office (03) 9820 3006 or 920 3008
fax (03) 9820 3007

Step 5 - Sampling


Cultural competence in:

Cultural competence in sampling

Selection of samples

Minority populations studied inappropriately or understudied: Some studies focus purely on ethnic issues when poverty may be more relevant.  Also, ethnic minorities group are under-represented in mainstream research.  Ethnically diverse research teams meeting regularly with communities and with good feedback systems will help to develop more appropriate research questions (Gil & Bob 1999).

“Prior to testing an intervention developed for members of the majority culture or an ethnic minority population, researchers should have adequate theoretical, clinical, practical, or empirical bases for hypothesizing that it will be effective in the particular ethnic group recruited for the study. This includes knowledge of that group’s cultural values, traditions, socioeconomic status, education, or other relevant variables that may facilitate or interfere with responsiveness or resistance to different treatment conditions under study” (Fisher, Hoagwood, Boyce, Duster, Frank, Grisso, Levine, Macklin, Spencer, Takanishi, Trimble, & Zayas 2002;Fisher & Wallace 2000).

Sample bias

Since 1993 guidelines issued by the US National Institutes of Health have mandated the proportionate representation of patients by race and ethnic group in clinical research funded by NIH. To the best of our knowledge, no similar requirement exists in Australia. What is the rationale for fair representation?  Obviously, deliberate exclusion is unethical.  But a common stated rationale is to allow subgroup analysis of mainstream trials based on race/ethnic group.  This is very tricky: race/ethnicity are not salient biological factors and ethnic stratification ‘perpetuates pseudoscientific rationalizations of the fundamentally social concepts of race and ethnic group’ (Chen et al. 2001;Rathore et al. 2000;Rathore & Krumholz 2003).

Examinations were conducted of trials on diabetes, cardiovascular disease, HIV/AIDS and cancer (all recognised ethnic-disparity diseases).  40% did not report race/ethnicity.  46% of the 60% that did used only one or two ethnic categories.  Results were analysed by race/ethnic group in less than 1% of trials.  Mandated inclusion of ethnic groups does not always result in reporting of inclusion and almost never results in analysis by race/ethnicity.  Therefore, the effect on interventions and outcomes is likely to have been negligible, even in disease areas where disparity (inequity?) is recognised (Corbie-Smith et al. 2003).

Corbie-Smith subsequently has described the range of possible interpretations of the phrase "appropriate representation" in relation to the practical application of the National Institutes of Health guidelines on the inclusion of minorities in research. She writes: ‘at least three goals might be reached by including minorities in clinical research: to test specific hypotheses about differences by race and ethnicity; to generate hypotheses about possible differences by race and ethnicity; and to ensure the just distribution of the benefits and burdens of participation in research, regardless of whether there are expected differences in outcome by race or ethnicity.’ (Corbie-Smith, Miller, & Ransohoff 2004).

In a recent review of RCTs in the UK,  Oakley et al., 2003 identified poor standards of inclusivity in health research samples and the negative impact of this on the generalisability of conclusions.

The case of clinical trials

Why are cultural minorities excluded from trials?

Exclusion of African-Americans and Latinos in lung cancer trials (Cabral, Napoles-Springer, Miike, McMillan, Sison, Wrensch, Perez-Stable, & Wiencke 2003). Much of the information about these factors comes from studies of cancer screening because little data are available on the factors that prevent and facilitate participation of minorities in clinical cancer trials specifically (Giuliano et al. 2000). The study design barriers are known (ms-Campbell, Ahaghotu, Gaskins, Dawkins, Smoot, Polk, Gooding, & DeWitty 2004) but the solutions continue to elude.

Murthy reports an exacting study on the racial and other disparities in the representation of racial and ethnic minorities, the elderly, and women in cancer trials sponsored by the National Cancer Institute (Murthy, Krumholz, & Gross 2004). Cross-sectional population-based analysis of all participants in therapeutic non-surgical NCI Clinical Trial Cooperative Group breast, colorectal, lung, and prostate cancer clinical trials in 2000 through 2002.

See issues and pathways to increase representation of diverse groups in clinical trials – Black Americans (Outlaw, Bourjolly, & Barg 2000). Black Americans are stricken disproportionately with cancer. However, they continue to be underrepresented in clinical trials.

What’s being carried forward in US to overcome these disparities?

U.S. National Cancer Institute. Reducing Cancer-Related Health Disparities. 

Special Populations Networks, for Cancer Awareness Research and Training that build relationships with community-based programs, foster cancer awareness activities, increase minority enrolment in clinical trials, pilot projects that will lead to the development of grant applications for new and innovative research, and develop junior biomedical researchers from minority and underserved communities.

See U.S. Radiation Oncology Program, Cancer Disparities Research Partnerships Program (CDRP) which will support the expansion of radiation oncology clinical trials in geographically dispersed and ethnic and culturally diverse institutions serving large numbers of Native Americans, African Americans, Hispanics and rural Appalachians. CDRP also is expanding dissemination and diffusion channels using novel telemedicine and teleconferencing architecture to connect CDRP and participating institutions. 

National Cancer Institute Initiatives and Resources Related to Minority Cancer Awareness, at