Incidence, Risk Factors and Impact on Survival among Cancer Patients Who Develop Mood Disorders: a Nationwide Population-Based Study

Extensive literatures demonstrate that psychosocial stress predispose toward the development of mood disorders after cancer diagnosis. An important meta-analysis concluded that depression predicts cancer mortality, but not cancer progress and suggested that depression should be measured and analyzed at multiple time points, and had ideal length of follow-up to capture a survival effect. Furthermore, the treatment adherence and mood disorders together on cancer mortality are seldom investigated comprehensively. In Taiwan, the availabilities of the National Health Insurance Research Database (NHIRD), Taiwan Cancer Registry (TCR) and Taiwan Death Record allow us to look at the mood disorder problem among cancer patients. We choose colorectal cancer (CRC) in this study because the second most common cancer, average survival, and relatively simple treatment plan. The aims of this study are: (1) to estimate the prevalence/magnitude and to examine the risk factors of mood disorders on different phases (pre-cancer, pre-treatment, during treatment, and after treatment for curative patients; pre-cancer and post-cancer for palliative patients or treatment refusal patients) in patients with CRC. (2) to examine how mood disorders on different phases affect the overall or cancer-specific mortality in patients with CRC. (3) to examine whether patients with cancers have higher prevalence/magnitude of mood disorders on different phases, when compared with the age-, sex-, comorbidity-matched control group. The CRC cohort are those who had newly diagnosed CRC from 2002 to 2011 (TCR). The matched non-cancer control cohort are those without cancers from NHIRD and 1:1 matched with CRC patient by age, sex, comorbidity before the index date, and same diagnosis month of the index date. Study phase is the important time point to look at the variation of mood disorders before cancer, during and after treatment. For the matched non-cancer control group, their study phase of each control individual will be the same as his/her matched CRC individual. It is expected that 140 000 new patients in the CRC cohort and another 140 000 patients in matched non-cancer control group. Mood disorders will be ascertained from NHIRD with the code of ICD-9-CM of depressive disorder, anxiety, bipolar, and prescription of antidepressants and mood stabilizers. Mood disorders will be treated as either continuous or binary variable. Univariate analysis and multivariate analysis will be performed on mood disorders in each study phase. To compare the problem of mood disorders among study phases, generalized estimating equation (GEE) for non-independent data will be used. Survival analysis will be used to look at mood disorders and other factors affect overall or cancer-specific mortality.

Project ID:PC10601-0590
External Project ID:MOST105-2314-B182-025-MY2