Diagnosis, treatment, and survival among older women with breast cancer: the effect of pre-existing mental illness.

Abstract

BackgroundPeople with a mental illness die 8-25 years earlier than people without a mental illness. Among women, decreased life expectancy is partially due to higher mortality from breast cancer; women with mental illnesses experience up to 367% increased hazards of mortality after breast cancers compared to women without mental illnesses. Yet, with timely detection and appropriate treatment breast cancer is very survivable. ObjectivesThis research focuses on how a pre-existing diagnosis of mental illness affects breast cancer outcomes across the cancer continuum, focusing on differences in stage of diagnosis, receipt of guideline-concordant care, and survival. I also examine whether outcomes vary by type and severity of mental illness. Finally, I evaluate whether the impact of mental illness varies by race, arguing that having marginalized identities associated with multiple forms of oppression may worsen outcomes. MethodsThis project uses the linked Surveillance and Epidemiology and End Results (SEER) cancer registry and Medicare claims (SEER-Medicare), a data national source that captures a significant proportion of breast cancer cases nationally among women over 65 years old. Mental illness is defined and categorized as having a diagnosis of serious mental illness (schizophrenia, bipolar disorder, depression with psychosis, and other psychotic disorders); depression or anxiety; and other mental illnesses excluding cognitive disabilities, in the two years prior to cancer diagnosis. Outcomes examined include AJCC staging, receipt of NCCN guideline-concordant care, focusing on surgery, radiation, and chemotherapy, and months of survival. A total of 96,034 women were included, though sample sizes vary depending on the analysis. ResultsPre-existing mental illnesses are common for women diagnosed with breast cancer. Overall, I found that 28.6% of women in this study had at least one diagnosis of a mental illness in the two years prior to their breast cancer diagnosis and 1.7 % had a serious mental illness. Women with SMI are more likely to be diagnosed with Stage II or III breast cancers than women without mental illnesses and women with depression or anxiety are less likely to be diagnosed with Stage IV cancers. In terms of treatment, I find that among women with AJCC stage I-III cancer, where either a mastectomy or breast conserving surgery and a full course of radiation is recommended, women with SMI less likely to receive surgery than women without. Women with mental illnesses (regardless of type) have a higher risk of not completing radiation after breast conserving surgery. Among women with AJCC stage I-III triple negative or HER2+ cancers, women with SMI are more likely to not complete surgery, radiation, and chemotherapy. Finally, I find that women with mental illnesses (regardless of type) have higher risks for mortality after cancer from both breast cancers and other causes, and that this risk is not entirely explained by stage of diagnosis and treatment received. Differences in survival are greatest for persons with SMI. I also investigated whether the effects of mental illness on cancer outcomes vary by race/ethnicity and find evidence that Black and Hispanic women may have especially high risks for later staging of breast cancer, incomplete treatment, and higher mortality. ConclusionsBreast cancer and mental illness are both common illnesses facing older women in the United States. This work provides new insights into issues of equity around breast cancer outcomes for older women with mental illnesses and provides entry points for considering interventions that will improve outcomes. This work also provides information for areas where inequities among Black and Hispanic women with mental illnesses experiencing breast cancer should be investigated further.