Participation of Minorities in Cancer Research

Abstract

Overall, participation rates in cancer clinical trials are very low, ranging from 3 to 20% of eligible participants. However, participation rates are especially low among the socially disadvantaged and racial/ethnic minority groups that have been historically underrepresented in clinical research. Structural factors such as study duration, treatment or intervention schedule, cost, time, followup visits, and side effects represent more of a barrier to participation among these groups compared with white, non-Hispanics. Attitudes, beliefs, perceptions, and knowledge regarding clinical research, and cultural characteristics of underrepresented minorities pose additional barriers to participation. This article focuses on the structural, cultural, and linguistic factors that affect participation in clinical cancer research for each major U.S. racial/ethnic group. Low socioeconomic status, speaking a primary language other than English, differences in communication styles, mistrust of research and the medical system, fear, embarrassment, and lack of knowledge about the origin of cancer appear to have a negative impact on clinical cancer research participation rates. Much of the information about these factors comes from studies of cancer screening because little data is available on the factors that prevent and facilitate participation of minorities in clinical cancer trials specifically. Such research is needed, and, given the heterogeneity within and between minority populations, should occur in several different geographic settings and with as many different minority subpopulations as possible.

Introduction

Overall, participation rates in clinical trials are very low, ranging from 3 to 20% (1). However, rates of participation in clinical cancer trials are especially low among the socially disadvantaged and racial/ethnic minority groups that have been historically underrepresented in clinical research in general 2, 3, 4. White, middle class, highly educated, married men are the most represented in treatment trials and cancer prevention research. Although participants in prevention research are more health conscious, vitamin users, more concerned about getting cancer, and have a family history of cancer 1, 5, 6, minority populations and the medically underserved are the least likely to participate in these trials. In recent years, racial/ethnic differences in participation rates in cancer treatment trials have decreased; however, prevention trial participation rates continue to be significantly lower for minority populations.

Lower minority participation in clinical cancer trials is a concern for both research and public health reasons. Without adequate minority enrollment in clinical trials, researchers cannot learn about differences among groups and cannot ensure the generalizability of results. In addition, participation in trials increases access to state-of-the-art cancer care, a critical factor in many minority communities that suffer disproportionately from cancer, particularly in terms of excess mortality from some cancers.

Most eligible individuals do not take part in clinical trials and little is known about the factors that affect their willingness to participate in clinical research or about successful recruitment strategies 1, 5, 7. Data on attitudes, perceptions, and beliefs of underrepresented populations regarding participation in clinical research and the effect on their expressed willingness to participate is limited. Most of the available information on cultural and attitudinal factors comes from cancer screening studies. Although this information is not specific to clinical trials participation or nonparticipation, many of the factors affecting cancer screening behaviors could be expected to have a similar effect on this decision process.

This article discusses the excess burden of cancer among minority populations that makes increasing their participation in clinical cancer studies an important research and health priority. Three types of factors are described for individual minority populations that can affect participation in clinical cancer trials: (i) structural factors; (ii) cultural factors; and (iii) linguistic factors.

Structural factors include concerns related to health and research systems and economic issues. Factors related to the research system include study duration, treatment or intervention schedule, cost, time, followup visits, and side effects, which overall represent more of a barrier to participation among minority groups compared with white, non-Hispanics 8, 9, 10, 11. Among socially disadvantaged individuals, whose struggle for day-to-day survival is a constant concern, these factors pose even larger barriers than for individuals of higher socioeconomic status (SES). In fact, the rate of participation in U.S. clinical trials is correlated with sociodemographics (e.g., educational attainment, income, employment status, insurance coverage). Regardless of race or ethnicity, low SES has repeatedly been shown to have a negative impact on clinical research participation 12, 13, 14, 15. There is tremendous heterogeneity in the sociodemographic characteristics both within and between minority populations, so this association is relevant to some minority subpopulations and not to others. Table 1 presents an overview of the SES of major U.S. minority populations.

Three reasons appear to account for the lower participation rates of people with lower SES. First, low SES results in limited access to health care 14, 16, which results in low cancer screening participation rates, low rates of prevention and diagnostic procedures (15), and a higher dependence on public hospitals where physicians are less likely to be involved in clinical research. Second, insurance issues also are related. Low SES patients are more likely to rely on Medicare and Medicaid insurance; however, their coverage of costs associated with participation in a clinical trial is inconsistent and often denied. Ethical concerns also arise in recruiting low SES patients and others who do not have health insurance or have inadequate coverage to prevention trials. If an abnormality is detected, these patients are usually without resources to receive followup care and cannot expect assistance from the trial because NIH funds for prevention trials cannot be used for followup care (15). Third, concerns about economic survival among lower SES individuals take precedence over those of health and well being. This reality often discourages patients from engaging in screening and prevention activities (3) and may reduce receptivity to health messages in general.

Culture is learned behavior shared by the members of a society, acquired by experience (17); it is a way of life that accounts for how and why individuals form beliefs, attitudes, and behavior, establish normative behavior, and organize social existence 14, 18. Cultural factors refer to the way of life for a group of people, the configuration of the patterns of learned behavior that is passed down to generations through language and imitation (17). Cultural beliefs and perceptions influence health behavior patterns and ultimately influence participation in health screening, prevention trials, diagnostic tests, compliance, as well as treatment-seeking behaviors.

The distribution of health determinants in diverse, multicultural populations has led to the examination of the effect of cultural variables on health outcomes (14). Although conceptualization of culture is complex, its elements exert profound impact on health beliefs, communication, attitudes, and health behaviors. The significance of culture in shaping behavior necessitates thoughtful, culturally sensitive planning and design of strategies to enhance participation of minority populations in clinical research.

Although there is tremendous cultural diversity both within and between minority populations, there are also several cultural and linguistic factors that are shared by many different populations that influence participation in cancer research. The problem of communication is often compounded by the historical experience of minority populations with research and the health care system that has led to a generalized high level of mistrust. Other factors that appear to influence participation in clinical research are fear, faith, embarrassment, emphasis on the family rather than the individual, and a lack of understanding of the origin of disease. Fear of cancer and an underlying belief that it cannot be cured is often based on the reality that minority populations experience; their family, friends, and neighbors often die soon after a diagnosis of cancer. Although these are some common themes, each of these factors is expressed in a different way in each of the different minority populations.

For many minority populations, especially immigrant populations, the dominant language spoken is not English. This poses a serious barrier to effective provider–patient communication and attempts at recruiting patients into clinical cancer trials. Communication styles (verbal versus written, body language) also affect overall clinical trial participation rates by influencing the level of understanding of the study, trust, and ability to understand and accept an informed consent form.

One reason for trying to increase minority participation in clinical cancer trials is a concern about the disproportionate burden of cancer that occurs in many of these populations. Of course, cancer can affect people of any age, race, gender, or SES. However, research suggests that the cancer experience of minorities and the underserved is very different from that of the majority of Americans, and within a racial group there are often large differences in cancer burden (15).

As Table 2, Table 3 illustrate, cancer incidence varies greatly among U.S. minority populations. The most alarming disparity occurs among African-American males, who experience cancer approximately 15% more frequently than white men and have the highest overall incidence of cancer among all racial groups (19). In general, other minority groups have a lower overall incidence rate of cancer than do white, non-Hispanic Americans. However, examination of cancer incidence by site of diagnosis and racial group suggests several areas of concern. All minority groups have higher rates of stomach cancer than the majority of Americans ( white, non-Hispanic Americans). Korean males, for example, experience up to a fivefold higher incidence of stomach cancer by comparison (Table 2). Liver cancer is very high among Chinese-American males, who experience 11 times the incidence of the U.S. white, non-Hispanic population (19). Similarly, disproportionate rates of cervical cancer are observed in many minority women (Table 3). For example, African-American and Hispanic women are twice as likely to receive a diagnosis of cervical cancer compared with white women in the United States. The rate of cervical cancer among Vietnamese women is an alarming 43.0 per 100,000 compared with 7.5 per 100,000 among U.S. whites (Table 3).

As illustrated in Table 4, Table 5, Table 6, many minority populations have higher mortality rates and lower survival rates for many cancers. Among all ethnic groups in the United States, African-American males have the highest overall cancer mortality rate (Table 4). Mortality rates from prostate cancer are notably high, at 2–5 times the level seen in other racial groups. Lung cancer mortality rates also are high among African-American males (30% higher than in U.S. whites).

Examination by site of diagnosis and racial group indicates that minorities have the highest rate of mortality from stomach cancer (similar to incidence). Mortality rates from cervical cancer also are elevated among most minority women (Table 5). Cancer mortality rates for Native Americans were not reported due to low representation (less than 25 cases), and information on rates among Koreans and Vietnamese also was not available.

Examination of survival data from the Surveillance Epidemiology and End Results (SEER) Study (1973–1981) gives a dismal picture of progress against cancer in minorities. Although overall incidence and mortality rates among many minority groups are low compared with those of the majority of Americans, survival from cancer is consistently poorer in these groups (Table 6) (20). Today, one-half of all whites with cancer will be alive 5 years from the time of diagnosis. In contrast, 5-year survival rates among African Americans and Native Americans are considerably reduced (38% and 34%, respectively). A similar pattern is seen, for example, in female breast cancer and cancers of the colon and rectum. The 5-year breast cancer survival rate among U.S. whites is 75%, versus 63% among African Americans and 53% in Native American women. Survival rates for cancer of the colon and rectum are 51% among U.S. whites, 41% among Filipinos, and 37% in Native Americans. Updated survival rates (1960–1992), which are available only for African Americans, support the suggested picture of poor survival among minority populations.

Thus, the major success in the “War on Cancer” in improving cancer survival appears to have been least effective among minorities and the underserved (15). This has largely been attributed to late stage at diagnosis, perhaps attributable in part to limited use of early detection services and access to state-of-the-art treatment. These data have led many researchers to focus their research on barriers to early detection services among minority populations. For minority populations, most of the knowledge available on barriers to recruitment into clinical cancer research has been derived from cancer screening studies and trials as well as behavioral intervention trials.

Although a low SES is a barrier to cancer screening and clinical research participation among subgroups of all minority (and majority) populations, the situation of each ethnic group, and varied populations within each group, is unique. This section profiles each of the major U.S. ethnic minority populations and describes the particular structural, cultural, and linguistic factors that affect their cancer screening behaviors and are likely to affect their participation in clinical cancer research. It is important to note that, even though we have tried to make the discussions of each group as comparable as possible, data limitations did not allow for completely consistent presentations.

Asian Americans presently make up approximately 2.7% of the U.S. population, but are one of the fastest growing minority groups in the country. It is estimated that they will comprise 10% of the U.S. population by the year 2050. This highly heterogeneous population comes from many countries of origin and has languages and cultures representing more than 30 different subgroups (21). A high proportion of the population are immigrants and refugees, yet many Asian Americans have been in the U.S. for several generations. This has resulted in a bimodal distribution of educational attainments, economic status, health status, and degree of integration into the U.S. culture. The commonly applied term “model minority” does not reflect this diversity and does a disservice to the large group of Asian Americans who are underserved. Any generalized statement made about this population must be interpreted with caution.

One important research-related problem is the difficulty in securing the cooperation and involvement of some Asian-American populations. In one study of Korean Americans, for example, participation rates were low: 48% of the eligible respondents declined to be interviewed. The reasons cited for not participating were lack of time due to long hours of work and that other members of the family did not allow an interview.

Asian-American women have a low rate of compliance with recommended cancer screening guidelines, with the lowest rates of Pap test screening, mammogram, and clinical breast examination of any ethnic group in the United States (22). Low cervical cancer screening rates have been reported among Vietnamese-American and other Asian-American women 23, 24.

Cultural beliefs may influence the willingness of women to undergo screening examinations, and these same beliefs may be relevant to clinical trials participation. Mo (25) studied the meaning of female cancers in Chinese-American women and found that only 6 out of 62 immigrant Chinese women thought it was important to have routine medical examinations. These women believed that the female organs ceased to function when they are not being used in procreation, and therefore, examination of the breast or cervix is not necessary when they are not bearing children. Additionally, for an unmarried woman to have such examinations would imply that she is sexually active, and such an admission would be untenable.

Kagawa-Singer (26) contrasted the cultural health values of the Euro-American cultural system with that of the Asian-American system. She observed that the Euro-American system holds that all life is sacred and must be preserved at all cost; autonomous decisionmaking must be maintained, and relief of suffering is a primary objective. In contrast, the analogous three values in most Asian-American cultures are that individual life is not sacred; the welfare of the group is the primary value; and decisionmaking is made by consensus. These cultural values may have impact on the individual's decision to undergo cancer screening or participate in research and the choice of treatment once cancer is found.

Kagawa-Singer et al. (27) also observed that in the dominant American culture, the breasts are symbolic of femininity, sexual appeal, and womanhood. Breast cancer seems to carry much greater symbolic significance for Euro-American women than it does for Asian-American women. It appears that Asian-American and Japanese women view the breast more functionally than Euro-American women, and the breasts are not as integrally woven into their sense of self.

Lee (28) looked at the barriers and incentives to cancer screening in Chinese American women in San Francisco. When the women were asked to cite their own reasons for not having breast and cervical cancer screenings, they cited: “nothing is wrong with me,” “because I am healthy, I do not need one,” “no need for it at my age,” and “the doctor did not recommend the test to me.”

This survey also showed that more than two-thirds of Chinese women had gone to traditional providers for preventive health. In health practices, they often shop for providers, going from one provider to the other. Additionally, Chinese-American women often went to temples to pray for their health and look to fortune-tellers for directions in all aspect of their lives.

McBride et al. (29) investigated factors associated with cervical cancer screening among Filipino women in California. Women who immigrated late in life and who preferred to use the Filipino language had lower screening rates. In the younger group, lack of Pap screening was associated with modesty and traditional gender role values. They considered it important to have a female doctor do the pelvic exam, felt embarrassed when a clinical breast exam was done by a male doctor, and perceived that male family members preferred a female doctor for their female relatives. As a predominantly immigrant population, Filipino-American women may continue to be influenced by the value of modesty imparted by their mothers and grandmothers.

Cervical cancer screening was studied in a Cambodian population in the Seattle, Washington, area by Taylor et al. (30). In this study, 24% of the women had never been screened for cervical cancer, and only 32% had received a Pap test in the previous year. Cervical cancer was an unfamiliar concept for many Cambodian women.

Certain cultural factors associated with cervical cancer screening were identified. These included a belief that life is a matter of karma or fate, and therefore destiny cannot be altered by early detection. There was a lack of familiarity with early detection concepts and the Pap test. Also, there was the perception that Cambodian women do not get cervical cancer if traditional postpartum practices have been observed. Other concerns were fear of cancer, which is perceived as incurable, and fear of surgery. There were also concerns of embarrassment, pain, and financial costs associated with Pap testing.

More than 63% of the total Asian-American population is foreign born (31), with approximately 40% having limited ability in English (21). This language barrier has clearly affected Asian-American participation in cancer screening, and it is reasonable to project that it negatively affects clinical trials participation as well. For example, among Chinese-American women in San Francisco (28), those who spoke English fluently had more regular screening tests than those who did not speak English well. In another cancer screening study of Korean women (32), investigators noted that only 22% of the respondents spoke English either moderately or very well. These immigrant women relied on Korean newspapers for their health-related information, and the authors concluded that they are not benefitting from the existing cancer education strategies in present use for the general public.

Pacific people reside in the State of Hawaii, the territories of American Samoa and Guam, the Federated States of Micronesia, the Commonwealth of the Northern Marianas, and the Republic of Belau and the Marshall Islands. Also, they are migrating to different parts of the United States (33).

Inhabiting the Pacific Islands are diverse people who find themselves confronting structural obstacles to health care and cancer clinical research. The political affiliation of the United States with various Pacific entities implies governmental responsibility for public health problems and solutions; however, too often, the needs of people in this geographically remote area are overlooked (34).

An important structural factor that affects participation in clinical trials among all groups is access to health care. Among Native Hawaiians and Pacific Islanders, access to health care, in many ways, is determined by geographical location and economic issues (35). Because research studies are typically affiliated with Honolulu and other West Coast hospitals and institutions, geographic location can be a major impediment to participation in clinical trials and cancer research studies. For example, although the majority of the 138,742 Native Hawaiians in the State of Hawaii live on the island of O`ahu (66%), many others live on other islands. For those who live on Hawai`i (17%), Maui (10%), Maui/Moloka`i (11%), and Kaua`i/Ni`ihau (6%), access to health care in urban O`ahu is impeded by costly travel, often requiring air transportation (35). Travel convenience and cost add other barriers to health services and research. Ground transportation presents a formidable access barrier throughout the Pacific. O`ahu has the only public transportation system in Hawaii, with bus routes only on major thoroughfares and schedules that are not always convenient.

Poor economic status is another structural barrier to health care and participation in clinical research. According to the indicators of poverty in the United States, such as participation in food stamp programs, public assistance, medical assistance, and unemployment as well as crowded living and homelessness, Native Hawaiians in Hawaii are economically disadvantaged compared with others. In 1989, 15% of all Native Hawaiian households received public assistance compared with fewer than 7% of other households in Hawaii 36, 37. Approximately 26% of food stamp recipients and 17% of individuals receiving medical assistance were Native Hawaiian 36, 37. In addition, one-fifth of Native Hawaiian households had annual incomes below $15,000, and Hawaiians averaged nearly $6,000 less per household than others in Hawaii 36, 37. Among all the major racial groups in Hawaii, at 10%, Native Hawaiians have the highest unemployment rate.

The unfamiliar and often alien nature of the research process for many Native Hawaiians and Pacific Islanders may be one of the key reasons they do not become involved. Scientists and academicians place great importance on research, but its significance is not clear to the “typical” person in a society, and is beyond the immediate priority and awareness of persons living in remote rural and island communities. Research has shown, however, that explaining the benefits of research to communities and families will enhance the potential for Pacific Islander participation in clinical trials and research studies (38). Often, personal contact through a respected community member or friend is an effective method (39).

In addition, contemporary paradigms of clinical research are in many ways opposed to traditional Pacific Island ways of learning. Traditionally, the development of knowledge in many Pacific Island cultures has been based on a close relationship between teacher and apprentice, in modeling or experiential learning and a “hands-on approach.” Yet, the relationship of cancer research and participant is mostly impersonal, and the nature of research is empirical, emphasizing written analyses. Furthermore, research results are not directly linked to a tangible skill or service. Altering the character of clinical research also may increase participation among these populations. This could include involving family members in support roles to improve individuals' understanding of how research is linked to family and community benefits.

Respect and understanding of the family system, cultural values, and the role of women in Pacific Island cultures are pivotal to increasing clinical cancer research participation. Among Pacific Islanders, the family is a cultural constant in a world view that highlights holistic affiliations over the importance of the individual 34, 40. Within many Pacific groups, the role of women is central to the family, and their identity often lies “within the bosom of the family and the group's performance” (41). For instance, among Chamorros, the importance of interdependence and family is reflected in the practices of sharing, desire for harmony, and discipline (42).

Although more women within the Samoan community are being diagnosed with breast and uterine cancer, women are reluctant to get mammograms and pap smears (43). Thus, the probability of women participating in clinical research becomes even more remote. Disease, disability, and death of a family member have great implications for the entire Pacific Islander family. A Pacific woman who recognizes the hardship and grief that her disability or death would bring to her family may be more motivated to participate in activities to enhance her well-being than the woman who is not so informed. The Wai`anae Cancer Avoidable Mortality Study found that the Native Hawaiian woman was more likely to get a mammogram because her family needed her, and it is important for her to stay healthy (44). This motivation also could prompt clinical trial participation.

A lack of trust in the health care system is another cultural barrier that can lower health service utilization as well as participation in clinical cancer research. For example, in a study sponsored by the National Cancer Institute, women in a high-density Native Hawaiian community identified negative perceptions of and experiences in the health care system (45) as key barriers to seeking screening services. The same can be expected to hold true for research participation. One cause of lack of trust in the health care system is receiving services that are not culturally compatible with the background of the patient. The lack of cultural sensitivity can have a variety of negative results. For example, resignation and avoidance may be two emotional reactions of the patient who is feeling uncomfortable in a system of health care or research. Noncompliance may be the result of an unsatisfactory interaction between medical provider/researcher and patient, rather than a reflection of a client who does not value his or her health.

Language also can be an obstacle for these populations to receive health services and participate in clinical research. In the United States, one-fourth of Pacific Islanders speak a language other than English at home (33). Among these groups, Tongans (72%) and Samoans (64%) have the highest proportion of persons speaking a non-English language at home. Among Samoans, language can prove to be a barrier in specific ways. For example, some Samoan words have multiple English meanings. Also, with some English words, there is no equivalent Samoan phrase or word (56). Finally, census data indicate that more than 10% of Pacific Islanders are considered “linguistically isolated” (33).

Although Native Hawaiians have the highest percentage of English-speaking people (33), language still may be a barrier. History indicates that Native Hawaiians became literate in English shortly after the arrival of the American missionaries in the 1800s, and English was used in all commerce and business. Later, as laborers for the sugar plantations were imported from Asia and Europe, the mixing of languages with English occurred, and a new “pidgin” English emerged as a common language of communication. As generations have passed, pidgin English has become commonly spoken by many; as a result, comprehension of standard English is sometimes a barrier, particularly among older, rural, and lower SES groups (37). Similarly, sentence structure appears to influence comprehension (46).

There are 511 federally recognized tribes in the United States, with a total of 1.9 million people (0.8% of the U.S. population) self-reporting American Indian or Alaska Native status in the 1990 U.S. Census. American Indian people reside in every state, with Oklahoma, California, Arizona, and New Mexico having the largest populations (more than 100,000 each). Most of the more than 300 reservation and trust lands in this country also are in the West (47). The largest tribes are Cherokee (308,000) and Navajo (220,000). Compared with white, non-Hispanics, American Indians have a significantly lower median age (27.8 versus 37.9 years) and a lower rate of high school completion (65.5% versus 83.0%) (Table 1).

Poverty remains an important issue facing all American Indians, regardless of tribal affiliation or geographic location. Unemployment is much more prevalent among American Indians than U.S. whites (14.4% versus 5.2%). The household income among American Indians is one-half to one-third that of U.S. whites, and the majority of these households do not have telephones or indoor plumbing 47, 48. They also have approximately three times the percent of households below the federal poverty line (30.9% versus 11.0%) compared with white, non-Hispanics (Table 1).

Another barrier to research participation is limited infrastructure for studies to reach and motivate these populations. American Indians are spread throughout the United States in both urban and rural areas. Each setting presents its own unique challenges to practitioners and researchers who desire to recruit from this special population into clinical studies. For example, 54% of all American Indians are located in metropolitan areas, yet less than 2% of the Indian Health Service (IHS) budget is allocated to urban clinics. This limited infrastructure for health care service delivery creates difficulty in reaching the urban population. Additionally, urban American Indians are often isolated from their extended families and tribal support system, further reducing the potential for networking in the community (49).

Approximately 20% of American Indians and Alaska Natives live in rural areas or on tribal trust lands, which are sparsely populated. Barriers in this setting include lack of transportation options, often compounded by long travel distances to medical facilities and the limited availability of medical care providers. These factors can negatively affect health care (50) as well as present challenges for researchers in recruiting and retaining American Indians in clinical studies.

The remaining portion of American Indians live on reservations, where both the tribal network and the availability of health care services are improved compared with urban and nonreservation areas. The IHS has 12 regional offices that coordinate the administration of hospitals, clinics, and contract providers throughout the country. IHS services are typically found on or near reservations, even though they are not limited to those who live there. Although people living on reservations are thus in some ways the most accessible of all American Indians in the United States, several system-related factors compromise the ability to recruit this population. First, it is estimated that less than one-half of all American Indians use IHS services (47). In addition, the recruitment of American Indians living on reservations into cancer research requires lengthy Institutional Review Board approvals that involve both the IHS or tribal facility, a second level of Federal approval, and most importantly, tribal approval. Similarly, publication of research findings requires multiple levels of approvals (15).

One of the key factors discouraging Native American participation in clinical research is distrust of white-dominated institutions. This long-time mistrust began many years ago when the U.S. government did not honor its treaty obligations. Questions regarding the adequacy of health care provided by the government perpetuated these concerns as evidenced by the increased mortality and lagging lifespan expectancies experienced by American Indians until 1955, when the IHS was transferred from the Bureau of the Interior to the Public Health Service. One result is that researchers who want to recruit American Indians for clinical studies must often undergo a strict approval process with tribal officials before project initiation. On a similar note, several tribes have rules regarding cultural appropriation, or the ability to retain certain beliefs and practices as “private property” of the community. Both approval processes may be time-consuming, and researchers must be cognizant of these necessary steps when projecting study timelines.

There are many diverse American Indian and Alaska Native tribes, each with its own history, culture, and beliefs. These beliefs, in turn, affect many lifestyle and behavioral health practices (51). For example, persons of varying tribal affiliations within the American Indian community handle cancer very differently. Although it is openly accepted and treated by all family members in some Southwestern tribes (52), Navajo women indicated that the causes of many deaths in their families were kept secret. Cancer was suspected in these cases, but never talked about (53). Many tribes do not even have a word for cancer in their native language.

Some general beliefs about life, health, and illness among the Indian population are very different from those in Western culture 54, 55. Members of the Western culture, for instance, focus on the individual, whereas American Indian communities emphasize the group or family. Stories and visual tools are often effective means of communication among American Indians, especially those that focus on the family 52, 56. Western culture plans for the future, whereas American Indians live in the present. In turn, American Indians are generally reluctant to dwell on illness and accept it as inevitable (57). Thus, the concept of disease prevention/research is a foreign concept to many in these populations.

Many cultural beliefs have been suggested to affect the use of preventive programs, and these same issues may affect participation in clinical research as well. For instance, fear, fatalism, and embarrassment have been reported as barriers to screening among minority women, including American Indians 51, 56, 57, 58, 59. However, recent studies completed among reservation and urban Southwestern American Indian women did not indicate the presence of significant cultural barriers to women's health care 53, 60. These findings further support the recognition of differences between American Indian communities and highlight the need for site-specific evaluation.

In 1997, African Americans numbered approximately 34.2 million, comprising approximately 12.8% of the total U.S. population. The geographic diversity of African Americans is reflected in census data showing the 10 states with the largest African-American populations as New York (3.2 million); California (2.4 million); Texas (2.4 million); Florida (2.3 million); Georgia (2.1 million); Illinois (1.8 million); North Carolina (1.6 million); and Maryland, Louisiana, and Michigan (1.4 million each).

As is the case with other minorities, poverty is a major structural barrier to participation in clinical research for many African Americans. In 1997, the poverty rate for African Americans averaged 26.5% compared with 11% for whites (61). The percent of African-American families with incomes below poverty level was approximately 31.3%; for individuals, this figure was 33%; (for whites, the percent of families and individuals in poverty are estimated at 9 and 12%, respectively) (62). Median income for African-American households was $25,050, 4% higher than the previous year, yet significantly lower than for whites, $38, 972. (61).

Labor force participation rates vary across educational achievement strata, starting at 49.4% among those with less than a high school diploma, increasing to 75.2% among high school graduates and to 82.5% and 89.5% for persons with less than a bachelors degree and college graduates, respectively (62). In this regard, 36% of African Americans have at least a high school diploma, 24% have 1–3 years of college, and approximately 13% have 4 or more years of college (62).

A related structural barrier is lack of health insurance. Only about 50% of African Americans are covered by private or government health insurance, and approximately 28% are covered by Medicaid (62).

Varied cultural issues can have a significant impact on the clinical trial participation of African Americans. For example, beliefs about disease causation and health maintenance, use of the lay referral network, folk remedies, and the role of the family are all important cultural factors that influence choices for prostate cancer treatment in African-American men 63, 64, 65. As with other populations, heterogeneity within the African-American population is an important consideration (66).

Cultural beliefs and perceptions about cancer can be expected to affect motivation to take part in cancer research. For many African Americans, hopelessness, fatalism, and pessimism undermine feelings of empowerment regarding cancer prevention and control 63, 67. These beliefs relate to the notion that cancer prevention measures are ineffective or nonexistent 68, 76.

Religious beliefs and spirituality also are important in understanding ideas regarding health, disease risk, and treatment. For example, among faith communities, God is considered the central healer of disease in addition to spiritual healers, physicians, and housewives 69, 70. Among breast cancer patients, African-American women had greater reliance on religion as a coping source than their white counterparts (71). Faith communities also provide psychological and physical support, serving as networks for communicating health messages and promoting research participation (72).

The perceived effectiveness of mainstream medicine in preventing, diagnosing, and treating disease also influences participation in medical research. For example, health beliefs regarding hypertension influenced perceptions of the usefulness of treatment and determined adherence and use of folk medicine among African Americans (73). Ignorance of these perceptions can make it difficult for investigators to work effectively within the context of ethnic cultural beliefs, and ultimately undermines study implementation (14).

Symbolic communication and nonverbal communication patterns are as important as language in communication (17). Symbolic communication structures can differ between African Americans and the rest of the population. Linear models of information are most frequently preferred among whites, but a cyclical model of consent and information is important among African Americans (15). In addition, clear, open, honest communication between research staff and African Americans is particularly important in light of negative historical experiences with medical research that has resulted in distrust (74).

Hispanics continue to be one of the fastest growing minority populations in the United States. It is estimated that by the year 2010, the Hispanic population will reach 39 million and will represent approximately 13% of the U.S. population. This growth, which is seven times greater than that of the general population, is due primarily to higher fertility rates and increased immigration (75).

Hispanics are a heterogeneous mixture of cultures, races, and ethnic groups with different economic and educational backgrounds and length of time living in the United States. Most Hispanics living in the United States are from Mexico, followed by Central and South America, Cuba, and Puerto Rico. Hispanics live primarily in large urban settings in eight states: Arizona, California, Colorado, Florida, New Jersey, New Mexico, New York, and Texas. Compared with non-Hispanics, Hispanics are younger; less likely to own their residence; have larger numbers of individuals per household; and have a larger proportion of two-parent families.

Like some other minority populations, Hispanics have problems related to poverty and lower SES. In general, they have lower educational attainment, lower economic status, higher rates of unemployment, and they more frequently lack health insurance and access to adequate health care 76, 77. As a result, Hispanics encounter more barriers to the use of health care services, which reduces the likelihood of their participation in clinical cancer research. Overall, Hispanics are less likely than some other groups to have routine medical examinations, dental care, prenatal care, family planning, Pap smears, and mammography 78, 79, 80, 81, 82, 83. Other common structural barriers to the participation of Hispanics in clinical research include transportation, child care, family/household responsibilities, lack of flexible clinic schedules, fear of not knowing whether they are receiving an experimental treatment or placebo, protocol requirements (number of clinics visits), and gender of the health provider.

Among Hispanics, several cultural characteristics have been found to affect utilization of preventive and medical health services as well as participation in clinical research. These characteristics include the concepts of family (familismo), community, religious/moral values, fatalism, empathy or sympathy, and expectation of personalismo or a more friendly interaction when dealing with health services and health providers (75). Lack of communication with health providers, insensitivity of staff to cultural values and linguistic needs, and lack of proper information about cancer or another medical condition have been reported to decrease the possibility of participating in clinical research (84).

Language has been determined as one of the principal barriers for underutilization of preventive and medical services among Hispanics. Research has shown that quality of care improves when patients have a language-concordant health provider.