Longer Life Foundation Final Report History of Mental Health Treatment and Survival Among Newly Diagnosed Women with Breast Cancer Principal Investigator:

Research has shown an association between psychological status and survival in some cancers. This study examined current distress, history of mental health problems (MHP), and history of mental health treatment (MHT) as predictors of overall survival among newly diagnosed patients with breast cancer. Methods: Eligible patients (N=302) were women, 18 and older, recently diagnosed with stage 0 – IV breast cancer, and had completed psychosocial surveys between May 2000 and October 2003. The survey included self-reported history of MHP and MHT. Survey data were linked with death records from the state departments of health, and with data abstracted from the outpatient medical record. Planned methods included Cox regression models of overall survival among patients with and without MHP and MHT. Models of MHP and MHT were constructed separately. Covariates in each model included stage of disease, comorbidities, treatment adherence, health behaviors, and a composite measure of general distress. Results: Patients were 73.5% Caucasian, with mean age of 53.5 years (SD=11.9), and 84.7% had stage II or earlier disease. Thirty-four patients (11.3%) had died by early 2007. Overall, 20.2% of patients reported a history of MHP, and 24.8% reported a history of MHT. Results for models of MHP and MHT showed that longer survival was associated with higher education, less comorbid disease, earlier stage of disease, and less distress at diagnosis. MHP and MHT both showed patterns in which patients with a positive history showed longer survival. The effect was nonsignificant for MHP, but significant for MHT (p=.016). Subsequent evaluation showed that the number of predictors used in the models was inappropriate for the number of observed events (deaths). Because the primary aims require covariates for control of relevant patient characteristics, revised statistical modeling cannot easily address this limitation. Results should be interpreted with caution. Conclusions: Findings support a link between either mental health treatment or treatment seeking, and breast cancer survival, and are at odds with some existing research. However, findings are subject to important methodological limitations. Lay Summary Research has shown that for some types of cancer, patients who are more psychologically distressed tend not to survive as long. However, little research has been done that examines whether mental health history might be related to survival. This study examined two characteristics of mental health history as predictors of survival among newly diagnosed patients with breast cancer. The two characteristics are history of mental health problems, and history of mental health treatment. Eligible patients included 302 women, 18 and older, who were recently diagnosed with stage 0 – IV breast cancer. All of the patients had completed surveys related to emotional distress and mental health history at the time of their diagnosis. Surveys were completed between May 2000 and October 2003. The survey information was linked with death records from the state departments of health, and with other information drawn from each patient’s medical record. The analysis took several other variables into