The ROC curve was constructed by plotting the sensitivity on the ordinate against 1-specificity on the abscissa.
The cutoff point was then selected on this graph from the values in the upper left-hand corner of the curve, seeking a good balance between sensitivity and specificity.
The area under the curve allowed estimation of the ability of the questionnaire to diagnose CVS; poor questionnaire performance would be indicated by a curve near the diagonal and, thus, a small area, and vice versa.