Moderating Effects of Poverty Between Interactions Via Social Network and Annual Mammography

To eliminate disparities in cancer incidence and mortality, women aged 40 to 64 in low-income household have been eligible for free/low-cost breast screening services since 1991. While these efforts have reduced the gaps due to race and ethnicity, the disparities in mammography use due to poverty have persisted over the decades (Wharam et al., 2014). The Internet has become a venue for social relations since its inception by connecting individuals without boundaries through synchronous (e.g., chat systems on blogs, message boards, and forums) and asynchronous (e.g., email, discussion boards, and social networking sites) communication channels (Pfeil, 2009).

Social network theories (DiMaggio & Garip, 2012) also suggest that there will interaction (=moderating) effects of social network with personal characteristics (e.g. poverty status). However, no research has examined the interaction effects between online network interaction (ONI) and poverty on annual mammography. To fill the gap, the research hypothesis this study examined is:

• There are moderating effects of poverty between ONI and annual mammography use among women (≥40).

Methods: This study analyzed the 2005 Health Information National Trends Survey data on U.S. women who were never diagnosed with breast cancer (unweighted N=2,639) after multiple imputations for missing data. After centering of ONI, interaction effects of poverty between ONI and annual mammography were estimated in multivariate logistic regression models employing the PROC SURVEY-procedures.

Results: Women in poverty were less likely by 19.1 percentage points than women not in poverty to use mammography (44.7% vs. 63.8%, p<0.001). The main-effect model shows that there are no main effects of ONI (OR=1.01, p=0.655), but significant main effects of poverty (OR=0.57, p<0.001), while controlling for the effects of covariates including socio-demographics, family cancer history, and health insurance. The interaction-effect model reports significant interaction effects of poverty (OR=0.80, p<0.01) between ONI and annual mammography. Further analysis for the size of the interaction effects reported that the relationship between ONI was not significant, but positively, for women not in poverty (OR=1.03), but significant, but negatively, for women in poverty (OR=0.82, p=0.011).

Implications: This study found that the effects of ONI on mammography use were moderated by poverty status (negatively related among women in poverty but positively related among their counterparts). The findings suggest that ONI will help women not in poverty to be empowered in health; however, such activities may cause poor women to compromise their health. Poor women tend to depend on ONI for their health information because healthcare use is expensive in terms of time and money (Jerome-D’Emilia & Chittams, 2014). However, without follow-up consultations with healthcare professionals, they may be misguided by false self-diagnoses and, thus, delay seeking medical care.

The strengths of this study include the use of a nationally representative sample, which contributes to the development of the knowledge on this topic and, accordingly, provides useful information to policy makers and researchers in planning community-based intervention. Still, further studies are needed to examine the mechanisms how ONI affects the psychological and behavioral aspects of women in poverty to be discouraged to have annual mammography.