Community-based recruitment engages multiple stakeholders in a collaborative approach.
Increased competition to bring the next blockbuster drug to the market has created an urgency to improve the efficiency of every facet of drug development. Subject recruitment, or lack thereof, has been considered to be one of the major factors delaying study completion. Recruiting patients in an efficient manner has become critical in clinical trials. Surprisingly, although "pharmaceutical companies invest heavily in marketing approved drugs, they often do not employ that same market research and marketing expertise in communicating the value of clinical trials to patients."1 This lack of support is unfortunate since investing in outreach to potential subjects has been shown to be advantageous.
Enrolling the necessary number of patients on an expedited timeline is necessary in conducting a successful trial. An underpowered study may lead to faulty or biased conclusions.1 Since regulatory agencies are requiring a greater number of efficacy and safety studies prior to a drug's release, the demand for subjects is also increasing, but is unlikely to be filled with the traditional approach of "physician-based recruitment." Community-based recruitment, either separately, or in combination with the traditional recruitment methods, may help fill this dearth of subjects. Community-based recruitment engages the multiple stakeholders in a collaborative approach by involving all partners in the research process and utilizing the unique strength that each brings.2 If patients are targeted through these potentially successful methods, patients can receive early access to promising treatments. This review aims to examine and analyze the effectiveness of various community-based recruitment strategies that were implemented in Type 2 diabetes studies that took place in the United States between 1975 and 2012.
Non-randomized controlled trials and a non-controlled trial utilizing various community-based recruitment methodologies to target diabetic adults (≥18 years of age) were included in this review. Excluded were studies that did not specifically address recruitment, enrolled children, and/or were observational studies. In addition, case reports, review articles, and duplicate reports were excluded.
A search was conducted utilizing PubMed, for articles published between 2002 and 2012 that met the inclusion/exclusion criteria. But due to the paucity of studies, the search was expanded to include publications as old as 1995. Search terms included recruitment, diabetes, Type 2, diabetic, community-based, strategies, methodologies, and adults. Additionally, bibliographies of articles retrieved were hand-searched for additional studies meeting the inclusion criteria. Only those studies that met the above criteria and were published in English were reviewed.
The literature search was performed according to the guidelines provided by the Thames Valley Literature Review Standards Group.3 The guidelines include steps on how to plan a literature search, identify key sources of information, guidance on carrying out an effective literature search, and documenting the research process.3 Grading (positive, negative, or neutral) according to quality criteria checklists was completed for each included study to assess the quality constructs as is recommended in the March 2002 Agency for Healthcare Research and Quality report on steps to rate the strength of scientific evidence.4 All studies included in this review received either a positive or neutral rating.
The best community-based recruitment strategy was abstracted from all studies that met the inclusion criteria, even when there were other outcomes presented in the results. When the information was provided, the percentage of patients enrolled with the strategy was standardized (number of patients enrolled via the strategy of interest divided by the total number of patients enrolled) to allow for comparison between the studies.
Community-based recruitment was defined as a general term describing those strategies that centered on a specific community and its community members. Strategies included focus on recruiting patients via community events or locations (such as churches or outpatient clinics) and utilization of various community members such as municipal health workers. Not all strategies were specifically mentioned in the articles, but community outreach may include setting up a booth at a health fair, attending patient support meetings, gaining support of a well-known leader in the community, and enlisting individuals who have volunteered for studies in the past.5 Other community-based strategies include targeted approaches (such as ads in local newspapers). The main focus of these strategies had to be a partner-led approach or where community resources were used as the basis for increasing interest among the desired patient population.
Physician-based recruitment was defined as those methods where the physician or site staff was the main facilitator targeting the patients. Furthermore, they may have recruited through their own patient database and printed their own recruitment flyers. The physician or site staff would be involved with initiating contact with the patient to provide information on the study and its benefits. In general, the site staff would contact the patient via phone or e-mail, or the physician would introduce the study to the patient during their office visit.
The number of studies screened, assessed, and included in this review are shown in Figure 1, and the design elements, methodology, and results are featured in Table 1.
A total of eight studies with 1,343 subjects were included in this review, seven of which were non-randomized controlled studies and one that was a non-controlled study. In seven of the studies, a subgroup analysis was also studied. More specifically, African-Americans (n=4), Mexican-Americans (n=1), only females (n=2), or a combination of different minority groups (n=2) were represented. Since diabetes is not restricted to one population, the multi-cultural nature of these studies was advantageous and would allow for some generalization to the population. In addition, they all used diverse recruitment strategies allowing for comparisons between individual community-based methods, and community-based methods versus physician-based recruitment strategies.
The rate at which each intervention recruited subjects was not analyzed in this review because the studies did not employ the recruitment strategies at the same time or specify how long each was implemented for and did not always note the number of subjects recruited for each method. Results on the most successful recruitment method is highlighted in Figure 2.
Outcome of community-based strategies. All eight studies emphasized the importance of community-based approaches to recruitment. The partner-led recruitment method entailed "identifying, reaching out, and motivating a vulnerable population into research.7 Four of these studies yielded the greatest number of patients from a specific community-based strategy (not in combination with other community-based strategies). However, one study utilized only a community health center system (a billing system and registry which was accessed to identify eligible participants in conjunction with a facilitator who worked within the community health center) so no comparisons could be made. The three other studies had the largest percentage of patients recruited via physician-based recruitment; more specifically, through healthcare providers, chart review, and the patient databases. Despite this, for all three studies, the second most successful method was community-based.
Community-based strategies proved to be essential in the recruitment process in three studies demonstrating between 53% and 68% accrual via this method.6, 7, 10 The community members, whether they were healthcare workers, community advocates, or community coordinators, were seen as a liaison between the community and the site. These members identified patients in a variety of settings including visiting churches and attending community events.
Printed materials and electronic media were also seen as top strategies to use (n=2). Falling in this category were newspapers and public service announcements at various locations within the community. Finally, community clinics were seen as a source of a large pool of potential subjects (n=2). In addition, one study noted that healthcare providers are a key resource for recruiting subjects but mentioned that relationships with faith-based organizations were also essential in establishing trust with the community.8
Community health workers. Partnering with community health workers was seen as one of the best ways to achieving the recruitment goal. Out of those studies that utilized more than one method of recruitment (n=7), three noted them as the most successful. In one study where it was utilized, it was able to attain almost 53% of the patients through this method.6 The community health workers' longstanding trusting relationships and experience helped break down barriers "to participation and engendered trust in the project via their dual role as research partners and advocates."7 Most of these community workers were already intricately involved in their community and were seen as trusted leaders. As stated by one author: "Researchers and diabetes educators should consider working with a community advisory board and with the healthcare providers known and trusted by people in the community."8 One of the biggest advantages was that they were able to work with existing social networks and various community organization members. Subjects approached were more apt to trust these individuals and express interest in the study.
Other community-based recruitment methods. There were several other community-based recruitment methods utilized for these studies. Providing disease-specific education as well as study details at community fairs and church events were the methods chosen in six studies. Methods falling under this category were giving presentations, having information tables, and distributing pamphlets as a way to spark conversation on the disease and/or study. While not being the most successful method, this did raise awareness of the disease process and the nature of clinical trials in locations most comfortable to the patients.
Searching through billing systems and registries at community health clinics was another recruitment method frequently employed (n=4). This method also yielded information regarding eligibility for participation.11 Attending support groups also helped expound the value of volunteering for a clinical trial. Out of the studies that utilized more than one recruitment method (n=7), one report showed the community clinic approach as the best community-based method.13
Utilizing community-based recruitment strategies is a fundamental technique of the recruitment process. In the review of the literature, all of the studies noted the importance of these types of strategies even though they did not always accrue the most patients. The findings in the articles were generally consistent and worked well for the analysis as the interventions could be applied to the general diabetic population.
In general, the community-based recruitment methods included utilizing community members and community events as a platform. The partner-led approach was seen as a way of having key community members serve as the facilitator with the population of interest by fostering trust. The community needs to be able to trust the information they are receiving, and in consequence, needs to trust those who are providing it.
One of the most important lessons discovered as a result of all these studies is the importance of collaboration. It is vital that the physician collaborates with the community to reach the potential patient pool. "Becoming visible in the community and having members from the community on the study team helps to increase the participation in research clinical trials and programs designed to promote care of Type 2 diabetes."8 As mentioned in one study, strategies that led to a productive collaborative research environment were composed of established relationships, fostered ownership of the research process, and built trust, which ultimately resulted in the successful recruitment, randomization, and retention of the study's participants.11 As is the case with many clinical studies, the fundamental key to success comes from establishing a strong working relationship between the research team and the referring agency.14 These agencies have their own set of patients from which they can draw potential subjects, whether it be hospitals, clinics, health maintenance organizations, etc. Their strong organizational infrastructure plays a central role in subject recruitment efforts since they are usually familiar with subject recruitment advertisement methods, in addition to being affiliated with other organizations such as university programs or health centers.
Another notable finding is that in almost all of the studies, more women were recruited than men.7, 10, 11, 12 (Two of the studies only recruited females).8, 13 Several explanations were given to explain this result such as less men received care from the participating health center, or in general, they lacked interest in the screening process.11 However, it should be noted that "the lower pool of potential male participants compared to females cannot be attributed to the overall prevalence of diabetes, which is similar by gender according to population-based studies."11 Also notable is that a number of participants from some of the studies came from a lower socioeconomic background.7, 11, 13 This is most likely due to the fact that they could potentially receive discounted or free medical care.
The present review has several limitations. First, some studies provided incentives while others did not. This could have affected the rate or number of patients that were recruited in the study. Previous studies have shown that providing incentives to subjects has led to the attainment of a larger number of participants or completing recruitment in a shorter amount of time. This leads to the second major limitation—standardization of each study's recruitment time period and duration of the recruitment methods was not possible. Since many of the studies did not note the full length of the recruitment period and/or when each recruitment method was executed and discontinued, there was an inability to analyze how long each recruitment method was implemented for. Another limitation is that each study focused on a different subgroup of patients. Those studies that focused on a smaller subgroup of patients found it more difficult to recruit patients (example: African-Americans living in a rural area compared to a study where race and population size are not specified) and introduced additional challenges. For example, African-Americans are less likely to participate in clinical trials and less likely to consider themselves informed about clinical trial opportunities.15
In the review of the literature, the importance of community-based recruitment was confirmed along with the advantages it can bring. In addition, these methodologies can prove to be effective by preventing delays in the research and minimizing costs. However, existing studies lack uniform study design and do not focus on standard populations, therefore, not allowing for more specific comparisons between different strategies among studies. There is a need for studies with clarity on their recruitment design along with detailed data on the various methods implemented. Future studies should also consider the effectiveness of community-based recruitment.
Saliha Akhtar* is a PhD canidate, e-mail: [email protected], and Barbara Gladson, PhD, MS, is Director, Biopharma Educational Initiative both at UMDMJ, Newark, NJ.
*To whom all correspondence should be addressed.
1. G. Frank, "Current Challenges in Clinical Trial Patient Recruitment and Enrollment," SoCRA Source, 2, 30–38 (2004).
2. C. R. Horowitz, M. Robinson, and S. Seifer, "Community-Based Participatory Research From the Margin to the Mainstream: Are Researchers Prepared?" Circulation, 119 (19) 2633–2642 (2009).
3. Thames Valley Health Libraries Network, "The Literature Searching Process Protocol for Researchers," (2006), http://www.ieph.org/hase/Downloads/Lit_Search_Protocol_Researchers.pdf.
4. American Dietetic Association, "Evidence Analysis Manual: Steps in the ADA Evidence Analysis Process," (American Dietetic Association, Chicago, 2010).
5. B. Oscherwitz, D. Anderson, and H. West, "Attitudinal Clustering of Research Participants to Optimize Patient Recruitment and Retention Programs: A Review of the Lliterature," Journal of Clinical Research Best Practices, 2 (6) 1-9 (2006).
6. M. A. Martin, et al., "Recruitment of Mexican American Adults for an Intensive Diabetes Intervention Trial," Ethn Dis, 21 (1) 7-12 (2011).
7. C. R. Horowitz, et al., "Effective Recruitment of Minority Populations Through Community-Led Strategies," Am J Prev Med, 37 (6S1) S195–S200 (2009).
8. D. Burns, et al., "Effective Recruitment and Retention Strategies for Older Members of Rural Minorities," Diabetes Educ, 34 (6) 1045–1052 (2008).
9. K. Wisdom, et al., "Recruitment of African Americans with type 2 diabetes to a Randomized Controlled Trial Using Three Sources," Ethn Health, 7 (4) 267–278 (2002).
10. R. M. Davis, et al., "A Collaborative Approach to the Recruitment and Retention of Minority Patients With Diabetes in Rural Community Health Centers," Contemp Clin Trials, 30 (1) 63–70 (2009).
11. L. A. Anderson, et al., "Recruiting From the Community: Lessons Learned From the Diabetes Care for Older Adults Project," Gerontologist, 35 (3) 395–401 (1995).
12. S. J. Andreae, et al., "Recruitment of a Rural, Southern, Predominantly African-American Population Into a Diabetes Self-Management Trial," Contemp Clin Trials, 33 (3) 499-506. (2012).
13. K. Newlin, et al.,"Recruitment of Black Women With type 2 Diabetes Into Self-Management Intervention Trial," Ethn Dis, 16 (4) 956-962 (2006).
14. P. G. Butterfield, et al., "Overcoming Subject Recruitment Challenges: Strategies for Successful Collaboration With Novice Research Agencies," Applied Nursing Research, 16 (1) 46–52 (2003).
15. L. A. Lynn, "AIDS Clinical Trials: Is There Access for All?" J Gen Intern Med, 12 (3) 198–199 (1997).