This investigation evaluated the effect of setting cut-off points for treatment allocation upon the reliability of the Index of Orthodontic Treatment Need (IOTN). Cut-off points were placed between the grades signifying ‘need’ and ‘moderate need’ and between ‘need’ and ‘no need’ for treatment. The reliability of each of these dichotomous decisions was evaluated with (i) the Kappa statistic and (ii) the false positive and false negative rates. We found that the reliability was decreased and the false positive rate was increased when the cut-off point was placed between the ‘need’ and ‘moderate need’ for treatment levels. The effect of this variation in reliability is discussed.