Reliability of a Masseter Muscle Prominence Scale and Lower Facial Shape Classification

Primary outcomes

1. Inter-rater Reliability Using a Masseter Muscle Prominence Scale (MMPS)

   Time frame: Day 1

   The MMPS is an ordinal tool to assess the masseter muscle prominence (jaw muscle) for each side of the face from 1=minimal to 5=very marked. Inter-rater (among raters) reliability was calculated separately for the left and right side of the face using Kendall's coefficient of concordance (Kendall's W). Kendall W statistics overall for the left and right sides of the face were derived using the average of assessment 1 and assessment 2 rounded to the nearest whole integer for each subject and each clinician. A total of 8 physicians rated each subject. The degree of agreement of the point estimates of Kendall's W was interpreted according to the reference range scale that was pre-defined as: ≤0: poor, >0 to ≤0.2: slight, >0.2 to ≤0.4: fair, >0.4 to ≤0.6: moderate, >0.6 to ≤0.8: substantial, and >0.8 to ≤1.0: almost perfect. The 95% confidence interval for Kendall's W is provided.

2. Intra-rater Reliability Using a MMPS

   Time frame: Day 1

   The MMPS is an ordinal tool to assess the masseter muscle prominence (jaw muscle) for each side of the face from 1 = minimal to 5 = very marked. Intra-rater (within raters) reliability was calculated separately for the left and right side of the face using weighted Kappa statistics. Weighted Kappa statistics were calculated for each of the 8 physician raters. The overall intra-rater agreement for Kappa statistics for all raters combined was estimated by pooling Kappa statistics for each rater using a chi-square statistic. The degree of agreement of the point estimates of Kappa statistics was interpreted according to the reference range scale that was pre-defined as: ≤0: poor, >0 to ≤0.2: slight, >0.2 to ≤0.4: fair, >0.4 to ≤0.6: moderate, >0.6 to ≤0.8: substantial, and >0.8 to ≤1.0: almost perfect. The 95% confidence interval for Kappa statistics is provided.

3. Inter-rater Reliability Using a Lower Facial Shape Classification (LFSC)

   Time frame: Day 1

   The LFSC is a qualitative tool to assess facial shape into one of 5 categories (A, B, C, D, and E). Inter-rater (among raters) reliability was calculated using Kappa statistics. Kappa statistics were calculated for each of the 5 facial categories. A total of 8 physicians rated each subject. The overall inter-rater agreement for Kappa statistics for all categories combined was estimated by pooling Kappa statistics for each category using a chi-square statistic. The degree of agreement of the point estimates of Kappa statistics was interpreted according to the reference range scale that was pre-defined as: ≤0: poor, >0 to ≤0.2: slight, >0.2 to ≤0.4: fair, >0.4 to ≤0.6: moderate, >0.6 to ≤0.8: substantial, and >0.8 to ≤1.0: almost perfect. The 95% confidence interval for Kappa statistics is provided.

4. Intra-rater Reliability Using a LFSC

   Time frame: Day 1

   The LFSC is a qualitative tool to assess facial shape into one of 5 categories (A, B, C, D, and E). Intra-rater (within raters) reliability was calculated using Kappa statistics. Kappa statistics were calculated for each of the 8 physician raters. The overall intra-rater agreement for Kappa statistics for all raters combined was estimated by pooling Kappa statistics for each rater using a chi-square statistic. The degree of agreement of the point estimates of Kappa statistics was interpreted according to the reference range scale that was pre-defined as: ≤0: poor, >0 to ≤0.2: slight, >0.2 to ≤0.4: fair, >0.4 to ≤0.6: moderate, >0.6 to ≤0.8: substantial, and >0.8 to ≤1.0: almost perfect. The 95% confidence interval for Kappa statistics is provided.