Oral cancer screening usefulness: between true and perceived effectiveness

Screenings, introduced in the 1920s, became rapidly popular in healthcare settings. Chance is an intrinsic screening characteristic. Roughly one half of screened subjects are correctly classified merely by chance, and in high-prevalence settings, even inaccurate screenings detect several diseased individuals by chance, thus appearing effective. The viewpoint is another variable affecting screening perceived effectiveness. For example, public health officers, who seek for mortality rate reductions, look for high sensitivity, which, in turn, is affected by disease prevalence. The relative mortality rate reduction attributable to screening may therefore be significant in high-prevalence areas and irrelevant in low-prevalence areas. This explains why oral cancer visual screening is perceived effective in high-prevalence countries and ineffective in low-prevalence countries. Patients seek for reliable outcomes. Therefore, they require screenings with high positive (PV+) and negative (PV-) predictive values. In high-prevalence areas, PV+ is high, while PV- is low. The reverse occurs in low-prevalence areas. Thus, even for accurate screenings, the perceived effectiveness due to misclassification is low among false-negative patients in high-prevalence areas, who are misclassified as unaffected by the disease, and among false-positive patients in low-prevalence areas, who are subjected to psychophysical sufferings. This article explains the reasons for these and other paradoxes engendered by screening.