Striking examples of learning from the failures of
a series of cause and effect correlations in treatments for
human disease and disorders are given in Wired,
January 2012, p. 102 - 109, by Jonah Lehrer.
‘Trials and Errors: Dead end experiments, useless
drugs, unnecessary surgery. Why science is failing us.’
Lehrer writes about unintended consequences in:
- cholesterol lowering torcetrapib (Pfizer)
- B vitamins reducing homocysteine-linked cardiovascular
Situations from more information from tests not producing
expected outcomes in:
- non specific back pain caused by swelling or degenerated
discs, revealed by MRI
Lehrer offers these as breakdowns in scientific correlation.
However, I offer that the ‘business perspective’ of correlation
and the ’science perspective’ of correlation are different.
Harford states this more clearly in his book, ‘Adapt’.
Harford’s arguments in Adapt that scientific experts are
continuously humbled by what they predict or believe and
the truth or outcome. Lehrer, while a very compelling writer,
might be overstating the case made by linking a series of
correlations to the development of a business outcome.
There are things to learn though in Lehrer’s article about the
industry and approaches to meet customers’ needs.
Lehrer provides examples where a change in R&D
approach is what job seekers should consider in
companies to work for. The business paradigm has
limited validity in complex scientific correlations such
are used in traditional pharmaceutic development.