Despite their higher incidence and mortality of cancer relative to their Caucasian counterparts, African American men are not well-represented in cancer screening trials. While cancer screening trial participation is of major importance for all individuals, it is of particular importance for African American men. Proper sampling of a heterogeneous population to ensure sample representativeness is a key component of valid epidemiological and clinical research. Without adequate numbers of African American men in cancer clinical trials, the extent to which trial results can be generalized to members of this population is in question. Swanson and Ward have developed a conceptual framework encompassing barriers related to the recruitment of members of minority groups to clinical trials (1). This framework includes four barriers: social-cultural barriers, economic barriers, individual barriers, and barriers inherent in study design. Social-cultural barriers include fear and mistrust of federally sponsored research. Economic barriers include the costs associated with participating in a research study, such as transportation and parking costs. Individual barriers include denial of disease, and barriers inherent in study design refer to barriers that are intrinsic to the design of research projects, such as the requirement that consent forms be returned by mail. The purpose of this presentation is to describe the lessons learned from a program of research designed to identify and overcome the barriers to cancer clinical trials enrollment and retention among older African American men that are identified in the Swanson and Ward framework (1).

In our first study, we conducted a four-arm randomized trial (the AAMEN Project) to test the effects of three increasingly intensive interventions vs. standard recruitment processes in recruiting African American men aged 55–74 years to a prostate, lung and colorectal cancer screening trial. The recruitment interventions addressed the four previously described barriers to clinical trial participation: social-cultural barriers, economic barriers, individual barriers and barriers inherent in study design. Social-cultural barriers were addressed by the inclusion of an enhanced recruitment letter written by, and including the photograph of, a prominent local African American business owner who employed many African American men and was a former professional basketball player. Social-cultural barriers were also addressed by the inclusion of research team members whose racial and ethnic backgrounds matched those of potential participants. We also included African American churches as project sites to increase a sense of community partnership among study staff, local community members, and potential trial participants. Economic barriers were addressed by the provision of transportation to the recruitment sites. Individual barriers were addressed in the enhanced recruitment letter, which included a statement focusing on the vulnerability of African American men to major health conditions. Barriers inherent in study design were addressed by allowing participants in one of the intervention arms to sign the consent form during a church-based project session, rather than requiring them to sign it and return it by mail. Participants were randomized to a control group or to one of the three increasingly intensive intervention arms, which used different combinations of mail, phone and church-based recruitment. Of the 39,432 African American men residing in the geographically defined study population (southeastern Michigan and Northern Ohio), 17,770 men (45%) could be contacted, and 12,400 (31% of 39,432) were found to be eligible to participate. No statistically significant differences in age, education or income level were found among participants in the four study arms. A significantly greater enrollment yield (3.9%) was seen in the most intensive, church-based intervention arm, compared to the enrollment yields in the other two intervention arms (2.5% and 2.8%) or the control group (2.9%) (p<0.01). Future studies could shed light on the effectiveness of specific recruitment strategies for African American men with different sociodemographic characteristics, such as age and income. The relative impact of various barriers to participation could be examined within these different sociodemographic groups.

We next conducted a study to examine adherence among older African American men (ages >= 55 years) enrolled in a cancer screening trial for prostate, lung and colorectal cancer. For this study, we defined adherence as completing the trial screenings. We used a randomized trial design. The intervention consisted of case managers (similar to patient navigators) who contacted intervention group participants (n=352) at least monthly by telephone and provided information and referral services. A central tenet associated with case management is the identification of needs by the clients. In addition, the client included not only the identified participants but also the participants' spouses or partners. In our recruitment trial, we found that the spouses or partners tend to serve as gatekeepers in terms of telephone access to the study participants. The case managers addressed social cultural, economic, and individual barriers to continued trial participation as well as barriers inherent in trial design, based on the Swanson andWard framework (1). To address social cultural barriers, the two case managers (who were African American women) served as sources of social support to the men assigned to the intervention group. The case managers addressed economic barriers by providing participants with referrals to free or low-cost transportation services for older adults, and by linking them with community resources to meet their other economic needs. The control group included 351 participants. The case managers addressed individual barriers by providing information about health issues affecting older African American men, such as diabetes and hypertension. Barriers inherent in study design were addressed when the case managers gave participants a verbal description of the procedures they would undergo in the cancer screening process, giving participants a chance to ask questions and express fears about the process. The case managers made 14,978 calls to participants, resulting in 780 referrals for health and human services. The 10 most frequent referrals were for assistance with scheduling medical appointments, health information, insurance information, legal aid, transportation, cancer screening information, information technology/computer information, employment, housekeeping/chore services, and food programs. Among participants with low income, those in the intervention group had higher screening adherence rates than did participants in the control group for (a) prostate-specific antigen test for prostate cancer (74.3% vs. 53.0%, p=0.001), (b) digital rectal exam for prostate cancer (66.2% vs. 46.1%, p=0.011), and (c) chest x-ray for lung cancer (70.9% vs. 51.3%, p=0.012). We found no statistically significant differences in adherence rates for flexible sigmoidoscopy screening for colorectal cancer. In contrast, among participants with moderate-to-high income, we found no statistically significant differences in adherence rates between intervention and control group participants for any of the screening tests. Thus, we discovered that the case management intervention was effective in enhancing adherence among participants with the lowest income, who in many studies are the most difficult to retain.

We also examined the effects of baseline comorbidities on screening adherence in a sample of older African American men (ages >= 55 years) enrolled in the case management intervention study. In general, participants with comorbidities were no less likely to adhere to trial screening than participants without comorbidities. Exceptions were current smokers and participants with chronic bronchitis. Current smokers were less likely than others to adhere to the prostate-specific antigen test (p=0.02) and the digital rectal examination for prostate cancer screening (p=0.01) to the chest ex-ray for lung cancer screening (p<0.01), and to the flexible sigmoidoscopy for colorectal cancer screening (p=0.04). Participants with chronic bronchitis had lower rates of adherence to the chest x-ray (p=0.06). Having a relative with cancer positively influenced adherence to the digital rectal examination (p=0.05). Thus, we found that older African American men with comorbidities are very good candidates for participation in longitudinal cancer screening trials. However, the study results show that current smokers were less likely to engage in cancer screening.

In summary, our results show that older African American men will participate in cancer clinical trials if intensive recruitment methods involving a significant amount of face- to-face contact are employed. Also, in order to retain men once they are recruited, a case management/patient navigation intervention can be successfully employed. Smoking status is a major barrier to adherence. In the future, it would be important to investigate which other health behaviors are associated with adherence to cancer screening.

Second AACR International Conference on the Science of Cancer Health Disparities— Feb 3–6, 2009; Carefree, AZ