Optimum cutoff points for biochemical validation of smoking status.

Selection of cutoff points for tests to validate smoking cessation should take account of the prevalence of deception. When the prevalence of deception is relatively low, the cutoff points to validate quitting should be relatively high. Many studies have used cutoff points that are too low and may have underestimated cessation rates. We present a method for determining the best cutoff points that takes account of the prevalence of deception.