Background: Women with larger social networks generally have better breast cancer survival. However, previous work has been conducted predominantly in white women. Therefore, we evaluated associations of social networks, social support, and breast cancer-specific and overall mortality in a large, Northern California cohort of women with breast cancer, evaluating whether associations differed in white and non-white women.
Methods: This study included 4,505 women from the Pathways Study, a prospective cohort study examining predictors of outcomes in women diagnosed from 2006-2013 with stages I-IV breast cancer in Kaiser Permanente Northern California. Participants responded to questions on social ties (marital/intimate partner, close friends and relatives, community and religious participation) and to the Medical Outcomes Study social support survey approximately two months following diagnosis. We used proportional hazards regression to evaluate associations between social networks, social support and recurrence, breast cancer-specific mortality, and overall mortality. We further stratified by white vs. non-white race since the number of outcomes were too small in specific non-white racial/ethnic groups, which included African-American (AA), Asian/Pacific Islander (API), and Hispanic women. We then evaluated evidence of effect modification by this variable.
Results: The cohort included 2,894 white and 1,611 non-white women. In multivariable-adjusted analyses of the full cohort, adjusted for sociodemographic characteristics, disease severity, treatment, and presence of comorbidity, women with small social networks had higher breast cancer-specific and overall mortality. However, associations differed substantially by race/ethnicity (p-value, test for interaction=0.02 for recurrence, p=0.07 for breast-cancer specific mortality, and p<0.001 for overall mortality). In white women, socially isolated women had higher breast cancer-specific (hazard ratio (HR)=2.41, 95% confidence interval (CI): 1.30-4.49, p-trend<0.001) and overall (HR=2.98, 95% CI: 1.94-4.58) mortality compared with those who were socially integrated. These associations were driven by the strong associations seen in white women with hormone-receptor positive cancer. Associations with recurrence were nonsignificant. By contrast, in non-white women, those who were moderately integrated had a significantly lower risk of recurrence (HR=0.68, 95% CI: 0.47-0.98) than those who were socially integrated. Associations with breast cancer-specific and overall mortality were nonsignificant but similar qualitatively. In additional exploration, larger social networks appeared beneficial in white and AA women but we noted possible direct associations between social network size and outcomes in API and Hispanic women. Though each of the different types of ties contributed to associations in white women, a larger number of close friends and relatives was related to worse outcomes in the non-white women in this sample. Social support was not significantly related to outcomes in either white or non-white women.
Conclusions: In white women, larger social networks were related to better survival, whereas among non-white women, those with moderate-sized networks had the lowest risks of recurrence and breast cancer-specific mortality. The influence of social networks on breast cancer outcomes may differ by race/ethnicity.
Citation Format: Candyce Kroenke, Scarlett Lin Gomez, Marilyn Kwan, Lawrence Kushi. Social networks at diagnosis and recurrence, breast cancer-specific mortality, and overall mortality in white and non-white women in the Pathways Study [abstract]. In: Proceedings of the Tenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2017 Sep 25-28; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2018;27(7 Suppl):Abstract nr C29.