Two faces looking in opposite directions with a question mark in between them..

Ethical Questions in Implementation Science - Workshop Summary

December 10 - 11, 2020
Virtual Zoom Event


The National Institutes of Health’s (NIH) National Heart, Lung, and Blood Institute (NHLBI) hosted a virtual workshop on December 10-11, 2020, to explore the unique ethical questions that arise in the conduct of implementation science which often includes research with underserved populations precisely in order to address health inequities.


Workshop Videos 

Workshop Day 1

Workshop Day 2 


On average, it takes 17 years for research findings to enter routine clinical practice, and even then only about 14% are commonly used. Implementation science aims to accelerate the sustained adoption, scale-up, and spread of proven-effective interventions. In this rapidly-growing field, new ethical dilemmas have arisen that demand resolution. For example, interventions are frequently cluster randomized within health care systems, so the “participant” (who gives consent) is often the practitioner (or site). Separate consent may be needed later to interact with or collect identifiable data from patients. In other studies, obtaining consent may be impossible, contraindicated (because people behave differently if they know they are being studied), or meaningless (if there is no way participants can opt out). The use of “gatekeepers,” trusted community leaders who may be asked for consent or permission, may be helpful but also problematic, and the moral status of groups and subpopulations remains poorly-defined. Additionally, the very fact that this research is often conducted (and is particularly needed) in underserved populations makes addressing the ethical issues even more imperative. We can draw some conclusions based on other types of research (such as comparative effectiveness studies, cluster-randomized trials, and other work with underserved populations), but the unique nature of implementation science demands that normative and empirical bioethics research be done to help clarify how to best move forward. This workshop was convened with the objective of bringing together experts in bioethics, research ethics, and implementation science to identify bioethics research opportunities in the conduct of implementation research.

Overarching Themes:

  • Identification of the research subject: Implementation science, in a sense, lies at the intersection between research and quality improvement; there is a lack of clarity around the boundaries of the two from ethical and regulatory points of view. In these studies, the researcher is intervening not on individuals but on units/systems and/or processes. Implementation science tends to be multi-level, including staff, patients, and caregivers, so the research needs to be examined through each of their perspectives. This also influences how the field approaches consent: is consent possible in the context of what the researchers want to accomplish? From whom is consent possible and necessary? Does a research proposal meet the requirements for a waiver of informed consent? If consent is needed, for what is consent being granted: permission to intervene or for the gathering of data separate from the intervention? (Consider a common intervention: signs are posted in some units but not others to remind staff to wash their hands. Rates of staff hand washing are observed and recorded, and subsequently infection rates are compared in the units.) Key take home messages from this session included the fact that openness and transparency are important to all concerned, and that guidelines provided to IRBs and DSMBs should support both transparency (and thus autonomy) and research integrity.
  • Influencing behavior while preserving autonomy: given that the purpose of implementation research is to discover how to increase the uptake of proven-effective interventions, it often includes efforts to alter the behavior of organizations, health care providers, or individuals. In this part of the workshop, participants considered the ethics of behavioral economics. Behavioral economics, credited to Richard Thaler decades ago, relies on these principles:
    • Decisions are strongly influenced by context, including the way choices are presented
    • Behavior and decision making are subject to cognitive biases, emotions and social influence
    Behavioral economics employs nudges. A nudge is a change in the way choices are presented or information is framed that alters people’s behavior in a predictable way without restricting choice. Nudges should be transparent and never misleading, easy to opt out of if needed, and aligned with the welfare of the person being nudged. Examples of the effectiveness of behavioral economics in changing health-related behaviors include 1) a study demonstrating that game-based incentives increase physical activity among overweight patients; and 2) an experiment in which financial incentives were used to promote smoking cessation, resulting in a significant difference in cessation among those offered the incentive vs. controls, an effect that remained true one year post-study. Nudges can help practitioners provide better care to patients, and can help individuals make the choices that they truly want to make. Indeed, nudging can be compatible with autonomy if they help the person make the choice they would like to make (were it not for fatigue, etc.) and if the person does not/would not repudiate or resist the nudge process that was used to influence their choice (second-order theories of autonomy)—this is an empirical question. Further, there are other goods in addition to autonomy—protection and promotion of individual autonomy is but one value in healthcare.
  • Working with underserved populations: Examples of critically important issues that implementation science needs to address include the U.S. maternal mortality rate, which is worsening, and the relative increase in mid-life mortality driven by Appalachia and other rural areas. Excess mortality is driven by specific locations and their challenges; this does not affect only one race or ethnicity. The challenges in addressing these issues center around trust/mistrust, and engagement in general. Panelist Dr. Aisha Langford shared an infographic from the South Carolina Department of Health and Environmental Control that identified aspects of meaningful community engagement, including plain language, information sharing, transparency, dialogue, attention and behavior, active listening and relationship building. Referring back to the previous day’s discussion of informed consent, she asked if implementation scientists were actually listening to underserved populations or simply doing town halls as a check-off for a report. The important thing is to build relationships over time, rather than to do so only on a specific project. Discussion in this part of the workshop also included special considerations for implementation science work in low- and middle-income countries.

Identification of Outstanding Opportunities/Gaps:

As the workshop drew to its conclusion, participants were encouraged to outline areas for future research in the ethics of implementation science research. The following were identified:

  • Empirical Research
    • Embed within implementation studies evaluations of the impact on subjects and processes for consent and communication
    • Evaluate how engagement of stakeholders can help to shape approaches to consent and communication with subjects—in a way to maintain trust
    • Evaluate the impact of engagement strategies on multiple dimensions of research conduct
    • Study different behavioral-science driven strategies from the perspective of key stakeholders (both how they should be created/the meaning/impact of being nudged or manipulated)
    • Categorize and study context-adaptive approaches to implementation science trials
  • Innovative Approaches
    • Standards for single IRBs, which are now required for multi-site federally funded research. There appears to be appreciable heterogeneity among IRBs regarding how to view different types of implementation science. Is there an opportunity to develop consensus standards (not regulatory change)?
    • Development of creative approaches within institutions to provide oversight and review of studies that straddle the border of quality improvement and research in ways that promote development of high quality data rather than incentivizing use of methods that don’t look like research
    • Explore opportunities for greater use of DSMBs as a source of oversight

Workshop Chair

Neal Dickert, MD, PhD, Associate Professor of Medicine, Division of Cardiology, Emory University School of Medicine

Workshop Leaders

Kathleen Fenton, MD, NHLBI Division of Cardiovascular Sciences (DCVS)
George Mensah, MD, NHLBI Center for Translation Research and Implementation Science (CTRIS)
Lora Reineck, MD, MS, NHLBI Division of Lung Diseases (DLD)
Mollie Minear, Mollie Minear, PhD (DCVS)
Cheryl Anne Boyce, Cheryl Anne Boyce, PhD (CTRIS)
Catherine M. Stoney, PhD (CTRIS)
Joseph Millum, PhD, NIH, Clinical Center, Department of Bioethics
Catherine Burke, MA (DCVS)

Workshop Speakers/Moderators in Order Listed in Workshop Agenda

George Mensah, MD, Director, Center for Translation Research and Implementation Science (CTRIS)
Neal Dickert, MD, PhD, Associate Professor of Medicine, Division of Cardiology, Emory University School of Medicine
Donna Spiegelman, ScD, Professor of Biostatistics, Statistics, and Data Science, Director, Yale Center for Methods in Implementation and Prevention Science, Yale School of Public Health
Steven Joffe, MD, MPH, Professor of Medical Ethics and Health Policy, University of Pennsylvania Perelman School of Medicine
Jeremy Kahn, MD, MS, MPH, Professor of Critical Care Medicine and Health Policy & Management, University of Pittsburgh School of Medicine
Nancy Kass, ScD, Vice Provost for Graduate and Professional Education, Johns Hopkins University
Scott Kim, MD, PhD, Senior Investigator, NIH Clinical Center, Department of Bioethics
Daniel Matlock, MD, MPH, Associate Professor of Medicine, Division of Geriatrics, Colorado University School of Medicine
Meeta Kerlin, MD, MSCE, Assistant Professor of Medicine, University of Pennsylvania Perelman School of Medicine
Jennifer Blementhal-Barby, PhD, MA, Cullen Professor of Medical Ethics and Associate Director of the Center for Medical Ethics, Baylor College of Medicine
Garth Graham, MD, MPH, Vice President of Community Health, CVS Health
Aisha Langford, PhD, MPH, Assistant Professor, Department of Population Health, Division of Comparative Effectiveness and Decision Science, New York University Grossman School of Medicine
Jim Lavery, PhD, MSc, Conrad N. Hilton Chair in Global Health Ethics, Professor, Hubert Department of Global Health, Rollins School of Public Health, Emory University Center for Ethics

Speaker Bios

For More Information

Contact Kathleen Fenton at