And this is why....

...we study tumor size and not metastasis... Tumor size is relatively easy to measure in experimental models and even in real patients. Prevention or inhibition of metastasis however, is very difficult to measure.  It is mainly for this reason (in my view) that we continue to plough most research funding into reducing tumor size rather than preventing migration of the cancer cells to new locations around the body. This is a great article (Why Scientists are so often wrong: the streetlight effect) on why science is often wrong. We look at end-points we can easily measure which may give us, at best, only a small piece of the picture and at worst, irrelevant findings. If we measure the measurable (and it is perfectly understandable why we would want to), then we must remember to re-insert the findings back onto the larger context. This is the value of the systems approach, which we have torn apart over the past 200 years in the name of Enlightenment. An over-emphasis on what we call things has led to an inappropriate level of categorization of data and terms which forces us to place something in one category when it might really belong in two or three or even in none. Medical diagnostic codes are a good example of this forced boxing which is done to allow communication between healthcare providers and insurers. The previous blog about cancer also illustrates the problem to some degree. We say a person has died from breast cancer when in actual fact they died from metastatic liver cancer for example. I have a friend whose disease sadly took this path. We remember her, and walk in her honor, to benefit all victims of breast cancer, but the liver and brain cancer that finally took her life is never mentioned.
Another example is appearing in the medical community at large. The pay-for-performance paradigm for doctors and hospitals depends on care professionals reporting certain elements that they are asked to report. Rewarding certain behaviors will ensure that they are reported more. It doesn't necessary follow that the patients are any healthier for it. For a balanced view of the implementation of 'quality' programs for health care see here. The author calls the programs 'garbage in, garbage out' or in other words, you get what you pay/ask for.  The whole system is forcing certain categorization that can make health data fairly meaningless.  For scientists and physicians alike, this should be cause for concern although, I must admit, there are no easy answers given that the system has evolved this way due to human nature.  Few are comfortable with ambiguities if a more concrete alternative is available, and the system does not allow the time for it anyway.  Perhaps in the future, it will.