Measurement in clinical vs. biological medicine: the Rasch model as a bridge on a widening gap.

In the dominant Bio-medical paradigm Medicine is mostly Biology applied to Man. Measurement in Biology stems from physical sciences and has established validity. This is not the case for whole-person variables such as behaviors and psychic conditions (disability, pain, knowledges). The very existence of these variables can only be inferred by observing samples of representative behaviors. The quantity of the inferred variable may only come from subjective and discrete counts (scores) of events (coming in a questionnaire). Contemporary statistics demonstrated that raw scores intrinsically lack fundamental properties for scientific measurement, whatever their algebraic manipulations. This adds to the stigmatization of Clinical Medicine as "soft science", compared to Bio-medicine. In the 1960s Georg Rasch inaugurated a new statistical approach allowing transformation of raw scores into objective linear measures comparable to physical measures. This may help the Bio-medical paradigm to redirect resources from laboratory bench back to bedside.