Influence of examiner differences on KIG-classification when assessing malocclusions.

Objective: Indication systems such as the German KIG system (Kieferorthopädische Indikationsgruppen = Orthodontic Indication Groups) presuppose the objective assessment of underlying malocclusions. In this survey, we aimed to investigate the degree of agreement among the findings of several examiners in the assessment of different malocclusions and their classification according to the KIG system.

Subjects And Methods: Calibrated examiners assessed in the clinical evaluation and on plaster models orthodontic malocclusions in 180 adults (aged 20-49, 64 male, 116 female) from the population-based Study of Health in Pomerania (SHIP). Clinical examination was carried out by an experienced orthodontist, and the plaster models were also analysed by an examiner experienced in orthodontics. To compare inter- and intra-individual model examiners, we had two examiners with differing orthodontic experience carry out additional analyses of 60 of the 180 models (29 male, 31 female).

Results: The examiner differences yielded various KIG classifications and hence different assessments (i. e., whether KIG case-costs should be borne by health insurance). The comparison "clinical examination versus model analysis" revealed differences regarding 16.7% of the study participants in the assessment of whether the expense would be borne by the statutory health insurance fund. At 5.0-8.3%, the number of participants whose assessments had differed was much smaller in the inter-individual comparison of model-examiners and was smallest (at 3.3-6.7%) when comparison was made between intra-individual assessments (by a sole examiner). With regard to overall malocclusion assessment, the greatest examiner differences were again revealed when comparing the clinical examination with the model analysis (median kappa 0.57). The model-examiner comparison revealed larger differences among examiners with less orthodontic experience (median kappa 0.61 and 0.62) than the comparison between examiners with orthodontic experience (median kappa 0.70).

Conclusions: There can occasionally be considerable examiner differences in the classification of participants according to orthodontic indication groups and hence varying assessments of whether such persons are KIG cases or not. Various means of data collection (clinical evaluation-plaster models) in the assessment of malocclusions by multiple examiners and by those with little orthodontic experience may negatively influence agreement among examiners.