Wilfred Y. Fujimoto, MD

Having carried out diabetes research in an Asian population for many years, I believe that minority participation in clinical research is both necessary and beneficial. We have learned much that is necessary to better understand the etiology and pathogenesis of diabetes in the target population, which has in turn benefited from this knowledge. However, I have learned, as have many others who have carried out research in minority populations, that there are many barriers to minority participation in clinical research. Many of the issues that affect minority participation in clinical research will be addressed in this review.

Minority Participation in Clinical Research—Required by Law
Inclusion of minorities in clinical research became a legal requirement in 1993. The National Institutes of Health (NIH) Revitalization Act of 1993 deemed it imperative that research determine whether interventions or therapies being studied differentially affected women or men or members of minority groups.

To assist investigators, guidelines were developed and published in the Federal Register.¹ These guidelines ensured that all future NIH-supported research involving human subjects would obtain information about individuals of both genders and from diverse ethnic and racial groups.

Requirements were most strict for Phase III clinical trials, for which evidence must be reviewed to determine whether clinically important gender or race/ethnicity differences in the interventions might be expected. An implication of this requirement for Phase III clinical trials is that increased attention should be given to gender, race, and ethnicity in earlier stages of research that provide the background information and preliminary data needed to make informed decisions when planning recruitment for Phase III clinical trials.

The Act specifically required that the director of NIH, in consultation with the director of the Office of Research on Women’s Health and the director of the Office of Research on Minority Health, engage in efforts to recruit women and minorities in NIH-sponsored clinical research. Furthermore, NIH funding components were specifically instructed to not award any grant, cooperative agreement, or contract or support any intramural project to be conducted or funded in fiscal year 1995 and thereafter that did not comply with this policy. For research awards that are covered by this policy, awardees must report annually on gender and race/ethnicity of research participants.

Costs of including women and minorities are not a permissible consideration for Phase III clinical trials. For other clinical research, costs are also not a permissible consideration unless the data regarding women and minorities that would be obtained if such inclusion were required have been or are being obtained through other means that provide data of comparable quality.

The only permissible exclusions to the requirement are if the inclusion of women and minorities is inappropriate with respect to the health of the subjects, with respect to the purpose of the research, or under other circumstances that the director of NIH might designate. Inclusion of women and minorities is not required in a Phase III clinical trial if there is substantial scientific evidence that there are no significant differences between a) the effects that the variables to be studied have on women or members of minority groups, and b) the effects that the variables to be studied have on individuals who would serve as subjects in the clinical trial in the event that these requirements were not in place.

In summary, beginning in 1995 there has been an increased opportunity for participation by women and minorities in clinical research in the United States. As a result, there are now increased opportunities for obtaining information concerning health and disease among all Americans. Furthermore, where differences existed between genders or among racial and ethnic groups, research will now be carried out to detect and explain these differences.

The types of studies affected range from the smallest observational studies and Phase I–II clinical trials to the largest Phase III clinical trials. Primary prevention trials involving disease-free participants, as well as secondary and tertiary intervention trials involving individuals with disease, are also covered.

Many investigators are novices when it comes to recruiting women and members of minority groups for clinical studies. The NIH recognized that understanding and incorporating these requirements would take some time. To assist investigators and their staff, the NIH published a primer that outlines key points concerning the recruitment and retention of participants in clinical research.²

Importance of Minority Participation in Diabetes Research
Type 2 diabetes mellitus has reached epidemic proportions in many minority populations.3 In virtually every minority group in the United States, diabetes prevalence is greater than in the majority white population. We also now know more about etiological factors that might be responsible for this increase.³ The information gathered so far, as well as additional information yet to be obtained, will be essential to the implementation of health policies and interventions that are ethnically specific.

Members of minority populations should be convinced to hold as an attainable long-range goal a reduction in the incidence and prevalence of diabetes in their populations. A more immediate goal might be the availability of state-of-the-art diabetes therapy that is culturally appropriate. To achieve these goals, minority participation in all aspects of diabetes research is highly desirable.

Minority participation in diabetes research is also desirable from a national perspective. The United States has one of the most ethnically diverse populations in the world, and minority groups have composed an increasingly larger proportion of this country’s population during each of the three most recent U.S. censuses.^4-6 Nearly a third of the current United States population belongs to a minority group, and this proportion is expected to increase steadily, such that by the early part of the 21st century, today’s minority groups are likely to make up more than half of the United States population.

Diabetes is already a leading contributor to total health care expenditures in the United States. The direct medical and indirect expenditures attributable to diabetes were estimated at $98 billion in 1997.⁷ Total medical expenditures by people with diabetes were $10,071 per capita, compared with $2,669 for people without diabetes.

Total health care costs in the United States may be anticipated to grow in the future since the number of people with diabetes is expected to increase. The growth of minority populations will contribute significantly to this anticipated increase in total costs attributable to diabetes since these are the populations that have the greatest number of affected and at-risk individuals. Greater minority participation in all aspects of diabetes research will help in learning more about how to reverse this trend.

Barriers to Recruitment
Working in minority populations is challenging, even for the most experienced clinical investigator. Barriers to recruiting minority participants fall into three general categories. The first of these are relatively simple logistical barriers that can be overcome by careful planning. Falling into this category are factors such as transportation, meals, child or elder care, and time of visits.

Participants may be keenly interested in volunteering for the study but may not immediately disclose barriers to participation. Staff members should be trained to pick up on clues that a participant may give and to find out whether they can help to overcome to these barriers.

As a study progresses, the staff will gain a clearer understanding of the more frequent barriers to participation and should develop solutions. In our own research, the main barrier has been the inconvenience of study visits, which we have met by adjusting the study schedule to meet the participants’ needs, including scheduling procedures for weekends, holidays, or vacations. In addition, as our study cohort has become older, transportation to and from the medical center has become an increasingly important barrier, which we have addressed through various solutions, including a clearer description of how to travel to and from the center, use of a special service available to elderly individuals through the public bus system, use of taxi vouchers, and provision of rides by staff.

More difficult to overcome, even with planning, are what may be considered to be complex logistical barriers. There may be a fear of large institutional settings. Staff may not be culturally diverse and may lack cultural sensitivity. Recruitment and instructional materials and assessment instruments may contain inappropriate language or be at an inappropriate literacy level.

Ways to overcome these barriers include having members of the target population visit the clinic, hiring appropriate staff, undertaking cultural sensitivity training, and including community involvement to ensure appropriate language and literacy level for all materials. It is important to assure that staff who interact with participants do so in a manner that is appropriate to the participants. In some studies, this may be facilitated by having at least some of the staff be of the same ethnic background as the target population.

Even more difficult to address are attitudes, beliefs, and inadequate knowledge. There may be fear or distrust of research. There may be unfamiliarity with research procedures such as randomization, masking, and placebo controls. There often are concerns about interference with primary care or with continuity of care.

Approaches to meet these barriers include having credible spokespeople, communication with and endorsement by community leaders, avoidance of jargon, and emphasis on both personal and ethnic group benefits from participation. Make sure that everyone’s agendas are clarified—investigators, community, participants. To maintain credibility, never suggest more than can be delivered, and never promote activities before they are ready. For example, never promise a complete physical examination unless you are prepared to perform one. The physical examination in many research protocols is relatively restricted and targeted. Yet some participants may believe that participation in a research study will substitute for the physical examination that they might receive during an annual visit to their health care providers. On the other hand, it is appropriate to indicate whether there are tests being done that might not be routinely provided by a participant’s health care provider and how participants may benefit from knowing the results from these.

Understanding the Target Population
An understanding of the target population is essential if an investigator is to succeed in recruiting in that population. The first step is to identify and characterize the population of interest within the community in which it resides. The size of the community, the geography of the area from which the study sample will be drawn, the size of the target population, its demography and history, and where it resides are important starting points. This information is required to estimate whether there are sufficient individuals with the appropriate basic characteristics for the research and whether travel to the research site might be a challenge.

Seldom is this sufficient information, however. Examples of additional information that is usually essential include ethnic characteristics, cultural customs and norms, socioeconomic level, language, education, literacy, health beliefs and knowledge, and needs and values, including economic, political, and religious. Although most researchers will recognize their importance when made aware of these issues, researchers often fail to recognize these factors during the planning process. Thus, researchers put inadequate effort into learning more about the target population, particularly when compared with the effort they put into learning more about the scientific background for the study.

Without this type of prior research, investigators may not adequately understand important differences within a population. "Southeast Asian," for example, does not take into account major differences among the peoples from this region (e.g., Filipino, Vietnamese, Thai, Indian, Laotian). Moreover, there are considerable differences even within a seemingly homogenous Asian ethnic group. For example, consider ethnic Chinese coming from northern China, southern China, Hong Kong, Taiwan, or Vietnam. Besides differences in cultural customs and norms, there are differences in dialect. In addition, there may be differences in socioeconomic level, education, literacy, and health beliefs and knowledge. Because of these differences, their needs and values will vary. Furthermore, in some communities a large proportion of the population may be quite uniform and have in common some or many of these characteristics, whereas for other communities, there may be considerable differences among individuals or families within the population. Similarly, "Native Americans" and "Hispanic/Latino" fail to portray the diversity within these populations.

There are additional factors to be considered, including the structure of the community. Recognition of how decisions are made is an important requirement. Identification of opinion makers within the community and their recruitment as spokespeople for the research are often immensely helpful. In addition, however, it is important to identify how decisions are made within the family and to learn what constitutes a family in the target population. Other important considerations are whether the target population is a stable one, an understanding of its migration pattern, and avenues by which it may be accessed.

Finally, the conventionally recognized mass media often may not be the most effective means of communicating with minority groups. Many minority populations have group-specific formal and informal channels of communication that may be much more effective. In our own research, the most effective channels have been through religious or social organizations, using print materials, oral presentations, or word of mouth. Word-of-mouth advertising should not be overlooked. This has turned out to be one of our most powerful recruitment tools, particularly when the "good word" is being given by current research participants or by members of our Community Advisory Board.

Information Gathering
To learn more about the target population, investigators must know how to effectively gather the desired information. The first step is to spend time in the community. In order to make this a valuable source of information, however, you must identify those individuals or groups from the target community who will be your entry. Remaining in the "ivory tower" of academia without going out into the community is not only culturally insensitive, but also a serious strategic error.

From a personal viewpoint, this was an important consideration when I first started our Japanese American Community Diabetes Study. Being racially and ethnically identified with the study population did not give me automatic entry into the community, and this fact was made clear to me. One of the first comments relayed to me from a member of our advisory board was that they could not support a research study in which the fruits of the research were not transmitted back to the community, and that my being Japanese American did not exempt me. The research staff of our study has been sensitive to this issue, and consequently we have provided information back to the community through biennial health fairs, regular articles in a community newsletter, information sessions at various community organizations, and most recently, by the establishment of a Web site.⁸

You may also want to consider doing focus groups and surveys, but you must be sure that the participants in these are representative of the target population in order for them to be sources of relevant information. In addition, you should do a library search, not only of the medical and other scientific literature related to the target population, but also of census data, government reports, maps, and newspapers to gain valuable demographic and socioeconomic information about the target population.

You should not hesitate to consult with experts who may be able to give you another perspective of the community. These experts may be individuals in government agencies and academic or research institutions who may have information or advice related to your research, and health professionals or others working in similar communities or on similar research problems.

Establishing Explicit Goals
Once you have obtained a better understanding of the study population, it is important to establish specific goals for recruitment and retention of study participants. The number of participants that will have to be recruited and the acceptable dropout rate will be determined from your best estimate of what is needed to detect significant differences for the primary outcomes (sufficient power).

Remember that effective recruitment and retention in one population or setting may be counterproductive in another. Involve important decision makers, organizations, and institutions in the community. Work the main communication channels, keeping in mind that many minorities are skeptical about participating in clinical research. Keep in mind also that community leaders can inform investigators about the social and economic needs of the population, factors that will affect both recruitment and retention. If it is necessary to pay participants, these payments and other incentives should not be of such magnitude as to be coercive. The level at which this occurs will depend upon the characteristics of the population.

In establishing lines of communication with and obtaining cooperation from the community, it is often useful to recruit minority investigators and staff. Consultation with hospital and clinic staff and community leaders during recruitment may allow investigators to fine-tune their recruitment strategies. The same type of interaction while conducting the research will help to fine-tune retention strategies. Identify recruitment and retention goals as explicitly as possible and share these with the community. Involving the community in these aspects of the research will often lead to a sense of responsibility for successful recruitment and retention among the community members.

It is also important that investigators, staff, and community all understand and agree to the design, methodologies, implementation, and completion of the study. It is useful to repeat the message that members of the target population must be included in clinical research if valid inferences are to be made about health and disease in that population. Investigators must continually strive to build the level of understanding and trust that will lead to a productive partnership with the community and successful conduct of the research.

Creation of advisory boards should be considered to review the research plans, recruitment, and retention. Advisory boards can also judge whether incentives are appropriate, coercion is used, and materials are appropriate in terms of language and literacy levels. Our own study has benefited immensely from the participation of our Community Advisory Board because of the partnership with the community that the members of the board have promoted.

Our board includes members from religious, social, business, and health care organizations that have close relationships to the Japanese-American community. Board members have been included in the planning phases of our grant applications and have been kept informed about the progress of our research. They have helped us in recruitment by placing articles in bulletins, newsletters, and newspapers, obtaining invitations to speak at the organizations that they represent, obtaining mailing lists and addresses, and organizing our biennial health fairs. They have advised us on how to provide greater outreach to members of the community not only about our research, but also about diabetes management and prevention.

As a result, our study has achieved very high visibility not only among the Japanese-American community but also among Asian-American communities in this region. We have subsequently formed a second board representing younger third-generation Japanese Americans because those on the main board, which comprises mainly second-generation Japanese Americans, believed that we should work more closely with this younger generation. One suggestion from this younger group that has been implemented is the development of a Web site.⁸

Approaches to Recruitment
There are two main approaches to recruitment. The first is for investigators to approach participants directly. This may include one or more of the following techniques: discussions at clinics, brochures, health fairs, direct mailings, talks to various groups, newspaper stories or advertisements, radio or television stories or advertisements, and telephone calls.

The second approach is indirect and uses physicians and other health care providers to serve as sources of participants. This indirect approach to recruitment may again include one or more techniques: telephone calls, letters to solicit referrals, seminars in hospitals or clinics, notices on bulletin boards or journals, letters to journals, talks at local and state medical societies or professional societies, and exhibits at conferences.

Recruitment strategies often include screenings at community health fairs or with targeted groups (e.g., clinic patients, work sites, churches, blood donors). Direct mailings may be effective using targeted lists (e.g., voter registrations, department of motor vehicles lists, payroll lists, purchased lists,) either as multiple mailings or staged mailings. Media include print media (e.g., daily and weekly newspapers, community newspapers, newsletters), radio, and television, and may include one or more of the following: news stories, public service announcements, and paid advertisements.

Consideration should be given to using multiple methods, either in combination or sequentially, to assure that the message is being received. A tracking system should be implemented to follow overall responses and yields by recruitment method and to allow simple cost analyses.

Obtaining Informed Consent
A significant component of the recruitment phase is the process of obtaining informed consent. Investigators working with minority populations should be particularly sensitive to this issue. If the proposed research is complicated and includes a lengthy screening process, careful consideration should be given to implementing a staged informed consent procedure that is integrated with screening. A staged informed consent process allows participants to gain enough information to make an informed decision about whether to proceed to the next step of screening. A consent form is available at each stage.

Using multiple approaches to transmit information to potential participants, including written and audiovisual materials, can assist participants in making an informed decision. Participants should be given ample opportunities to ask questions. Methods of assessing participant understanding may need to be implemented. All materials should be in the appropriate language and at the appropriate literacy level. Pre-testing in the target population of informational and informed consent materials should be seriously considered, particularly in research involving complex protocols.

Retention and Adherence
Retention and adherence are not synonymous. Retention refers to attendance at scheduled outcome visits and completion of outcome measurements. Adherence refers to attendance at scheduled intervention visits and completion of prescribed interventions.

Moreover, it should be recognized that although measures to promote one will generally assist the other, this is not always the case. Enthusiasm in promoting adherence may fail to take into account the possibility that this may cause participants who are having problems in adhering to the protocol to fail to show up for follow-up visits. The requirement that participant dropout be kept to a minimum may necessitate flexibility in adherence to the intervention.

Both retention and adherence start during recruitment. A screening process should be developed to identify potential participants who are likely to pose future retention and adherence problems. From their initial contact with the study, participants should be able to trust that they are being treated in an honest and forthright manner. This means that staff who interact with participants should be knowledgeable about the study.

Other important considerations concern time and travel and convenience to participants. The location of the clinic should be convenient to participants, and the hours of operation should be flexible. The participants should also feel connected to the study. Thus, staff should be available at all times. Regular communication with participants is to be encouraged, and consideration should be given to having a newsletter for participants.

Potential obstacles should be anticipated. There should be clear instructions about procedures and directions to the clinic. Transportation and parking may be significant problems for some participants. Other participants may have young children or elderly parents who will require care while they are visiting the clinic for their scheduled appointments. Just as was the case in recruitment, attention should be paid to language and literacy level and possible cultural issues.

Finally, incentives are sometimes helpful, although they should never be used to substitute for the above. Free medical tests are usually possible. In some cases, free medical care may also be feasible. Gifts that serve as symbols of participation in the study are often quite effective and need not be expensive. In many studies, honoraria for participating are also used. Attention should be given, however, to assure that these cannot be interpreted as being coercive. Incentives should never be overemphasized at the expense of other components of the program.

Incentives that we have found to be most effective are the timely return of laboratory test results; souvenirs, such as a T-shirt or coffee mug that symbolized participation in the study; potluck socials to which participants and staff are invited; newsletters containing not only educational information but also features, such as recipes; and honoraria.

Dropouts will occur but should be minimized as much as possible. The best approach is to anticipate these from the beginning and develop measures to identify potential dropouts, and transform them into adherent participants. Learn what factors might be responsible for dropping out, and develop strategies to address these issues.

Design and Conduct Evaluations
The importance of designing and implementing an evaluation plan to assess how well recruitment and retention strategies are working should be recognized. This is best done in cooperation with community leaders and potential participants to ensure that the strategies selected conform to the needs and values of the participants and their community. An advantage of this is that the community will share with the investigators the responsibility for effective recruitment and retention.

A monitoring process should be in place in order to follow subject accrual and retention on a frequent and regular basis and to compare these findings with established goals. This information should be shared with the community. This will allow investigators to determine which strategies work, show that progress has occurred, and identify what is not working.

In planning recruitment and retention strategies, evaluation is an integral part of the process and provides investigators the tools for addressing several tasks. This evaluation process should begin at the start of the study and be continuous. Whether the procedures being followed are feasible must be ascertained in order to modify procedures when required. Evaluation will also establish the fact that investigators and staff are accountable for specific tasks. This will assist in providing information that will encourage acceptance and response of the community and participants. It will also allow the establishment of feedback mechanisms that can be implemented to avoid or counter participant dropout.

Costs
In planning studies, the costs associated with recruitment, retention, and adherence should not be overlooked since these are usually not trivial. There should be sufficient personnel to plan and implement recruitment, screening, obtaining informed consent, and problem solving. Staff will also need to be assigned to planning and ordering incentives, writing newsletters, communicating with participants, and removing obstacles to retention and adherence, such as transportation, parking, child care, and elder care.

There are also direct costs, such as the printing of brochures, flyers, ads, letters, and consent forms; mailing costs; ad costs; and telephone costs. Travel expenses should be included. These may be related to meetings and screenings for both staff and participants.

Minority Participation in Clinical Research—Final Points to Consider
There are several points to consider when planning minority participation in clinical research. Researchers should ask themselves the following questions:

1. Is the subject population adequately drawn? Will minority subjects be appropriately and adequately represented? If not, is homogeneity of the study population justified?

2. Are subject recruitment strategies appropriate for obtaining a diverse subject population?

3. Have the special needs of prospective participants been addressed in order to minimize or remove obstacles?

4. Has undue influence or coercion been eliminated?

5. Does the proposed consent process ensure open and effective commu- nication between researchers and prospective participants? Are the consent documents written in language that will be easily accessible to subjects? Are documents in a foreign language necessary? Is foreign language facility on the part of the research staff necessary?

There are several useful sources of information about recruiting minorities in clinical research, including some excellent reviews to which the reader is referred.^9-11

References
¹Department of Health and Human Services, National Institutes of Health: NIH guidelines on the inclusion of women and minorities as subjects in clinical research. Federal Register 59:14508-13, 1994.

²National Institutes of Health: Outreach Notebook for the NIH Guidelines on Inclusion of Women and Minorities as Subjects in Clinical Research, Bethesda, MD, NIH, 1994.

³National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases: Diabetes in America. 2nd ed., Bethesda, MD, NIH, 1995. (NIH Publ. No. 95-1468).

⁴Bureau of the Census: Census of the Population: 1970. Washington, D.C., U.S. Govt. Printing Office, 1970.

⁵Bureau of the Census: Census of the Population: 1980. Washington, D.C., U.S. Govt. Printing Office, 1980.

⁶Bureau of the Census: Census of the Population: 1990. Washington, D.C., U.S. Govt. Printing Office, 1990.

⁷American Diabetes Association: Economic consequences of diabetes mellitus in the U.S. in 1997. Diabetes Care 21:296-309, 1998.

⁸http://weber.u.washington.edu/~jacds/

⁹Freedman LS, Simon R, Foulkes MA, Friedman L, Geller NL, Gordon DJ, Mowery R: Inclusion of women and minorities in clinical trials and the NIH revitalization act of 1993—the perspective of NIH clinical trialists. Controlled Clin Trials 16:277-85, 1995.

¹⁰Spilker B, Cramer JA: Patient Recruitment in Clinical Trials. New York, Raven Press, 1992.

¹¹Gonzalez VM, Gonzalez JT, Freeman V, Howard-Pitney B: Health Promotion in Diverse Cultural Communities. Health Promotion Resource Center, Standford, Calif., Stanford Center for Research in Disease Prevention, 1991.

Wilfred Y. Fujimoto, MD, is a professor of medicine in the Division of Metabolism, Endocrinology, and Nutrition of the Department of Medicine at the University of Washington in Seattle.

Return To American Diabetes Association Home Page

Copyright © 1998 American Diabetes Association

Last updated: 9/98
For Technical Issues contact webmaster@diabetes.org