Jump to navigation
Identifying typologies of breast cancer patients based on multiple individual and contextual factors for timely treatment initiation
Baik, Sharon Hyun
Malcarne, Vanessa L.
Wells, Kristen J.
McDonald, Carrie R.
Sadler, Georgia Robbins
Roesch, Scott C.
Rationale: Breast cancer is the most commonly diagnosed cancer, excluding skin cancers, and is the second leading cause of cancer death among women in the United States. Despite advancements in screening, early detection, and cancer treatments, not all women have benefited equally. Racial and ethnic minorities, particularly African American women, and those of low income have higher breast cancer mortality rates compared to the general population. Previous research has identified a number of demographic (e.g., race/ethnicity, age, health insurance, income), medical (e.g., comorbidities with other illnesses, family medical history), environmental (e.g., geographic area), and health system (e.g., type of cancer-related services available) factors associated with breast cancer disparities. However, these factors have largely been examined individually, and no study has comprehensively evaluated how multiple individual and contextual factors impact breast cancer outcomes. Therefore, this dissertation project had two primary aims: 1) to identify distinct subgroups of breast cancer patients based on demographic, medical, environmental, and health system factors that have been shown to influence timeliness of breast cancer care, and 2) to examine differences among emergent classes in timely initiation of breast cancer treatment. Design: The proposed study used archival data from the control arm of the Patient Navigation Research Project (PNRP), a five-year 10-site clinical trial of adult patients from medically underserved populations with an abnormal cancer screening or a new diagnosis of breast, cervical, colorectal, or prostate cancer. For this study, the sample included 198 patients with newly diagnosed Stage I-III breast cancer who received usual standard of care (control arm) from four PNRP sites, and who received a treatment for breast cancer (e.g., surgery, chemotherapy, radiation, hormonal therapy). Control participants were primarily recruited via medical record abstraction for which informed consent was waived. Exploratory Latent Class Analysis (LCA) was used to identify subgroups of breast cancer patients based on demographic (race/ethnicity, age at diagnosis, health insurance status, annual household income), medical (comorbidities [Charlson Comorbidity Index], family history of cancer), environmental (geographic residence [urban vs. rural], and health system (cancer-related services available onsite) factors associated with timeliness of breast cancer care. For the second aim, the study conducted logistic regression analyses to examine if class membership significantly predicted timely breast cancer treatment initiation, defined as initiation of any treatment for breast cancer (e.g., surgery, chemotherapy, radiation, hormonal therapy) within 30 or 60 days of diagnosis, controlling for type of breast cancer treatment. Results: Three classes of breast cancer patients were identified with varying patterns of patient demographic, medical, and health system characteristics. The first class was distinguished by its high endorsement of indicators associated with timely breast cancer care; patients in this class were most likely to be White, have private health insurance, and have a family history of cancer. The second class was characterized by individual and contextual factors associated with treatment delays, including having public health insurance, not having a family history of cancer, and receiving care at a facility with the least amount of breast cancer services available onsite. The third class represented breast cancer patients with the oldest average age at diagnosis and the greatest number of medical comorbidities. Binomial logistic regression analyses demonstrated that the emergent classes did not significantly differ in the likelihood of initiating breast cancer treatment within 30 days or 60 days from breast cancer diagnosis, controlling for type of treatment. Conclusions: The present study used LCA to derive classes of breast cancer patients based on simultaneous evaluation of demographic, medical, environmental, and health system factors associated with timely breast cancer care. However, the emergent classes did not significantly differ in terms of timely initiation of breast cancer treatment following definitive diagnosis of breast cancer. The relatively small and homogenous study sample may have obscured differences in timeliness of breast cancer treatment initiation. Future studies should utilize LCA with larger, more diverse samples of breast cancer patients to identify distinct classes with unique combinations of individual and contextual characteristics that influence timeliness of breast cancer care. Identification of distinct typologies of breast cancer patients provides a deeper understanding of how the combination of factors synergistically impacts breast cancer outcomes and can help target interventions to specific subgroups of patients that are most likely to experience delays in breast cancer care.
Doctor of Philosophy (Ph.D.) University of California, San Diego and San Diego State University, 2018
© 2015 SDSU Library & Information Access. All Rights Reserved.