injection-related complications of substance use. These are also individuals with some of the greatest risks of becoming infected with HIV, which increases our concern about the alarmingly low PrEP prescribing we found.

To achieve the public health goal of reducing new HIV infections, we need to meaningfully engage with people who inject drugs and employ evidence-based prevention strategies. We need better ways of identifying those who would benefit from PrEP in claims data, but even an imperfect approach shows striking gaps. Aggressive expansion of PrEP for people who inject drugs is urgently needed.

The study, “Prevalence of HIV Preexposure Prophylaxis Prescribing Among Persons With Commercial Insurance and Likely Injection Drug Use,” was published in JAMA Network Open on July 12, 2022. Authors include Carl G. Streed Jr, MD, MPH; Jake R. Morgan, PhD; Mam Jarra Gai, MPH; Marc R. Larochelle, MD, MPH; Michael K. Paasche-Orlow, MD, MPH; and Jessica L. Taylor, MD.

Where to Find CHERISH This Fall

In October, CHERISH researchers and colleagues are heading abroad and across the states to share their latest research on harm reduction, syringe service programs, and HIV and HCV treatment.

Overseas, researchers are attending the 13th National Harm Reduction Conference in San Juan, Puerto Rico, and the 10th International Conference on Health and Hepatitis Care in Substance Users (INHSU) in Glasgow, Scotland. Presenters include Cristina Chin, research manager within the Department of Population Health Sciences at Weill Cornell Medicine; CHERISH Research Affiliates Xiao Zang, Shashi Kapadia, Czarina Behrends, Natasha Martin, and Tyler Bartholomew; and CHERISH investigator Alexander Walley.

In the states, CHERISH researchers and colleagues will make appearances at the 2022 Polysubstance Use Symposium in New York City, the American Public Health Association 2022 Annual Meeting in Boston, and the Society for Medical Decision Making (SMDM) 44th Annual North American Meeting  in Seattle. Look out for CHERISH Research Affiliates Thanh Lu, Xiao Zang, and Ali Jalali; Danielle Ryan, senior research coordinator in the Division of Comparative Effectiveness and Outcomes Research at Weill Cornell Medicine; and Erminia Fardone, assistant scientist at the University of Miami Miller School of Medicine.

Learn more about their presentations below.

graphic for harm reduction conference 2022

The 13th National Harm Reduction Conference

Presented by the National Harm Reduction Coalition

October 13 – 16, 2022, San Juan, Puerto Rico


logo for the INHSU conference

The 10th International Conference on Health and Hepatitis Care in Substance Users (INHSU)

October 19 – 21, 2022, Glasgow, Scotland, United Kingdom


logo of nyu langone health

2022 Polysubstance Use Symposium

Presented by NYU Langone’s Center for Opioid Epidemiology and Policy

October 20 – 21, 2022, New York, NY

logo of Society for Medical Decision Making

Society for Medical Decision Making 44th Annual North American Meeting (SMDM)

October 23 – 26, 2022, Seattle, WA


Banner of APHA 2022 Annual meeting

APHA 2022 Annual Meeting

November 6 – 9, 2022, Boston, MA


A banner presenting details of the AMERSA 2022 conference

AMERSA 46th Annual Conference

November 10 – 12, 2022, Boston, MA


Paying Patients Can Improve Engagement in Mental Health Treatment

This article originally appeared and was produced in conjunction with the University of Pennsylvania Leonard Davis Institute.

Gabriela Khazanov, PhD, is a postdoctoral fellow at the Penn Center for the Prevention of Suicide at the Perelman School of Medicine.

James McKay, PhD, is a Professor of Psychology in Psychiatry at the University of Pennsylvania and a CHERISH Advisory Board member.

Financial incentives are known to increase patient engagement in a wide range of health programs, including programs for maintaining abstinence from substances, attending vaccination and screening appointments, and increasing physical activity. However, they have rarely been used for mental health care beyond treating substance use disorders.

Showing up and being involved in mental health treatment is half the battle for patients, and our new study, a meta-analysis of existing research, shows that financial incentives can help. The study synthesized findings from 39 papers and found that patients who receive such payments are more likely to attend their mental health treatment, adhere to their medications for mental health disorders, and complete their mental health treatment goals like doing homework.

Effects of the financial incentives were large, with findings suggesting multiple ways in which incentives could be used, such as at the start of treatment to improve attendance for initial sessions, for preliminary treatment goals like signing a treatment plan, and for reducing dropout from psychotherapy, case management, or medication management. Findings also indicated that incentives to enhance treatment engagement might have beneficial downstream effects on mental health symptoms, functioning, and quality of life.

However, efforts to put these findings into practice have run into significant barriers. Some people worry that incentives may coerce patients into seeking treatment or undermine their autonomy, reduce their intrinsic motivation to engage in treatment, or cost too much for the health system to maintain. Research demonstrates, however, that providers and patients typically value these interventions once they are put into practice, and patients’ intrinsic motivation shows no sign of decreasing after receiving incentives. These financial interventions generally are cost-effective in the long term, particularly when factoring in patients’ potential work productivity or criminal justice costs.

Next Steps

What will it take to move this promising and effective strategy forward? A variety of strategies would help:

  • Incentive programs should expand beyond individuals with only substance use disorders to those with other mental health disorders. For example, financial incentives can be used to improve medication adherence among individuals with depression and anxiety.
  • Insights from behavioral economics should be used to tailor how incentives are presented to patients and how providers are motivated to implement interventions.
  • Corporate partnerships can help overcome financial barriers. These partnerships would decrease implementation costs and connect corporations with members of their communities, allowing their philanthropic contributions to impact patient care directly. Additionally, corporations can use these programs to promote employee mental health by incentivizing outcomes such as mental health treatment attendance, just as many already do to incentivize other healthy behaviors such as exercise.
  • Regulatory limits on incentives for patients treated through federally and state-funded programs like Medicaid and Medicare need to be lifted to make the widespread implementation of incentive programs possible. The regulatory landscape is shifting in the right direction, but broad policy modifications are necessary to allow programs offering incentives to be used to their full capacities.
  • To further encourage implementation, federally and state-funded programs could provide direct funding for incentives while also enabling oversight of incentive programs to prevent the potential for fraud and abuse that regulatory limits were intended to address.

Research shows that paying patients can significantly improve their engagement in mental health treatment. It is time for practice to catch up with the evidence and for payers and providers to utilize financial incentives to improve mental health outcomes.

The review, “Do Financial Incentives Increase Mental Health Treatment Engagement? A Meta-Analysis,” was published in the June 2022 issue of the Journal of Consulting and Clinical Psychology. Authors include Gabriela K. Khazanov, Paige E. Morris, Alexander Beed, Shari Jager-Hyman, Karoline Myhre, James R. McKay, Richard S. Feinn, Elaine M. Boland, and Michael E. Thase.

Cost-effectiveness of Extended-release Injectable Naltrexone among Incarcerated Persons with Opioid Use Disorder Before Release from Prison Versus After Release

interior of a sleeping quarter within criminal legal system

Compared to the general population, individuals with opioid use disorder (OUD) who are involved with the criminal-legal system have a much higher risk of a fatal overdose after release from incarceration. Yet, stigma, costs, and preferences for non-narcotic treatments in criminal-legal settings have stalled the implementation of life-saving treatment within the criminal-legal system.

Extended-release naltrexone (XR-NTX), methadone, and buprenorphine-naloxone are three widely studied and evidence-based medications for opioid use disorder. Given the preference for non-narcotic treatments in many carceral settings, XR-NTX is often selected rather than methadone or buprenorphine-naloxone despite its relatively high cost.

In a new study published in the Journal of Substance Abuse Treatment, researchers examined whether the combination of XR-NTX before release and linkage to community-based treatment has the potential to be a cost-effective treatment, compared to a referral to community-based treatment after release only. While there are documented barriers to meeting detoxification requirements before starting XR-NTX, randomized studies have shown XR-NTX to be just as effective as buprenorphine-naloxone among patients who successfully start treatment.

Ali Jalali

Lead author Ali Jalali, PhD, MA.

To reduce the health consequences of OUD among incarcerated populations and lower the economic burden of OUD on society, Ali Jalali, lead author, a CHERISH Research Affiliate, and assistant professor of population health science at Weill Cornell Medicine, led a comprehensive economic evaluation with colleagues at Weill Cornell Medicine, the Crime Prevention Research Center, Johns Hopkins University, and the University of Pennsylvania. They compared health economic outcomes of administrating XR-NTX before release plus linkage to community-based care, versus providing only a referral after an individual with OUD is released from incarceration. The incremental cost-effectiveness ratio was the primary measure of cost-effectiveness and was calculated at 12- and 24-week periods. Effectiveness measures included both quality-adjusted life-years (QALYs) and a clinical outcome, time free from opioids, as recommended by guidelines and the team’s own research.

summary of study findings

Based on the study, Jalali and colleagues concluded:

  1. XR-NTX before release from incarceration led to higher average OUD-related and total costs from a state policymaker perspective; however, the high rate of patient attrition observed in the trial limited the study’s conclusions on other health economic outcomes.
  2. XR-NTX before release, plus referral, is likely cost-effective compared to referral-only after release, depending on stakeholders’ willingness-to-pay for a year free from opioids.

The authors found that administering XR-NTX to individuals who are incarcerated with opioid use disorder, prior to their release, may provide value for stakeholders and bridge a well-known treatment gap for this vulnerable population. The study provides a comprehensive estimate of the healthcare and non-healthcare costs for incarcerated populations following their re-entry into their community, and evidence that supports connecting individuals receiving pre-release XR-NTX for OUD to medical and social services. Additional research is needed to determine effective methods that will improve treatment retention after release.

The study, “Cost-effectiveness of Extended-release Injectable Naltrexone among Incarcerated Persons with Opioid Use Disorder Before Release from Prison Versus After Release,” was supported by funding from National Institute on Drug Abuse and published in the Journal of Substance Abuse Treatment on July 1, 2022.

Co-authors include Philip Jeng, research manager at Weill Cornell Medicine; Daniel Polsky, CHERISH Advisory Board member and Bloomberg Distinguished Professor at Johns Hopkins University; Sabrina Poole and George E. Woody from the University of Pennsylvania; Yi-Chien Ku from the Crime Prevention Research Center; and Sean M. Murphy, CHERISH Methodology Core co-director and associate professor of population health science at Weill Cornell Medicine.