Wednesday, March 08, 2017

Evidence Of A Correlation Between Depression and Anxiety and Cancer Mortality

     On January 25, 2017, The BMJ, a respected British scientific journal, published the results of a research study in which the unpublished data from 16 prospective cohort studies initiated between 1994 and 2008 were pooled to examine the role of psychological distress (anxiety and depression) as a potential predictor of site specific cancer mortality. The participants in this research, 163,363 men and women, 16 years of age or older at the time of study induction, make up nationally representative samples drawn from health surveys for England and Scotland. Participants were initially free of a cancer diagnosis, and were asked to fill out questionnaires with items that were designed to capture symptoms of depression and anxiety over the previous four weeks. According to the study's authors, these questionnaires are widely used in population research and have been validated against standardized psychiatric interviews.

     The results of this pooling of unpublished data are interesting. They showed that those with the most psychological distress (depression and anxiety), had a higher rate of death from selected cancers than those with the lowest psychological distress. This was after adjusting for other known risk factors such as adverse health behaviors and reverse causality (it is well known that a diagnosis of cancer can give rise to psychological distress so these authors attempted to control for this by excluding participants with self reported cancer malignancy at study entry). The most consistently robust effects were evident for cancer of the colorectum, prostate, pancreas, esophagus and for leukemia.

     These authors state that their findings could be important in advancing understanding of the role of psychological distress in cancer etiology and cancer progression. They go on to explain that psychological distress could be considered along with other factors such as smoking and obesity to develop an algorithm that would have predictive utility for common cancer presentations such as colorectal, breast and prostate cancer.

     Among the limitations of this study cited by the authors is that the assessment of psychological distress with the instrument that they used, referred to the preceding four week period. A short bout of depression or anxiety is highly unlikely to be relevant to cancer etiology, but the authors state that there is evidence that rates of recurrence are high for psychological distress. Therefore they state that a single administration of a distress inventory seems to capture cases of long term depression and anxiety. Nevertheless, if serial assessments of depression and anxiety had been done over a period of time, that would lend more credence to the findings.

     The results of this study suggest associations between depression and anxiety and cancer mortality, and lend support to the idea that psychological distress could have some predictive capacity for the occurrence and lethality of some types of cancer. It is important to note that further research is required to clarify the extent to which each of the associations between psychological distress and cancer are likely to be causal and to examining the precise mechanisms by which they exert their influence. We would need to investigate exactly how depression and anxiety affect cancer. For example, how much of any effect that we see is related to poor lifestyle choices that a depressed person might make, such as compliance with treatment screenings or treatment regimens. Alternatively, we need to learn how much of the impact of psychological distress is associated with biological mechanisms, such as prolonged inflammatory responses.


Howard Covitz said...

Linda ... Do we know the size of the treatment effect? I have a physician friend who died last year from a Blood Cancer that he fought for 25 years ... he was an upbeat guy ... married to an upbeat lady with upbeat kids. He would tell me and others how such studies felt to him "blaming" ... He knew and I know that they are not intended as such. But the comments that he needed to keep a positive outlook would piss him off. "Hey ... if I'd've had a positive outlook I wouldn't have gotten this Lymphoma." ... Especially, if, indeed, the Tx effect is small ... one has to wonder the value. ??? Just sayin' ... as the kids say. With regard ... Howard Covitz

Linda L. Guerra, Ph.D. said...

Howard: Thanks for your question/comments. Regarding the size of the treatment effect: death rates in the most distressed group (highest occurrence of anxiety and depression) were consistently raised for all types of cancer (multivariable adjusted hazard ratio 1.32, 95%confidence interval 1.18to 1.48. Stepwise associations across the full range of distress scores were observed for colorectal and prostate cancer. That said, all studies of this nature have to be very carefully evaluated. As a psychologist and breast cancer survivor, I felt and still feel that a source of implicit bias in these types of studies comes out of the need for control that all humans have, i.e. we are disposed to wanting to find associations between say mood and cancer occurrence and /or mortality so that we feel we have more control over whether or not we get cancer or if we do get it, how we can increase our chances of surviving it.