Analyzing a Moderating Role of Gender in the Link between Online Health Information Seeking and Colorectal Cancer Knowledge Among Asian Americans

Background and Purpose: Inadequate knowledge of cancer is a substantial barrier to use of cancer screening. Research has indicated a lack of cancer knowledge especially among underrepresented racial and ethnic groups in the U.S., addressing their underutilization of cancer screening. Empirical studies showed that colorectal cancer (CRC) screening outcomes for Korean Americans (KAs) remain suboptimal (20-50%), falling short of the Healthy People 2030 goal (74.4%). A growing body of literature demonstrates that online health information seeking (OHIS) behaviors are associated positively with cancer knowledge. However, little is known about how the association differs by gender for this ethnic group. Thus, this study aims to examine the role of gender in moderating the link between OHIS and CRC screening knowledge among KAs. The research question includes: Does gender significantly moderate the association between OHIS and CRC knowledge among the participants?

Methods: A cross-section study design was employed to collect quantitative data regarding socio-demographics, health-related information (i.e., health status, history of cancer and CRC screening, screening decisional balance), OHIS, and CRC screening knowledge. Purposive sampling was performed to recruit KAs aged 50 and 75 from a metropolitan area in the Southeastern U.S. A total of 421 participants completed a survey questionnaire. Hierarchical linear regression analyses were implemented to investigate if three blocks of variables significantly reduce unexplained variance in CRC screening knowledge. The three blocks of variables were added to the regression model in the following order: [1] sociodemographics and health-related information, [2] gender and OHIS, and [3] an interaction term of gender and OHIS.

Results: The first model of Block [1] accounted for 26.2% of the variance in CRC screening knowledge (F_{[20, 352]} = 6.250, p < .001). Next, the second model of Block [1] + [2] increased the portion of explained variance in CRC screening knowledge (28.5%; F_{[22, 350]} = 6.250, p < .001), with being females (β = -.098, SE = .210, p = .046) and OHIS (β = .117, SE = .047, p = .019), respectively, predicting CRC screening knowledge. The final model of Block [1] + [2] + [3] accounted for 29.3% of the variance in CRC screening knowledge (F_{[23, 349]}=6.297, p < .001). In the final model, the effect of OHIS disappeared after adding the interaction term, while being females (β = -.186, SE = .281, p = .005) and the interaction term (β = .143, SE = .091, p = .044), respectively, predicted CRC screening knowledge.

Conclusions and Implications: This study’s novelty lies in being the first to examine a significant moderating role of gender in the association between OHIS and CRC knowledge among KAs. The findings add to the existing body of knowledge in social work practice and research regarding health education and promotion especially among KAs. The findings can be used to develop gender-specific OHIS-based interventions for reversing the screening outcomes among KAs by advancing their CRC knowledge. Also, the findings suggest that future research should addressing gender disparities in accessing online health information and cancer knowledge for improving CRC screening outcomes.