The Development of the Index of Complexity, Outcome and Need (ICON)

Abstract

This paper is based on the winning submission for the 1998 Chapman prize awarded by the British Orthodontic Society for an essay on a subject promoting the interests of orthodontics. The aim of the investigation is to develop a single index for assessing treatment inputs and outcomes. An international panel of 97 orthodontists gave subjective judgements on the need for treatment, treatment complexity, treatment improvement, and acceptability on a diverse sample of 240 initial and 98 treated study models. The occlusal traits in the study models were scored according to a defined numerical protocol. Five highly predictive occlusal traits were identified (IOTN Aesthetic Component, crossbite, upper arch crowding/ spacing, buccal segment antero-posterior relationships, and anterior vertical relationship) and then used to ‘predict’ the panellist's decisions using regression analysis. Cut-off values were determined for the dichotomous judgements by plotting specificity sensitivity and overall accuracy. Twenty percentile ranges were used to determine 5 grades of complexity and improvement. The index prediction of decisions for treatment need, had specificity 84•4 per cent, sensitivity 85•2 per cent, and overall accuracy 85 per cent. When used to predict treatment outcomes, the new index had specificity 64•8 per cent, sensitivity 70•1 per cent, and overall accuracy 68•1 per cent. The index could explain 75•6 per cent of the variance in the mean casewise complexity score and 63•5 per cent of the mean casewise improvement score. A new orthodontic index is proposed to assess treatment need, complexity, and outcome. It is based on international orthodontic opinion.

Keywords:

• Icon
• Occlusal Index
• Orthodontics

Acknowledgments

We would like to acknowledge the assistance of our academic colleagues, Professor Berg (Germany), Professor Adamidis (Greece), Professor Rehák (Hungary), Professor Miotti (Italy), Professor Prahl-Andersen (the Netherlands), Professor Stenvik (Norway), Professor Canut (Spain), and Professor Proffit (USA), who facilitated this study, and our professional colleagues who took part. We are grateful for the financial support of the European Union and The University of Wales College of Medicine, and Dr Frank Dunstan of University of Wales, Medical Computing Department for statistical advice.