Statistical versus fuzzy measures of variable interaction in patients withstroke

Evidence-based medicine, founded in probability-based statistics, applies w hat is the case for the collective to the individual patient. An intuitive approach, however, would define structure in the (physiologic) system of in terest, the human being, directly relevant to other systems (patients) comp osed of similar variables. A difference in measure of variable interaction in the patient from that in the collective would show how extrapolation of information from the latter to the single patient is counterintuitive. Meth ods: We compare statistical to 'fuzzy' measures of variable interaction. Th ree diagnostic variables are considered in 30 stroke patients who underwent the same diagnostic tests. 'Fit' (fuzzy information) values [0, 1] for deg ree of variable severity were expertly assigned by 2 blinded raters for rea l and fabricated patients. Fabricated patients were composed of real patien t 'fit' values after shuffling. Real and fabricated patients were each nume rically represented as a set {blood, heart, vessel}. Three groups of fabric ated patients and the real patient group were studied. Statistical [Pearson 's product-moment (regression analysis) and Spearman's rank correlation] an d three different fuzzy strong in real patients, and weak after one shuffle , using all fuzzy measures. By comparison, the same interaction was found i n real patients by only 1 rater (Rater 2) using 1 statistical technique (Sp earman's rank correlation) which, as did Pearson product-moment correlation , found a 'significant' interaction between blood-heart in fabricated patie nts. Conclusion: Our study suggests that the measure of variable interactio n in nature - as combined in the individual treat) patient - is captured ro bustly by fuzzy measures and not so by standard statistical measures. Copyr ight (C) 2001 S. Karger AG, Basel.