Lambda

What wavelengths of light skin absorbs and reflects is something I've come to consider as useless, as it only rests on society's shared assumptions as to its implications. It has no meaning beyond what someone mentally attaches to it. Which is why a recent New England Journal of Medicine article irritates. From the relevent full text of the article:

Inclusion and Exclusion Criteria Patients 18 years of age or older, self-identified as black (defined as of African descent),...

Basically, the authors used a phenotypic proxy (race) for pharmocologic responsiveness that has nothing to do with the selected phenotype. They assume that this one phenotype denotes a shared genotype (DNA background). Fortunately for the authors, their result reached statistical signficance, and the racial angle guaranteed them a prominent publication. The authors acknowledge the questionable utility later on in the paper:

A future strategy would be to identify genotypic and phenotypic characteristics that would transcend racial or ethnic categories to identify a population with heart failure in which there is an increased likelihood of a favorable response to such therapy.

One question that comes into play with evidence-based medicine is "Does this paper apply to my patients?" I don't think you can say that with a high degree of confidence for you individual patient. Does the same imperfect skin color phenotype-heart failure phenotype hold true for ibo, zulu, rwandas, ethiopians, much less the descendants of slaves after many generations of admixture with other american settlers? The absolute risk reduction for deaths from any cause is only 4%. The number of patients you need to treat to save one patient from death of any cause is 25. Until a molecular mechanism can be identified and tested in patients, I would hesitate to automatically follow the paper's recommendations, as you may be doing the other 24 patients a disservice.