The influence of measurement precision on medical findings

Medical researchers, epidemiologists and clinicians are all concerned with the observed incidence rates of various diseases and diagnostic findings. Such findings are necessarily based on imprecise procedures and instrumentation. The present study describes the theoretical extent to which the precision of instruments and procedures contributes to reported distribution curves and incidence rates of diagnostic findings. A numerical method is described for separating underlying distribution curves from the observed distribution curves which are distorted by measurement error. The method is illustrated by two examples: blood pressure measurements and QRS axis distributions in the ECG.