Abstract
Reply
Mosmuller DGM, Mulder FJ, Don Griot JPW. Concerning the article entitled “judgment of nasolabial esthetics in cleft lip and palate is not influenced by overall facial attractiveness”. Cleft Palate Craniofac J. In press.
A “Letter to the Editor” is usually valuable to authors because, first, it demonstrates that the publication attracts attention of the readers and, second, remarks included in such a letter might help improve future studies.
We read the letter by Mosmuller et al. (in press) regarding our recent publication in Cleft Palate-Craniofacial Journal (Kocher et al., 2016) with both interest and disappointment. Our disappointment results from the fact that the claims of Mosmuller et al. (in press) are mostly beliefs and prejudices not supported by facts.
Authors expressed criticism regarding three aspects of the study: (1) design, (2) statistics, and (3) discussion of findings.
First, the study design of our investigation was based on the old principle of experimental work that causation cannot be proved without manipulation. The manipulation in our study was fusion of nasolabial areas of 11-year-olds with UCLP with facial features of peers without the cleft, followed by esthetic evaluation of the nasolabial areas in fused images. In our opinion, the core of the design was relatively simple. A suggestion by Mosmuller et al. (in press) that, this way, “the effect of a cleft on the facial attractiveness was determined instead of the other way around” can be easily declined because raters were asked to focus on evaluation of the nasolabial area. This was clearly explained in the “Subject and Methods” section. Furthermore, the authors imply that selection of photographs used for fused images might have been biased. They ignored, however, all measures undertaken to minimize selection bias, such as participation of different raters for subjects with and without the cleft or the long time between rating sessions. Finally, Mosmuller et al. (in press) proposed an alternative study design that actually was the one used 25 years prior by (Asher-McDade et al. 1991) during development of the esthetic index. We explained in the “Introduction” of our paper why this design is not optimal.
Second, comments regarding our statistical methods are rather surprising because we followed the recommendation of (Mosmuller et al. 2013) who advocated “calculating the interobserver and intraobserver reliability [with ICC] in combination with the Cronbach”. Moreover, we find the remark that “the very high ICC (0.90) in this study needs to be questioned” as completely inappropriate in the light of arguments of Mosmuller et al. (in press). The authors claim that-based on their own experience and earlier publications-“it is very unlikely the interobserver reliability is this high when eight different observers judge 72 photographs with a visual analog scale”. However, three systematic reviews (Al-Omari et al, 2005; Sharma et al, 2012; Mosmuller et al., 2013) showed that no study focusing on assessment of the nasolabial area in subjects with UCLP used visual analogue scale (VAS) and ICC for evaluation of rater reliability. In addition, (Mosmuller et al. 2014, 2015) did not use VAS in their studies; instead, they applied a 5-point Likert-type scale and 200-point numerical scale in their works. Therefore, claims of an excessively high ICC score in our study are entirely unsubstantiated.
Third, regarding our “Discussion”, we offered a possible explanation for why the perception of nasolabial esthetics in UCLP is not influenced by overall facial attractiveness. We understand that one can disagree with our arguments, but the point made by Mosmuller et al. (in press) would be much stronger if supported by any alternative explanation.
In conclusion, we would not like to dissuade Mosmuller et al. (in press) from cropping photographs if they “believe” that this is still necessary. However, based on our results, the use of uncropped photographs of girls and boys with UCLP for esthetic assessment of nasolabial area seems to be justified.