Abstract PL02-04: Considerations of culture in behavior change interventions

Increasing access to and participation in cancer early detection and screening programs can contribute to substantial progress in reducing cancer morbidity and mortality burden experienced by underserved and/or ethnic minority communities. Regular screening facilitates early diagnosis of cancer, which, in turn, contributes to the reduction of morbidity and mortality. Colorectal cancer (CRC) represents an important focus for reducing cancer health disparities due to the widespread availability of effective early detection screening methods. Despite the known benefits of early detection, screening rates across the U.S. remain sub-optimal. CRC incidence presents a paradox in that about 75% of CRC is diagnosed in persons with no known risk factors. About 90% of CRC is diagnosed in people aged 50 years or older. Incidence and mortality rates for CRC differ substantially by race and ethnicity with African American men having the highest mortality in the U.S. These facts together suggest the need for broad-based efforts to increase screening among older minority groups. In this presentation, we will present an integrated model of health behavior change that incorporates cultural variables relevant to various minority groups such as African Americans, Korean Americans, and South Asians. Results of both quantitative and qualitative research studies will be detailed.

Study 1. Purpose: African American men are at the highest risk for mortality from CRC, and, correspondingly, African American men also have very low screening rates. Using latent class analysis, we identified groups of African American men at high and low behavioral risk for screening nonadherence.

Methods: Data were from baseline interviews of a 5 year, RCT designed to increase CRC screening.

Sample: A subset of 260 African American men, with a mean age of 56.2; 45% reported some college education, 33% were single, and 82% were not employed.

Measures: Validated measures on CRC screening related self-efficacy (SE), benefits and barriers, knowledge, perceived risk, trust in provider, and screening history were used based on the Health Belief Model.

Analysis: We fit LCA models to summarize behavioral risk and protective factors (high knowledge, SE, benefits, risk and trust, and low barriers) related to past use of screening tests, and LCA regression models to assess the significance of behavioral risk group and screening history.

Results: Risk groups were: high, moderate high, moderate low, or low risk for nonadherence with screening. Four LCA models provided the best fit across the three screening tests (FOBT, sigmoidoscopy and Colonoscopy). African American men at highest risk for not being screened by colonoscopy (34%) had low knowledge, SE, benefits, risk, and trust, and high barriers. For stool blood tests, 42% of men fell into high risk for not having the test. In contrast, only 10% of men were at high risk for sigmoidoscopy nonadherence; the majority were at moderate risk (55%). Latent class structure (controlling for age) was related only to endoscopy screening. Men nonadherent with colonoscopy screening were 8 times more likely to be in a high risk group compared to moderate or low risk groups (p<.05). Similarly, for sigmoidoscopy, men who have never been screened were 2.8 times less likely to be in the high risk group (p<.05).

Conclusions: There is potential for clustering individuals into behavioral risk groups and targeting interventions relevant to their risk for not getting screened. African American men at higher risk for not getting screened may need interventions focused on addressing the relevant beliefs that would move them into lower risk groups.

Studies 2, 3, and 4 are qualitative studies examining cultural differences with regard to beliefs and knowledge. For all three studies:

Purpose: Identify knowledge and perceptions related to CRC screening among Korean Americans, Latinos/Latinas, and South Asians.

Samples: Men and women, aged 50 or older, who self-identified as Latino/Latina, South Asian, or Korean American.

Methods: Focus group discussions were conducted guided by semi-structured interview guides. The Latino/a and South Asian groups were separated by gender.

Results: While many beliefs, especially barriers, were common to both the majority and minority groups, several barriers related to patient provider communication, medical mistrust, and cultural sensitivity were recurrent themes. Results underscore the need to integrate culturally appropriate messaging into educational programs on CRC screening.

Examples of culturally sensitive tailored interventions designed to increase CRC screening and the pilot testing will also be described. Strategies to incorporate linguistic and socio-cultural sensitivity into community-based interventions will also be discussed.