- Email: [email protected]
Overcoming subject recruitment challenges: Strategies for successful collaboration with novice research agencies
Clinical Methods Overcoming Subject Recruitment Challenges: Strategies for Successful Collaboration With Novice Research Agencies Patricia G. Butterfield, Shawna M. Yates, Bonnie Rogers, and Julia M. Healow
In most studies, the investigator reaches out to locate potential research subjects using direct strategies such as targeted radio advertising, recruitment posters, and newspaper advertisements). However, other studies may depend on indirect methods of locating potential subjects and work through intermediary contacts in clinics and hospital outpatient departments. Some agency personnel may not have had prior experience with clinical research protocols and may be unfamiliar with screening potential subjects; these personnel are likely to be employed in agencies unaffiliated with academic health sciences centers. In cases in which agency staff members are new to clinical research, special attention is required to keep agency personnel updated about the status of the research and subject recruitment efforts. This article provides an overview of practical tips designed to engage and sustain the interest of novice research agencies in subject recruitment. The article concludes with a case overview of recruitment issues that occurred during a clinical trial addressing occupational low back pain. Copyright 2003, Elsevier Science (USA). All rights reserved.
C
LINICAL RESEARCHERS frequently depend on clinics or other health agencies to help identify potential subjects. The reason is obvious; agencies that are providing services to the population of interest have already established a variety of methods to contact the people whom they serve. In many situations, the agencies main-
Patricia Butterfield, PhD, RN, Professor, College of Nursing, Montana State University, Bozeman, MT; Shawna M. Yates, MS, Western University of Health Sciences, College of Osteopathic Medicine of the Pacific, Pomona, CA; Bonnie Rogers, DrPH, RN, FAAN, Director, Occupational Health Nursing, University of North Carolina at Chapel Hill, Chapel Hill, NC; Julia M. Healow, BS, Research Associate, College of Nursing, Montana State University, Bozeman, MT. Supported by the National Institute of Nursing Research (Project #R29 NR04450-05) and the Center for Research on Occupational and Environmental Toxicology at Oregon Health & Science University. Address reprint requests to Patricia G. Butterfield, PhD, RN, College of Nursing, 207 Sherrick Hall, Montana State University, Bozeman, MT 59717. Copyright 2003, Elsevier Science (USA). All rights reserved. 0897-1897/03/1601-0007$35.00/0 doi:10.1053/apnr.2003.50004 46
tain comprehensive files on past and present clients, including their phone numbers, home addresses, and employer information. However, in most cases, identifying information on potential research subjects cannot be released to researchers outside the organization without the consent of those persons. Before they can provide a researcher with the name of a potential subject, agency personnel are required to ask each person if they are willing to have his/her name and phone number released to the research team. Because of this role, clinics and other agencies play the central role in linking the researcher with persons who may be interested in becoming research subjects. Because obtaining an adequate number of subjects is predicated on an adequate number of referrals, the agency performs an important role, albeit a defacto one, in ensuring the success of the research. The purpose of this article is to outline practical tips on working with agencies that are new to clinical research. An overview of methodologic challenges in subject recruitment is provided and is followed by a case review addressing a longitudinal study of occupational low back pain. In some Applied Nursing Research, Vol. 16, No. 1 ( February), 2003: pp 46-52
OVERCOMING SUBJECT RECRUITMENT
situations, it may be the first time that agency personnel have participated in a research project. The preponderance of nursing literature addressing subject recruitment emphasizes direct subject recruitment activities such as advertisements, telephone calls, or approaching persons during a clinic visit. In contrast, this article provides a summary of how to develop and implement a subject recruitment plan in partnership with agencies that are novices to research. The term research novice is not used in a disparaging manner; in fact, many such agencies are highly sophisticated in the coordination or delivery of health care. However, either by design, geography, target population, or service mix, participation in research has not been a central role in the agency’s mission. A major theme throughout the article is the importance of listening to agency personnel and providing them with materials (e.g., referral forms and screening and criteria handouts) that meet their needs for efficiency and clarity. 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.
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.
DIRECT VERSUS INDIRECT ACCESS TO POTENTIAL SUBJECTS
In examining issues addressing subject recruitment, Walson (1999) noted that it is helpful to break down the recruitment process into a series of steps so that potential problems with each step can be systematically analyzed. Steps proposed by the author include (1) identifying the eligible population, (2) explaining the study, (3) obtaining true informed consent, (4) maintaining ethical standards, (5) recruiting an adequate and representative sample, (6) retaining subjects until study completion, and (7) minimizing the risk/benefit ratio. Step one, identifying persons in the eligible population, can oc-
47
cur directly through contacts with members of the research team, or indirectly, through agency personnel. When direct access to potential subjects occurs, members of the research team have the opportunity to approach potential subjects and explain the study to them. However, in situations in which potential subjects’ names cannot be released outside an agency without their direct consent, potential subjects are first contacted by agency personnel and asked if they are willing to have their name released to the research team. Eaves (1999) referred to agency personnel as intermediaries and noted that those persons frequently included discharge planners, medical social workers, and casemanagement specialists. Because most potential subjects need to have a brief overview of a study before making a decision about releasing their name, it is important to provide agency intermediaries with clear scripted information about what to say about the study. A clear differentiation between providing a potential subject with background information and obtaining informed consent for study participation needs to be made; written and verbal consent to participate can be given only after a subject has been briefed by the research team and provided with an explanation of the risks and benefits of participation.
Because most potential subjects need to have a brief overview of a study before making a decision about releasing their name, it is important to provide agency intermediaries with clear scripted information about what to say about the study.
FACILITATING THE RECRUITMENT OF SUBJECTS FROM UNDERSERVED AND/OR DISENFRANCHISED GROUPS
Prior studies and methodologic articles have provided evidence that some population subgroups have been under-represented in clinical research and pose special challenges to investigators seeking balanced and representative samples in their research. Underrepresented groups may include inner-city residents (Evans, Mejı´a-Maya, Zayas,
48
Boothroyd, & Rodriquez, 2001), ethnic and racial minorities (Eaves, 1999; Hodge, Weinmann, & Roubideaux, 2000; Berman, Grosser, & Gritz, 1998), rural citizens (Ulrich, Browning, Gaffney, Schoter, & Chesnut, 1997; Shreffler, 1999), persons with physical disabilities (Bach & McDaniel, 1995), persons with low incomes (Berman, Grosser, & Gritz, 1998); the elderly (Leader & Neuwirth, 1978), and persons with traumatic memories of a health-related event (e.g., perinatal loss) (Coˆ te´ -Arsenault & Morrison-Beedy, 1999). For a clinical trial that was conducted in the United Arab Emerates, Glasgow and colleagues (1996) reviewed recruitment and dropout rates at different study points and concluded that clinical trials in developing nations may take up to twice as long as originally anticipated. The authors suggested that researchers working in isolated areas need to devote additional resources to sustain subject recruitment and retention throughout the data collection process. On a related issue, Walson (1999) noted that treatment for any condition that may directly or indirectly lead to denial of insurance coverage for a condition is a strong disincentive to research participation. When the investigator has the ability to directly access potential subjects through advertising, Internet groups, clinical contacts, or support groups, recruitment activities can be framed in a manner that speaks directly to the target population. Investigators use a variety of verbal and visual strategies to catch the attention of potential subjects. Shreffler (1999) discussed the importance of developing culturally sensitive research materials in the design of studies that are conducted in rural and frontier communities. The author suggested that selecting a graphic or logo that is emblematic of a specific community may help increase research participation; examples of such graphics for rural communities include crop motifs, a farm scene, or the town logo. In an article addressing subject recruitment of American Indians, Hodge and colleagues (2000) emphasized that recognizing the distinctiveness of each tribal nation was key to successful recruitment for clinical trial research. In a study involving rural residents with osteoporosis, Ulrich and colleagues (1997) used a mobile clinic van to travel to Montana and Wyoming communities for osteoporosis testing. The authors attributed the success of their recruitment efforts to use of the mobile clinic and suggested this
BUTTERFIELD ET AL
approach as a method of overcoming low research participation in rural communities. When the investigator is working to access potential clients through agency intermediaries, efforts need to focus on enlisting the ongoing attention and support of agency personnel. When appropriate, financial incentives may be offered, although ethical guidelines in the use of “finder’s fees” for subjects need to be implemented before any contact with agency personnel (Walson, 1999). Other incentives may include books for the agency library (e.g., new nursing textbooks), desk items such as pens or coffee mugs with a project or university logo, or letters to employees’ supervisors that recognize the employee for their contributions to identifying research subjects.
When the investigator is working to access potential clients through agency intermediaries, efforts need to focus on enlisting the ongoing attention and support of agency personnel.
UNDERSTANDING THE AGENCY PERSPECTIVE
Because of their economy of scale and strong organizational infrastructure, large hospitals, clinics, and health maintenance organizations have historically played a central role in subject recruitment efforts for descriptive studies as well as clinical trials. Employees in such agencies who attended university programs affiliated with academic health sciences centers are generally familiar with subject recruitment advertisements that are frequently posted in hospital elevators and waiting rooms. Frequently, they have participated in clinical trials or assisted with other research-related activities. On the other hand, smaller agencies and those located in rural communities are less likely to have employees that have worked in a researchintensive environment. To those employees, the research role is new, and they may not be familiar with research journals or internal publications that showcase research results. Engaging personnel in novice research agencies requires ongoing attention to employees’ workloads and interest in research. Because personnel
OVERCOMING SUBJECT RECRUITMENT
may be extremely busy, it is important not to assume that more information about research is better than less. In a discussion of obstacles to the development of a rural health promotion program, Wagner (1994) discussed the importance of building collaborative relationships among multiple community agencies. The author identified several barriers to program participation among agency staff members; these barriers included (1) skepticism about the potential success of the program, (2) no prior experience in program development, and (3) a lack of personnel with available time to take on a new initiative.
Engaging personnel in novice research agencies requires ongoing attention to employees’ workloads and interest in research.
The plan for subject recruitment should reflect elements of the organizational culture from both the clinical agency as well as the agency sponsoring the research. It is imperative to obtain the support of agency supervisors, who not only can provide entree into an agency but also can provide valuable insights into how the agency functions. Although research publications are valued in academic circles, agency personnel may or may not want to be provided with background publications and reports that detail each phase of the research process. They may, in contrast, want information that is distilled into a concise handout, similar to an executive summary for a scientific report. It is critical to ask agency personnel what they want and need to assist with identifying potential subjects for study participation. The importance of brief but regular contacts with agency personnel to troubleshoot problems or clarify any questions about subjects’ participation cannot be overestimated. Contact should be made using the preferences of agency personnel; some groups may want to have regular faceto-face meetings, whereas others may prefer a regular e-mail or telephone update about the progress of the research.
49
The plan for subject recruitment should reflect elements of the organizational culture from both the clinical agency as well as the agency sponsoring the research.
CASE EXAMPLE: OCCUPATIONAL BACK PAIN IN SMALL INDUSTRIES
The goal of the Reducing Occupational Disability in Rural Workers Study is to systematically examine feasibility issues in delivering a clinical intervention on behalf of workers with an occupational low back injury. Because the vast majority of occupational health studies have recruited workers from large industries (e.g., ⬎1,000 employees), this research specifically targets workers employed in smaller companies. Such companies are unlikely to have occupational health professionals onsite or have a risk reduction plan for the company; in addition, smaller companies are less likely to develop a comprehensive return-to-work plan for an injured worker. Previous research by Butterfield and colleagues (1998, 1999) has provided evidence that back-injured workers employed in small companies (100 or ⬍ workers) have significantly more lost workdays than those employed in larger companies (mean small companies ⫽ 56 days paid absenteeism, mean larger companies ⫽ 32 days, t test ⫽ 2.31, P ⫽ .02). Workers in the current study are randomly assigned to either standard care (i.e., control group) or standard care plus four home visits focusing on back injury recovery and return to work (i.e., treatment group). Because potential workers for the study are referred from workers’ compensation carriers in Oregon and Montana, claims adjusters in these two organizations had been involved in earmarking files on new claims that may meet the study inclusion criteria. Claims adjusters did not play any role in screening potential workers or enrolling them in the study, but they did need to call workers and obtain permission to have workers’ names and phone numbers released to members of the research team. Because back claims in the target areas occurred on an intermittent basis, referrals of potential subjects required the ongoing attention of claims adjusters, many of whom had never been involved in clinical research before. Study personnel
50
BUTTERFIELD ET AL
Figure 1. Reducing occupational disability in rural workers: Montana worker’s comp referral flow sheet.
made site visits to the insurance carrier in Montana before the beginning of the subject enrollment period; however, after several months, referrals of potential subjects decreased significantly. In assessing the rea-
sons for the problem, the research team noted several organizational factors (within either the university or the insurance organizations) that may have contributed to the decline, including the following:
OVERCOMING SUBJECT RECRUITMENT
51
1. Low numbers (i.e., low incidence) of claims for low back injuries in the participating rural communities, leading to claims adjusters having to relearn the process with each new claim. 2. Seasonal changes in claims for low back injuries, with more injuries occurring in the summer. 3. A corporate reorganization at the insurance agency during the data collection period, leading to a temporary changes in departmental assignments and daily work tasks of staff. 4. Implementation of a new and more sophisticated computer system within the insurance agency, leading to new training needs for the claims adjuster staff. 5. Normal variability in the market share of different workers’ compensation carriers, leading to changes in the number and type of companies (and workers) served by the carrier. 6. Internal staff changes at the insurance agency (e.g., employees transferring into a new department). 7. Annual turnover in university research associates (e.g., graduate students) working on the project. In response to the decrease in referrals, the principal investigator and research staff worked to reduce obstacles for the claims adjuster to screen for study referrals. In addition, the occupational health consultant and a managed-care administrator were contacted to make suggestions that may be unique to the workers’ compensation system. A major theme involved the research team’s perception that the busy nature of the claims adjusters’ day kept them from thinking about the study on a regular basis. In addition, the claims adjusters may not have seen how findings from the study could lead to improvements in care for low back injuries or occupational injuries in general. After discussion with the research team and workers’ compensation personnel, the following actions were implemented. 1. The research team met with the chief executive officer of the company and the Board of Directors to update them on the status of the research and thank them for sharing in a strong collaborative effort. 2. Several books on health and back care were
purchased and given to the claims adjusters to show the appreciation of the study team for ongoing efforts of the claims adjusters. 3. Information about what to do with a new claim was simplified and reprinted on a one-page flow sheet and redistributed to the claim adjusters. Information from this flow sheet is listed in Figure 1. 4. Weekly e-mails were sent asking agency personnel to review their files for potential subjects; in addition, a quick “thank you” mail was sent to the claims adjuster after a referral was received. 5. Revised instructions to claims adjusters provided them with a script to use in speaking with potential subjects. From these efforts, referrals of potential subjects increased over the course of the study. The frequency of back injury workers’ compensation claims in the target areas is generally low, and referrals continue to come on an intermittent basis; in addition, a number of temporal patterns with referrals (e.g., more injuries in the summer) have been noted. Because of these factors, subject recruitment continues to be a daily challenge; however, organizational barriers have been removed with a commensurate increase in subject yield. The research team has a better appreciation of the time obligations of claims adjusters and works to make the referral process as clear as possible. Finally, there is greater attention to the shared vision of the research team and workers’ compensation carriers, which is to promote the fullest, healthiest, and most timely recovery for all injured workers. This goal encompasses informing workers about the availability of a clinical research project as well as understanding the complimentary roles that researchers and agency personnel play in the subject recruitment process. Through their daily efforts on behalf of quality care, agency personnel provide the critical link between applied clinical research and individualized client care. ACKNOWLEDGMENTS The authors would like to thank Saif Corporation, Salem, OR, and the Montana State Fund, Helena, MT, for their contributions to the success of this research.
REFERENCES Bach, C.A., & McDaniel, R.W. (1995). Techniques for conducting research with quadriplegic adults residing in the community. Qualitative Health Research 5(2), 250-255. Berman, B.A., Grosser, S.C., & Gritz, E.R. (1998). Recruit-
ment to a school-based adult smoking-cessation program: Do gender and race/ethnicity make a difference? Journal of Cancer Education, 13(4), 220-225. Butterfield, P.G., Spencer, P.S., & Yates, S. (1999). Back-pain
52
related absenteeism: Differences between small and large companies. Journal of Agricultural Safety and Health, 5(2), 215-225. Butterfield, P.G., Spencer, P.S., Redmond, N., Feldstein, A., & Perrin, N. (1998). Low back pain: Predictors of absenteeism, residual symptoms, functional impairment, and medical costs in Oregon workers’ compensation recipients. American Journal of Industrial Medicine, 34, 559-567. Coˆ te´ -Arsenault, D., & Morrison-Beedy, D. (1999). Practical advice for planning and conducting focus groups. Nursing Research, 48(5), 280-283. Eaves, Y.D. (1999). Family recruitment issues and strategies: Caregiving in rural African Americans. Nursing Research, 48(3), 183-187. Evans, M.E. Mejı´a-Maya, L.J., Zayas, L.H., Boothroyd, R.A., & Rodriquez, O. (2001). Conducting research in culturally-diverse inner-city neighborhoods: Some lessons learned. Journal of Transcultural Nursing, 12(1), 6-14. Glasgow, N.J., Murdoch, J.C., Baynouna, L., Al Sabosy, M.M.A., Al Jabiri, O.N.K., & Harron, D.W.G. (1996). Doing randomized controlled trials in a developing country: Some practical realities. Family Practice, 13(1), 98-103.
BUTTERFIELD ET AL
Hodge, F.S., Weinmann, S., & Roubideaux, Y. (2000). Recruitment of American Indians and Alaska Natives into clinical trials. Annals of Epidemiology, 10, S41-48 (suppl). Leader, M.A., & Neuwirth, E. (1978). Clinical research and the noninstitutional elderly: A model for subject recruitment. Journal of the American Geriatrics Society, 26(1), 27-31. Shreffler, M.J. (1999). Culturally sensitive research methods of surveying rural/frontier residents. Western Journal of Nursing Research, 21(3), 426-435. Urich, U., Browning, M., Gaffney, E.V., Schoter, K.H., & Chesnut, C.H. (1997). Implementation of an osteoporosis research program with a mobile dual-energy X-ray absorptiometry unit: The Montana/Wyoming experience. Osteoporosis International, 7(6), 583-8. Wagner, D.I. (1994). Health promotion cooperatives: An alternative delivery system for rural health promotion programs. Journal of Health Education, 25(2), 77-81. Walson, P.D. (1999). Patient recruitment: U.S. perspective. Pediatrics, 104(3), 619-622.
In most studies, the investigator reaches out to locate potential research subjects using direct strategies such as targeted radio advertising, recruitment posters, and newspaper advertisements). However, other studies may depend on indirect methods of locating potential subjects and work through intermediary contacts in clinics and hospital outpatient departments. Some agency personnel may not have had prior experience with clinical research protocols and may be unfamiliar with screening potential subjects; these personnel are likely to be employed in agencies unaffiliated with academic health sciences centers. In cases in which agency staff members are new to clinical research, special attention is required to keep agency personnel updated about the status of the research and subject recruitment efforts. This article provides an overview of practical tips designed to engage and sustain the interest of novice research agencies in subject recruitment. The article concludes with a case overview of recruitment issues that occurred during a clinical trial addressing occupational low back pain. Copyright 2003, Elsevier Science (USA). All rights reserved.
C
LINICAL RESEARCHERS frequently depend on clinics or other health agencies to help identify potential subjects. The reason is obvious; agencies that are providing services to the population of interest have already established a variety of methods to contact the people whom they serve. In many situations, the agencies main-
Patricia Butterfield, PhD, RN, Professor, College of Nursing, Montana State University, Bozeman, MT; Shawna M. Yates, MS, Western University of Health Sciences, College of Osteopathic Medicine of the Pacific, Pomona, CA; Bonnie Rogers, DrPH, RN, FAAN, Director, Occupational Health Nursing, University of North Carolina at Chapel Hill, Chapel Hill, NC; Julia M. Healow, BS, Research Associate, College of Nursing, Montana State University, Bozeman, MT. Supported by the National Institute of Nursing Research (Project #R29 NR04450-05) and the Center for Research on Occupational and Environmental Toxicology at Oregon Health & Science University. Address reprint requests to Patricia G. Butterfield, PhD, RN, College of Nursing, 207 Sherrick Hall, Montana State University, Bozeman, MT 59717. Copyright 2003, Elsevier Science (USA). All rights reserved. 0897-1897/03/1601-0007$35.00/0 doi:10.1053/apnr.2003.50004 46
tain comprehensive files on past and present clients, including their phone numbers, home addresses, and employer information. However, in most cases, identifying information on potential research subjects cannot be released to researchers outside the organization without the consent of those persons. Before they can provide a researcher with the name of a potential subject, agency personnel are required to ask each person if they are willing to have his/her name and phone number released to the research team. Because of this role, clinics and other agencies play the central role in linking the researcher with persons who may be interested in becoming research subjects. Because obtaining an adequate number of subjects is predicated on an adequate number of referrals, the agency performs an important role, albeit a defacto one, in ensuring the success of the research. The purpose of this article is to outline practical tips on working with agencies that are new to clinical research. An overview of methodologic challenges in subject recruitment is provided and is followed by a case review addressing a longitudinal study of occupational low back pain. In some Applied Nursing Research, Vol. 16, No. 1 ( February), 2003: pp 46-52
OVERCOMING SUBJECT RECRUITMENT
situations, it may be the first time that agency personnel have participated in a research project. The preponderance of nursing literature addressing subject recruitment emphasizes direct subject recruitment activities such as advertisements, telephone calls, or approaching persons during a clinic visit. In contrast, this article provides a summary of how to develop and implement a subject recruitment plan in partnership with agencies that are novices to research. The term research novice is not used in a disparaging manner; in fact, many such agencies are highly sophisticated in the coordination or delivery of health care. However, either by design, geography, target population, or service mix, participation in research has not been a central role in the agency’s mission. A major theme throughout the article is the importance of listening to agency personnel and providing them with materials (e.g., referral forms and screening and criteria handouts) that meet their needs for efficiency and clarity. 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.
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.
DIRECT VERSUS INDIRECT ACCESS TO POTENTIAL SUBJECTS
In examining issues addressing subject recruitment, Walson (1999) noted that it is helpful to break down the recruitment process into a series of steps so that potential problems with each step can be systematically analyzed. Steps proposed by the author include (1) identifying the eligible population, (2) explaining the study, (3) obtaining true informed consent, (4) maintaining ethical standards, (5) recruiting an adequate and representative sample, (6) retaining subjects until study completion, and (7) minimizing the risk/benefit ratio. Step one, identifying persons in the eligible population, can oc-
47
cur directly through contacts with members of the research team, or indirectly, through agency personnel. When direct access to potential subjects occurs, members of the research team have the opportunity to approach potential subjects and explain the study to them. However, in situations in which potential subjects’ names cannot be released outside an agency without their direct consent, potential subjects are first contacted by agency personnel and asked if they are willing to have their name released to the research team. Eaves (1999) referred to agency personnel as intermediaries and noted that those persons frequently included discharge planners, medical social workers, and casemanagement specialists. Because most potential subjects need to have a brief overview of a study before making a decision about releasing their name, it is important to provide agency intermediaries with clear scripted information about what to say about the study. A clear differentiation between providing a potential subject with background information and obtaining informed consent for study participation needs to be made; written and verbal consent to participate can be given only after a subject has been briefed by the research team and provided with an explanation of the risks and benefits of participation.
Because most potential subjects need to have a brief overview of a study before making a decision about releasing their name, it is important to provide agency intermediaries with clear scripted information about what to say about the study.
FACILITATING THE RECRUITMENT OF SUBJECTS FROM UNDERSERVED AND/OR DISENFRANCHISED GROUPS
Prior studies and methodologic articles have provided evidence that some population subgroups have been under-represented in clinical research and pose special challenges to investigators seeking balanced and representative samples in their research. Underrepresented groups may include inner-city residents (Evans, Mejı´a-Maya, Zayas,
48
Boothroyd, & Rodriquez, 2001), ethnic and racial minorities (Eaves, 1999; Hodge, Weinmann, & Roubideaux, 2000; Berman, Grosser, & Gritz, 1998), rural citizens (Ulrich, Browning, Gaffney, Schoter, & Chesnut, 1997; Shreffler, 1999), persons with physical disabilities (Bach & McDaniel, 1995), persons with low incomes (Berman, Grosser, & Gritz, 1998); the elderly (Leader & Neuwirth, 1978), and persons with traumatic memories of a health-related event (e.g., perinatal loss) (Coˆ te´ -Arsenault & Morrison-Beedy, 1999). For a clinical trial that was conducted in the United Arab Emerates, Glasgow and colleagues (1996) reviewed recruitment and dropout rates at different study points and concluded that clinical trials in developing nations may take up to twice as long as originally anticipated. The authors suggested that researchers working in isolated areas need to devote additional resources to sustain subject recruitment and retention throughout the data collection process. On a related issue, Walson (1999) noted that treatment for any condition that may directly or indirectly lead to denial of insurance coverage for a condition is a strong disincentive to research participation. When the investigator has the ability to directly access potential subjects through advertising, Internet groups, clinical contacts, or support groups, recruitment activities can be framed in a manner that speaks directly to the target population. Investigators use a variety of verbal and visual strategies to catch the attention of potential subjects. Shreffler (1999) discussed the importance of developing culturally sensitive research materials in the design of studies that are conducted in rural and frontier communities. The author suggested that selecting a graphic or logo that is emblematic of a specific community may help increase research participation; examples of such graphics for rural communities include crop motifs, a farm scene, or the town logo. In an article addressing subject recruitment of American Indians, Hodge and colleagues (2000) emphasized that recognizing the distinctiveness of each tribal nation was key to successful recruitment for clinical trial research. In a study involving rural residents with osteoporosis, Ulrich and colleagues (1997) used a mobile clinic van to travel to Montana and Wyoming communities for osteoporosis testing. The authors attributed the success of their recruitment efforts to use of the mobile clinic and suggested this
BUTTERFIELD ET AL
approach as a method of overcoming low research participation in rural communities. When the investigator is working to access potential clients through agency intermediaries, efforts need to focus on enlisting the ongoing attention and support of agency personnel. When appropriate, financial incentives may be offered, although ethical guidelines in the use of “finder’s fees” for subjects need to be implemented before any contact with agency personnel (Walson, 1999). Other incentives may include books for the agency library (e.g., new nursing textbooks), desk items such as pens or coffee mugs with a project or university logo, or letters to employees’ supervisors that recognize the employee for their contributions to identifying research subjects.
When the investigator is working to access potential clients through agency intermediaries, efforts need to focus on enlisting the ongoing attention and support of agency personnel.
UNDERSTANDING THE AGENCY PERSPECTIVE
Because of their economy of scale and strong organizational infrastructure, large hospitals, clinics, and health maintenance organizations have historically played a central role in subject recruitment efforts for descriptive studies as well as clinical trials. Employees in such agencies who attended university programs affiliated with academic health sciences centers are generally familiar with subject recruitment advertisements that are frequently posted in hospital elevators and waiting rooms. Frequently, they have participated in clinical trials or assisted with other research-related activities. On the other hand, smaller agencies and those located in rural communities are less likely to have employees that have worked in a researchintensive environment. To those employees, the research role is new, and they may not be familiar with research journals or internal publications that showcase research results. Engaging personnel in novice research agencies requires ongoing attention to employees’ workloads and interest in research. Because personnel
OVERCOMING SUBJECT RECRUITMENT
may be extremely busy, it is important not to assume that more information about research is better than less. In a discussion of obstacles to the development of a rural health promotion program, Wagner (1994) discussed the importance of building collaborative relationships among multiple community agencies. The author identified several barriers to program participation among agency staff members; these barriers included (1) skepticism about the potential success of the program, (2) no prior experience in program development, and (3) a lack of personnel with available time to take on a new initiative.
Engaging personnel in novice research agencies requires ongoing attention to employees’ workloads and interest in research.
The plan for subject recruitment should reflect elements of the organizational culture from both the clinical agency as well as the agency sponsoring the research. It is imperative to obtain the support of agency supervisors, who not only can provide entree into an agency but also can provide valuable insights into how the agency functions. Although research publications are valued in academic circles, agency personnel may or may not want to be provided with background publications and reports that detail each phase of the research process. They may, in contrast, want information that is distilled into a concise handout, similar to an executive summary for a scientific report. It is critical to ask agency personnel what they want and need to assist with identifying potential subjects for study participation. The importance of brief but regular contacts with agency personnel to troubleshoot problems or clarify any questions about subjects’ participation cannot be overestimated. Contact should be made using the preferences of agency personnel; some groups may want to have regular faceto-face meetings, whereas others may prefer a regular e-mail or telephone update about the progress of the research.
49
The plan for subject recruitment should reflect elements of the organizational culture from both the clinical agency as well as the agency sponsoring the research.
CASE EXAMPLE: OCCUPATIONAL BACK PAIN IN SMALL INDUSTRIES
The goal of the Reducing Occupational Disability in Rural Workers Study is to systematically examine feasibility issues in delivering a clinical intervention on behalf of workers with an occupational low back injury. Because the vast majority of occupational health studies have recruited workers from large industries (e.g., ⬎1,000 employees), this research specifically targets workers employed in smaller companies. Such companies are unlikely to have occupational health professionals onsite or have a risk reduction plan for the company; in addition, smaller companies are less likely to develop a comprehensive return-to-work plan for an injured worker. Previous research by Butterfield and colleagues (1998, 1999) has provided evidence that back-injured workers employed in small companies (100 or ⬍ workers) have significantly more lost workdays than those employed in larger companies (mean small companies ⫽ 56 days paid absenteeism, mean larger companies ⫽ 32 days, t test ⫽ 2.31, P ⫽ .02). Workers in the current study are randomly assigned to either standard care (i.e., control group) or standard care plus four home visits focusing on back injury recovery and return to work (i.e., treatment group). Because potential workers for the study are referred from workers’ compensation carriers in Oregon and Montana, claims adjusters in these two organizations had been involved in earmarking files on new claims that may meet the study inclusion criteria. Claims adjusters did not play any role in screening potential workers or enrolling them in the study, but they did need to call workers and obtain permission to have workers’ names and phone numbers released to members of the research team. Because back claims in the target areas occurred on an intermittent basis, referrals of potential subjects required the ongoing attention of claims adjusters, many of whom had never been involved in clinical research before. Study personnel
50
BUTTERFIELD ET AL
Figure 1. Reducing occupational disability in rural workers: Montana worker’s comp referral flow sheet.
made site visits to the insurance carrier in Montana before the beginning of the subject enrollment period; however, after several months, referrals of potential subjects decreased significantly. In assessing the rea-
sons for the problem, the research team noted several organizational factors (within either the university or the insurance organizations) that may have contributed to the decline, including the following:
OVERCOMING SUBJECT RECRUITMENT
51
1. Low numbers (i.e., low incidence) of claims for low back injuries in the participating rural communities, leading to claims adjusters having to relearn the process with each new claim. 2. Seasonal changes in claims for low back injuries, with more injuries occurring in the summer. 3. A corporate reorganization at the insurance agency during the data collection period, leading to a temporary changes in departmental assignments and daily work tasks of staff. 4. Implementation of a new and more sophisticated computer system within the insurance agency, leading to new training needs for the claims adjuster staff. 5. Normal variability in the market share of different workers’ compensation carriers, leading to changes in the number and type of companies (and workers) served by the carrier. 6. Internal staff changes at the insurance agency (e.g., employees transferring into a new department). 7. Annual turnover in university research associates (e.g., graduate students) working on the project. In response to the decrease in referrals, the principal investigator and research staff worked to reduce obstacles for the claims adjuster to screen for study referrals. In addition, the occupational health consultant and a managed-care administrator were contacted to make suggestions that may be unique to the workers’ compensation system. A major theme involved the research team’s perception that the busy nature of the claims adjusters’ day kept them from thinking about the study on a regular basis. In addition, the claims adjusters may not have seen how findings from the study could lead to improvements in care for low back injuries or occupational injuries in general. After discussion with the research team and workers’ compensation personnel, the following actions were implemented. 1. The research team met with the chief executive officer of the company and the Board of Directors to update them on the status of the research and thank them for sharing in a strong collaborative effort. 2. Several books on health and back care were
purchased and given to the claims adjusters to show the appreciation of the study team for ongoing efforts of the claims adjusters. 3. Information about what to do with a new claim was simplified and reprinted on a one-page flow sheet and redistributed to the claim adjusters. Information from this flow sheet is listed in Figure 1. 4. Weekly e-mails were sent asking agency personnel to review their files for potential subjects; in addition, a quick “thank you” mail was sent to the claims adjuster after a referral was received. 5. Revised instructions to claims adjusters provided them with a script to use in speaking with potential subjects. From these efforts, referrals of potential subjects increased over the course of the study. The frequency of back injury workers’ compensation claims in the target areas is generally low, and referrals continue to come on an intermittent basis; in addition, a number of temporal patterns with referrals (e.g., more injuries in the summer) have been noted. Because of these factors, subject recruitment continues to be a daily challenge; however, organizational barriers have been removed with a commensurate increase in subject yield. The research team has a better appreciation of the time obligations of claims adjusters and works to make the referral process as clear as possible. Finally, there is greater attention to the shared vision of the research team and workers’ compensation carriers, which is to promote the fullest, healthiest, and most timely recovery for all injured workers. This goal encompasses informing workers about the availability of a clinical research project as well as understanding the complimentary roles that researchers and agency personnel play in the subject recruitment process. Through their daily efforts on behalf of quality care, agency personnel provide the critical link between applied clinical research and individualized client care. ACKNOWLEDGMENTS The authors would like to thank Saif Corporation, Salem, OR, and the Montana State Fund, Helena, MT, for their contributions to the success of this research.
REFERENCES Bach, C.A., & McDaniel, R.W. (1995). Techniques for conducting research with quadriplegic adults residing in the community. Qualitative Health Research 5(2), 250-255. Berman, B.A., Grosser, S.C., & Gritz, E.R. (1998). Recruit-
ment to a school-based adult smoking-cessation program: Do gender and race/ethnicity make a difference? Journal of Cancer Education, 13(4), 220-225. Butterfield, P.G., Spencer, P.S., & Yates, S. (1999). Back-pain
52
related absenteeism: Differences between small and large companies. Journal of Agricultural Safety and Health, 5(2), 215-225. Butterfield, P.G., Spencer, P.S., Redmond, N., Feldstein, A., & Perrin, N. (1998). Low back pain: Predictors of absenteeism, residual symptoms, functional impairment, and medical costs in Oregon workers’ compensation recipients. American Journal of Industrial Medicine, 34, 559-567. Coˆ te´ -Arsenault, D., & Morrison-Beedy, D. (1999). Practical advice for planning and conducting focus groups. Nursing Research, 48(5), 280-283. Eaves, Y.D. (1999). Family recruitment issues and strategies: Caregiving in rural African Americans. Nursing Research, 48(3), 183-187. Evans, M.E. Mejı´a-Maya, L.J., Zayas, L.H., Boothroyd, R.A., & Rodriquez, O. (2001). Conducting research in culturally-diverse inner-city neighborhoods: Some lessons learned. Journal of Transcultural Nursing, 12(1), 6-14. Glasgow, N.J., Murdoch, J.C., Baynouna, L., Al Sabosy, M.M.A., Al Jabiri, O.N.K., & Harron, D.W.G. (1996). Doing randomized controlled trials in a developing country: Some practical realities. Family Practice, 13(1), 98-103.
BUTTERFIELD ET AL
Hodge, F.S., Weinmann, S., & Roubideaux, Y. (2000). Recruitment of American Indians and Alaska Natives into clinical trials. Annals of Epidemiology, 10, S41-48 (suppl). Leader, M.A., & Neuwirth, E. (1978). Clinical research and the noninstitutional elderly: A model for subject recruitment. Journal of the American Geriatrics Society, 26(1), 27-31. Shreffler, M.J. (1999). Culturally sensitive research methods of surveying rural/frontier residents. Western Journal of Nursing Research, 21(3), 426-435. Urich, U., Browning, M., Gaffney, E.V., Schoter, K.H., & Chesnut, C.H. (1997). Implementation of an osteoporosis research program with a mobile dual-energy X-ray absorptiometry unit: The Montana/Wyoming experience. Osteoporosis International, 7(6), 583-8. Wagner, D.I. (1994). Health promotion cooperatives: An alternative delivery system for rural health promotion programs. Journal of Health Education, 25(2), 77-81. Walson, P.D. (1999). Patient recruitment: U.S. perspective. Pediatrics, 104(3), 619-622.