Embedding research recruitment in a community resource e-prescribing system: lessons from an implementation study on Chicago’s South Side

Abstract

Objective

The study sought to implement and assess the CommunityRx e-prescribing system to recruit research participants from a predominantly non-Hispanic Black community on Chicago’s South Side.

Materials and Methods

CommunityRx integrates with electronic medical record systems to generate a personalized list of health-promoting community resources (HealtheRx). Between December 2015 and December 2016, HealtheRxs distributed at outpatient visits to adults with asthma or chronic obstructive pulmonary disease also incentivized participation in a pulmonary research registry. Usual practices for registry recruitment continued in parallel.

Results

Focus groups established acceptability and appropriateness among the target population. Pulmonary research registry recruitment information was included on 13 437 HealtheRxs. Forty-one (90% non-Hispanic Black) patients responded with willingness to participate and 9 (8 non-Hispanic Black) returned a signed consent required to enroll. Usual recruitment practices enrolled 4 registrants (1 non-Hispanic Black).

Discussion

Automating research recruitment using a community e-prescribing system is feasible.

Conclusions

Implementation of an electronic medical record–integrated, community resource referral tool promotes enrollment of eligible underrepresented research participants; however, enrollment was low.

patient recruitment, health information technology, electronic health records, minority health, health disparities

BACKGROUND

Inclusion of underrepresented minority populations is essential to producing research generalizable to a diverse national population; however, many clinical researchers experience difficulty recruiting historically marginalized, “hard-to-reach” populations.^1,2 Among other strategies to promote health equity, the National Institutes of Health has called for innovation in recruitment strategies to increase participation of minority populations.³

Lack of awareness about opportunities to participate in research is a prevalent, modifiable barrier to minority recruitment.^4–6 Several targeted outreach strategies to increase awareness of opportunities among minority groups have been described.⁷ Successful methods include time- and resource-intensive strategies, such as community-engaged and direct outreach efforts.^8–10

Electronic medical record (EMR) systems are emerging as promising and cost effective to identify and recruit patients for research.^11,12 Primary uses of EMR systems for recruitment include cohort identification,¹³ real-time recruitment support,^14–17 and messaging via patient portals.^18–20 However, EMR portal use is low among historically marginalized populations, including African Americans.^21,22 Community-driven strategies that leverage clinical informatics to promote research recruitment are needed.²³

This study presents the development and first test of CommunityRx, a patient-centered, EMR-integrated community resource referral platform, for its ability to promote clinical research recruitment among traditionally underrepresented groups.^24,25 CommunityRx was developed to promote population health and reduce health disparities by increasing all patients’ knowledge about and access to local health-promoting community resources. To achieve this goal, CommunityRx used an asset-based community-engaged process that included clinicians, researchers, community practitioners, and residents.^24,26–28 The main patient-facing intervention involved delivery, during healthcare visits, of a “HealtheRx,” a personalized list of community-based programs and services for basic and other health-related social needs (Figure 1).^24,29 We used CommunityRx to recruit eligible patients to the University of Chicago Pulmonary Research Registry (PRR). This case study describes implementation of the CommunityRx e-prescribing system to recruit research participants in a predominantly non-Hispanic Black community.

Figure 1.

Sample HealtheRx with information about registry. COPD: chronic obstructive pulmonary disease.

MATERIALS AND METHODS

CommunityRx was developed with a 2012-2015 Health Care Innovation Award from the Center for Medicare and Medicaid Innovation and tested in a 16 contiguous ZIP code region (population 983K, 62% non-Hispanic African American or Black) on Chicago’s South Side. This case study took place in the same geography. The HealtheRx (Figure 1) was automatically generated using novel algorithms that linked computable phenotypes (eg, individual characteristics, health and other conditions)³⁰ documented in the patient’s EMR to indicated community-based self-care resources (Figure 2). The clinician or staff presented the printed HealtheRx to the patient at the point of care. Additional details on the CommunityRx system are described elsewhere.^24,25

Figure 2.

High-level technical integration workflow. EMR: electronic medical record.

Focus groups

Using plan-do-study-act methodology,³¹ the idea to include research opportunities among other community resources on the HealtheRx emerged from observations made during CommunityRx implementation at academic medical center sites. This idea was deliberated by the research team (about 30 people, including community leaders, community health workers, clinicians, and researchers). To gain wider community input on the idea before launching a study, we were able to opportunistically add 2 questions to already planned focus groups of HealtheRx recipients to elicit their thoughts about getting information about research studies on the HealtheRx. Facilitators presented participants with sample HealtheRxs that included, among other community resources, information about health-related research studies. Sessions (May 2015) were audio-recoded and transcribed. Responses to targeted questions were analyzed using conventional content analysis.³² This inductive analytic method, “used when limited research or theory exists about the phenomenon,”³³ was applied to systematically generate new insights and key concepts, grounded in the data, that emerged from targeted questions. Participant responses were analyzed by 2 trained researchers who independently generated open codes from line-by-line analysis of the transcribed text, documented the codes in a key, adjudicated the codes to achieve consensus, applied the codes to a final analysis of the data, and generated counts to quantify responses. Sample size was determined by a priori objectives of the focus group.

E-prescribing research

Informed by focus group findings, we sought to partner with a study that was actively enrolling adult patients. The PRR was selected for this study because it had few exclusion criteria and a need to accelerate enrollment, including among underrepresented minority groups. Between December 2015 to December 2016, the HealtheRx alerted eligible patients about the opportunity to join the PRR and incentivized patients to “Call or text to complete a 15 minute phone survey as part of a research study. You will be paid $25 for your time and participation” (Figure 1). All people 18 years of age and older, English speaking, and living in the study area who received care at ≥1 of 28 CommunityRx trial sites, with an asthma or chronic obstructive pulmonary disease (COPD) diagnosis, were eligible to receive a HealtheRx with PRR information.

HealtheRx recipients who called in response to the PRR information were administered a brief, structured phone survey that elicited sociodemographic characteristics, interest in joining the PRR, attitudes about participation in health research more generally,³⁴ and willingness to participate in and be contacted for future research opportunities. Survey respondents received a small monetary incentive. After the first 10 surveys, questions were added, including an open-ended question, “When you think about the idea of participating in medical or health research, what are some of the thoughts or concerns that go through your head?” and a question to assess interest in engaging as a patient partner on research teams. All survey participants received a check incentive via mail (increased from $10 to $25 when the previous questions were added). Respondents who expressed interest in joining the PRR were mailed a blank consent form and prepaid, addressed return envelope with their monetary incentive. Additionally, these individuals were randomized, using a random number generator, to receive a $5 bill either (1) with their mailed materials (front-end incentive) or (2) upon the research team’s receipt of a signed PRR consent form (completion bonus).

We present descriptive analysis of close-coded survey questions (frequencies, percents) and of open-ended survey questions, which were open-coded thematically and adjudicated by the coauthors to achieve consensus.

The University of Chicago Institutional Review Board approved recruitment to the PRR in June 2010. At the time of this study, the PRR had enrolled 758 registrants toward a target of 2000. During this study period, usual PRR recruitment strategies included (1) public advertisements like printed fliers requiring the patient to contact the research team and (2) education of eligible patients by a research coordinator during routine clinical care or at the time of enrollment to another study.

RESULTS

Focus groups

Nineteen people participated in 2 focus groups, which met the needs of the larger study. Most participants were women (n = 11), non-Hispanic Black (n = 18), and not employed (n = 15). Participant mean age was 51 years (range, 24-68 years). The question probing participants’ thoughts about getting research opportunity information from a HealtheRx generated the theme of “affirmation”. Nearly all (n = 14) participants reacted positively to seeing research opportunities included on the HealtheRx (eg, “I think it’s wonderful” and “it’s perfect”) and expressions of affirmation (eg, “yes’s” and “mmhms”); no participants raised objections. “Financial compensation” emerged as a subtheme: 8 participants endorsed the value of information specifying compensation (eg, “I love that,” “that makes it even betters [sic],” and “a fair exchange is no robbery”) and none objected. The theme of “share with others” also emerged. Three participants expressed that they might share the information with others: “I think it’s fine, cuz I’ll refer to somebody that does [fit the opportunity]. I’ve got aunts and uncles and cousins that love to smoke and some have asthma and I would tell them.” One participant voiced reluctance, specifically about participating in experimental medication studies: “Well my first thought, anytime I see anything in regards to medical research, is how much medication are they going to expect me to take? You know that is something that is frightening for me.”

E-prescribing research

During this study, 13 437 HealtheRxs with PRR information (8762 for asthma, 3842 for COPD, and 833 for both asthma and COPD) were generated: 41 patients called for the phone survey (Figure 3). Fifteen (37%) respondents reported having asthma only (65% of all HealtheRxs with the PRR information were generated for asthma only), 9 (22%) reported a diagnosis of COPD only (28.5% of all HealtheRxs with the PRR information were for COPD only), and 17 (41.5%) reported both asthma and COPD diagnoses (6.2% of all HealtheRxs with the PRR information were for asthma and COPD). Table 1 summarizes phone survey participant characteristics; 90% identified as non-Hispanic Black.

Figure 3.

Flowchart of participant recruitment and enrollment through the CommunityRx system. PRR: Pulmonary Research Registry.

Table 1.

Characteristics of phone survey participants (N = 41)

Age
24-49 y	10 (24)
50-56 y	10 (24)
57-64 y	10 (24)
65-77 y	11 (27)
Race/ethnicity
Non-Hispanic Black	37 (90)
Non-Hispanic White	4 (10)
Sex
Female	28 (68)
Male	13 (32)
Unknown
Self-reported asthma/COPD diagnosis
Asthma only	15 (37)
COPD only	9 (22)
Both asthma and COPD	17 (42)
Unknown
Education
Elementary or some high school	15 (37)
High school or GED	16 (39)
Some college or associate’s degree	8 (20)
College graduate	2 (5)
Employment status
Not employed	39 (95)
Employed part time	1 (2)
Employed full time	1 (2)

Values are n (%).

COPD: chronic obstructive pulmonary disease.

The majority of participants (n = 31) reported no prior participation in health research and most (n = 25) were not previously aware of such opportunities. For those who reported awareness of current opportunities, 6 learned of opportunities from their clinician, 5 from flyers, 4 from family or friends, and 1 from online. The large majority of participants strongly disagreed or disagreed that it would be difficult to find time to participate in research (n = 37) and that there was not anything for them (n = 38) or their community (n = 39) to gain by participating in research. Nearly one-third of participants (n = 12) endorsed (strongly agreed or agreed) that they would not want to participate in a study that would require taking medication. This sentiment was further elaborated in the open-ended survey item that asked participants to reflect on their thoughts or concerns about research participation (Table 2).

Table 2.

Thoughts and concerns regarding participation in health or medical research among survey participants (phone survey results)

General reflections on e-prescribing research

“I think it's a good thing to know about research opportunities.”

“Some of the thoughts that go through my head is that they're trying to find a cure for whatever it is that they're researching.”

“I see in the clinic they have flyers but my doctors and nurses don't tell me about it.”

“Does my insurance coverage matter to participate in a study?”

“Maybe they'll test me like a lab rat or something.”

Helping others

“The only concerns that I have is that they can find a cure in the world and that after being in the study they might be able to help someone else.”

“Will they find a way to help other people, other patients?”

“That I'm really trying to help someone that has the same condition that I have or has a similar condition.”

“I would be interested to help the community and to help others.”

Medication

“Do I have to take any medication or will it mess with my asthma?”

“What would I be doing and how is the medication going to affect my body?”

“When you go through medical research they give you different medicine and see how you might respond, right? So I would be concerned about the side effects…”

“I think ‘are they trying to help me or are they trying to use me as a guinea pig?’ It depends what kind of medicine it is.”

“I just don't want to take nothing that they're researching - experimental.”

“I'm not sure if I would want to take medicine because I would have to talk to [my other] doctor.”

“I'm not really sure if I would want to take any medication.”

“That they may have me take experimental drugs.”

Personal benefit/improvement

“If it helps me to quit smoking, that's what I want to do.”

“It is going to help me out and improve myself I guess.”

“…kind of excited to see what's going to happen and what it's all about…I’m doing some things on my own and this is different thing to try.”

All 41 participants agreed to be contacted for future research. Of the subset of participants (n = 31) that were asked about their interest to engage as patient partners in research, 26 (84%) expressed interest and agreed to be contacted for such an opportunity.

Figure 3 summarizes enrollment, including by incentive randomization group. All 41 participants who called in to learn more about the PRR expressed interest in joining and were mailed a blank PRR consent form and a prepaid return envelope with their survey compensation check. All but 1 participant’s checks were cashed. Nine participants (8 non-Hispanic Black) returned a signed consent form. During the same period, 4 patients (1 non-Hispanic Black) enrolled in the PRR via usual recruitment methods.

DISCUSSION

EMR systems are near ubiquitous in academic and other medical research settings and are increasingly being used to accelerate clinical research enrollment.^18,35–37 In parallel, with growing adoption of value-based payment strategies in healthcare, EMRs are increasingly being used to monitor, screen for and intervene to address modifiable “upstream” health risk factors by connecting people to community-based resources.^24,38 In this study, residents of a predominantly African American community endorsed the new idea of thinking about the opportunity to participate in research studies as a health-related community resource. Implementation of this idea creatively leverages both the research recruitment and community referral applications of EMRs. We find that it is both feasible and acceptable to promote health research participation opportunities along with other health-promoting community resources in a personalized, auto-generated community e-prescription delivered during a healthcare visit. We also find that the EMR can be meaningfully used to raise awareness about research among people commonly underrepresented in health studies. In our study, three-quarters of 41 people who responded reported no prior history of research participation, 90% identified as non-Hispanic Black. In comparison, usual recruitment practices over the same period yielded 4 respondents, 1 of whom was non-Hispanic Black.

This case study, using a highly pragmatic implementation approach, yields several lessons that can inform future efforts to leverage community resource referral systems for research recruitment. While it is feasible and acceptable to non-Hispanic Black patients to use a community resource referral tool for recruitment, only a small fraction of eligible people responded. In this study, patients received only the auto-generated HealtheRx; no additional coaching or information was provided. High-yield patient recruitment using EMR-driven methods likely requires not only the information about eligibility but some encouragement or explanation from clinicians or staff.³⁹ Furthermore, we learned that information about research should specify whether taking medication^39,40 will be required, any costs will be incurred, and whether participation would potentially benefit others.⁴⁰ Interestingly, the importance of the opportunity to benefit others corroborates the finding in 2 prior CommunityRx studies that nearly half the people who receive a HealtheRx share the information to help others.^24,29

The consent process was also a barrier to enrollment. Every phone survey respondent expressed intent to join the PRR and agreed to receive a consent form. Fewer than a quarter of consents were returned. Explanations for this gap could include social desirability bias, ineffective or burdensome consent forms, and loss in the mail delivery process. Tighter coupling of the research recruitment and consent processes can be effected using EMR workflows.⁴¹ Oral, phone-based consent for this low-risk observational registry study would also likely have yielded higher enrollment.

Cost, including incentives, is an important consideration for any recruitment strategy. Our findings corroborate the feasibility of conducting randomized incentive experimentation in an informatics-based research recruitment study.⁴² Overall low enrollment numbers make it challenging to quantify differences in enrollment based on type of incentive delivery; however, the front-end incentive yielded more participants than the completion bonus.^43,44 Once a health system decides to adopt a community resource referral system,⁴⁵ the cost to add research opportunity information to the informatics tool is marginal and could easily be implemented to complement other recruitment strategies. We have previously presented a quantitative comparison of overall cost of the 2 recruitment strategies, which favors e-prescribing,⁴⁶ but financial modeling beyond the scope of this study is warranted. This case study will inform further experimentation to increase the impact of the intervention and to assess its cost effectiveness.

CONCLUSION

Implementation of an EMR-integrated, community resource referral tool promotes enrollment of eligible non-Hispanic Black research participants; however, enrollment was low overall. More research is needed to understand the value of patient-facing EMR-integrated solutions, including community resource referral solutions, to recruit minority populations for research participation.

FUNDING

This work was supported by U.S. Centers for Medicare & Medicaid Services Grant No. 1C1CMS330997 (to STL), National Institute on Aging Grant No. 1R01AG047869 (to STL), and a pilot grant (to STL) from the University of Chicago Institute for Translational Medicine (ITM) Grant No. UL1TR000430 (to JS); ITM is a member of the National Institutes of Health Clinical and Translational Science Awards consortium. The research described is registered on ClinicalTrials.gov (NCT02653066). The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the Department of Health and Human Services or any of its agencies. The contents of this article describe research that was conducted by the authors.

AUTHOR CONTRIBUTORS

All those designated as authors meet all 4 criteria of authorship recommended by the International Committee of Medical Journal Editors.

ACKNOWLEDGEMENTS

The authors would like to acknowledge Ms Chenab Navalkha and Ms Kelsey Paradise for their assistance in preparing this manuscript for publication as well as the community members and patients who participated in this research.

CONFLICT OF INTEREST STATEMENT

CommunityRx was created with support, in part, from the U.S. Center for Medicare and Medicaid Innovation Health Care Innovation Award, which expected that a sustainable business model would be implemented to continue the innovation after the completion of funding. To this end, STL is the founder and owner of NowPow, LLC. GF, who was an employee at the University of Chicago Lindau Laboratory, Department of Obstetrics and Gynecology, at the time this research was conducted, is currently an employee at NowPow, LLC. Neither the University of Chicago nor University of Chicago Medicine is endorsing or promoting any NowPow entity or its business, products, or services. The other authors have no competing interests to declare.