1 INTRODUCTION

The past decade has seen a proliferation of pragmatic clinical trials (PCTs) conducted in health care delivery settings. PCTs can be more informative than traditional randomized clinical trials because they attempt to answer questions related to the effectiveness of therapies in the “real world.”1 However, given the proximity of PCTs to the point of care, they require some level of involvement by healthcare professionals. A key lesson from the first cohort of PCT demonstration projects funded through the National Institutes of Health (NIH) Health Care Systems Research Collaboratory is that engagement and buy-in at the organizational level is necessary but not sufficient for successful PCT implementation.2

Despite the importance of practicing clinicians to research conducted in the context of health care delivery, we know surprisingly little about how they perceive their roles in such research. A 2017 National Academy of Medicine (NAM) paper notes that the current health care ecosystem contains numerous impediments to clinician participation in learning activities, including misalignment of priorities, productivity pressures, and lack of training and incentives.3 Studies assessing clinician perspectives on research participation have reported generally favorable views toward research but also identify multiple barriers to participation, with lack of time and competing priorities topping the list.4, 5 Common motivations for clinicians to participate in research include acquiring knowledge to improve patient care, contributing to scientific knowledge, and professional development.4, 6, 7

Previous efforts to engage front-line clinicians in research activities have suggested a range of strategies for improving engagement. For example, the importance of selecting topics of mutual interest to researchers and clinicians is frequently highlighted.8-10 Other recommendations for improving clinician engagement include involving clinicians in adapting study protocols to fit the practice environment, procedures, and workflow11, 12; providing comprehensive and ongoing training;13 identifying site-level study champions;12 and recognizing individual study contributions.14, 15

Empirical evidence and practical guidance concerning clinician engagement in research remain limited and there is a fundamental gap in understanding the extent to which practicing clinicians should be involved in PCTs. We conducted a qualitative study to better understand clinician views on their potential role in research conducted in the context of health care delivery and to develop guidance for researchers on engaging front-line clinicians in PCTs. In phase one, we conducted focus groups with clinicians who spend the majority of their time providing direct patient care and have limited research experience. Phase two involved key informant interviews with PCT research teams and clinicians participating in the ADAPTABLE (Aspirin Dosing: A Patient-centric Trial Assessing Benefits and Long-Term Effectiveness) trial.16, 17

2 METHODS 2.1 Focus groups 2.1.1 Overview

A focus group approach was chosen for this phase of the project so that participants would have the opportunity to hear perspectives from their peers that they may not have otherwise considered. We identified focus group sites that were diverse with regard to the type of health system.

2.2 Participants

Four organizations agreed to participate in focus groups, including: (a) Clinical Director's Network, Inc. (CDN)—a practice-based research network (PBRN) in New York City; (b) AllianceChicago—a network of community health centers serving primarily low-income and uninsured patients; (c) Edward Hines Jr. VA Hospital—a Veterans Health Administration hospital serving veterans in the Chicago area; and (d) Rush University Medical Center—an academic medical center located in Chicago, IL.

The project advisory committee included representatives from each of the participating sites who, along with designated site coordinators, assisted with recruitment and meeting logistics. Invitations to participate in a 90-min focus group were distributed via multiple channels (newsletters, email blasts, fliers). Physicians, nurses, nurse practitioners, physician assistants, and related professionals who spend the majority of their professional time providing direct patient care were invited to participate. Participants were offered $100 honoraria (except for one site where this was not permitted) and were provided a meal appropriate for the time of each group.

2.2.1 Procedures

Participants completed informed consent upon arrival followed by a brief, anonymous questionnaire intended to characterize the group in terms of professional roles and research experience. Focus groups were audio-recorded and transcribed for qualitative analysis.

Discussions were led by experienced facilitators using a discussion guide with questions and probes that were flexible enough to adapt to the flow of conversation within each group. Following an introduction to the goals of the study, participants were asked a series of questions pertaining to the roles that clinicians are sometimes asked to play in comparative effectiveness and patient-centered outcomes research (CER/PCOR) studies.

Based on the assumption that participants may have had very limited research exposure, we began by discussing quality improvement initiatives before transitioning to questions about hypothesis-driven research and the difference between traditional randomized clinical trials and pragmatic trials. The remainder of the discussion focused on clinician involvement across the phases of pragmatic trials.

To facilitate a more in-depth discussion around pragmatic trial implementation, we introduced a hypothetical case study modeled after the ADAPTABLE trial. As shown in Figure 1, the roles of patients, clinicians, and researchers in the study were described as: (a) research team identifies eligible patients using EHR, (b) clinician provides permission to contact eligible patients, (c) research team recruits and enrolls eligible patients, (d) patient completes informed consent process and is randomized via patient portal, (e) patient obtains prescribed aspirin dose and takes as instructed, and (f) data is collected from EHR by research team and from patients via the portal. After reacting to this initial protocol, participants were asked to consider how their opinions might change with various modifications.

Pragmatic clinical trial case study

2.3 Interviews 2.3.1 Overview

The goal of this project phase was to examine clinician engagement in the context of a real-life pragmatic trial. The timing of our project offered a valuable opportunity to learn from the ADAPTABLE trial. ADAPTABLE, the first pragmatic trial to leverage PCORI's National Patient-Centered Clinical Research Network (PCORnet), aims to identify the optimal dose of aspirin for secondary prevention in patients with atherosclerotic cardiovascular disease.

2.4 Participants

The Mid-South and REACHnet CDRNs agreed to participate in our project during the proposal process and additional CDRNs were invited once the project was underway.

For each participating CDRN, we invited members of the ADAPTABLE study team, including CDRN- and site-level investigators and project managers, to participate in interviews. In addition, we asked study team members to help identify clinicians who might be willing to participate in brief interviews.

2.4.1 Procedures

One-hour, semi-structured telephone interviews with ADAPTABLE team members were conducted by experienced members of the project team and, except for one group of investigators, were one-on-one. Interviews were audio-recorded and transcribed for qualitative analysis.

The discussion guide included open-ended questions regarding experience with the ADAPTABLE trial and directed questions regarding the role of clinicians in the trial and specific challenges related to clinician engagement.

Twenty-minute semi-structured telephone interviews with participating clinicians included a series of open-ended questions regarding their experience with the trial.

2.5 Qualitative analysis

De-identified transcripts were checked for accuracy and uploaded to NVivo 12.

Two study team members independently coded a subset of focus group transcripts using a codebook based on a preliminary literature review. After resolving discrepancies and discussing additional codes, the codebook was revised, and the remainder of the transcripts was coded by a single team member.

The same codebook was applied to the interview transcripts with additional codes added to reflect themes that did not emerge during focus group discussions.

4 DISCUSSION

In phase one of this study, focus group discussions with physicians, nurses, and other care providers highlighted the importance of involving front-line clinicians in PCT planning. PCTs are intended to inform clinical decision-making, and therefore should address questions that are relevant to clinical practice. Given their motivation to improve patient care, having a say in what research topics are selected is likely to increase clinician willingness to play a role in PCTs. Focus group participants also noted that there are clinicians who are interested in research but do not know how to get involved. Expanded opportunities for clinicians and researchers to collaborate via multiple platforms and modalities were encouraged.

Focus group participants also emphasized the importance of involving clinicians in designing studies that are feasible to implement in the context of clinical care and adapting study protocols to fit workflows that are often unique to a specific clinic. More active involvement by a range of clinical stakeholders in the PCT planning stage may help researchers avoid common barriers to trial implementation. Funding mechanisms that allow adequate time and flexibility for protocol development (eg, the UG3/UH3 model employed for the NIH Health Care Systems Research Collaboratory Demonstration Projects) can allow for more meaningful use of clinician feedback.

Engagement of front-line clinicians in PCT implementation occurs along a spectrum. At one extreme, “low-touch” PCTs may involve direct recruitment and enrollment of eligible patients by members of the research team with no need for clinician involvement in delivering the study intervention or conducting follow-up. A “low-touch” approach may be viable for PCTs involving a simple, low-risk intervention where there is a large population of eligible patients, an electronic infrastructure to facilitate patient outreach, and a willingness to accept a very low rate of enrollment. Based on their experience recruiting for the ADAPTABLE trial, Pfaff et al18 suggest a combined approach that includes electronic and in-clinic recruitment and recommend future work to understand the demographic skew that results from electronic recruitment approaches.

At the other end of the spectrum, “high-touch” PCTs can be defined as those in which clinicians play an active role in any or all of the following: identification of eligible patients, recruitment and enrollment, delivery of the study intervention, and data collection and participant follow-up. A “high-touch” approach may be necessary when there are a limited number of eligible patients, barriers to identifying patients via EHR, institutional policies that limit researchers' direct access to eligible patients, a patient population that requires extra support to participate, or complex study interventions.

We propose that many PCTs, such as ADAPTABLE, can be categorized as “medium-touch” with regard to clinician involvement. A “medium-touch” approach works best when there are an adequate number of eligible patients at participating sites, an EHR that is functional for identifying eligible patients, site-level study champions, and adequate space to accommodate research activities without impeding clinic workflow. The optimal research team for this type of trial has adequate resources to provide “boots on the ground” and carefully selected and trained research staff (eg, those with adequate emotional maturity and relationship-building skills). Successful trial implementation is more likely when researchers take the time to build awareness and buy-in at every level of the organization, identify clinician and staff champions, develop a detailed understanding of site-level operations, and adapt study protocols to accommodate individual preferences and workflow.

With increasingly high demands on clinician time and the fact that their first priority is, as it should be, providing the best possible patient care, research teams should take on as much of the burden of trial implementation as possible. A “medium-touch” approach to recruitment and enrollment is grounded in the idea that clinician effort should be focused where it is most critical—to reassure eligible patients that trial participation is a safe alternative for them. A common theme from both focus group and interview participants was that patients place a lot of trust in their clinicians. Thus, as is the case for traditional randomized clinical trials, clinician support may be a critical element of successful PCT recruitment and protocols that optimize this role while limiting clinician burden may be the most successful.

With the increasing focus on learning health systems as a nexus for real-world research, the need for optimization of PCTs will only grow in importance. We have attempted to offer solutions to some common barriers to engaging clinicians in PCTs but given that this is a relatively nascent addition to the research landscape, many questions remain. Ongoing and recently completed trials may provide additional insights regarding successful and unsuccessful strategies for PCT implementation and efforts to capture these learnings such as the NIH Collaboratory Living Textbook of Pragmatic Clinical Trials remain critical.19 As noted in a recent article by Simon et al,20 the goal of a health care system where the vast majority of clinical decisions are based on reliable evidence can only be achieved with commitment by researchers, health system leaders, research funders, and regulators in partnership with patients, caregivers, and clinicians.

ACKNOWLEDGEMENTS

The authors would like to thank the members of our advisory committee for their helpful input and guidance throughout this project. Special thanks to all of the clinicians who took time out from their busy schedules to participate in focus groups and interviews. This study was funded by Patient-Centered Outcomes Research Institute Engagement Contract #3518.

CONFLICT OF INTEREST

The authors have no conflicts of interest to declare.