Juliana Eugênia Caixeta, Vitor Guo Chen, Carolina Sponchiado Miura
We thank Martinelli and Berretin-Felix for their interest in our Task Force document.Our review screened more than 300 articles, but the journal limits citations to 75, which constrained the works individually referenced.The original protocol by Martinelli et al. was cited under the same criterion applied to all other instruments reviewed. 1The 2016 studies -"Validation of the Lingual Frenulum Protocol for Infants" and "Validity and reliability of the Neonatal Tongue Screening Test"describe the same population in two journals, which raises a legitimate concern of duplicate publication. 2, 3 Regarding the validation process itself, several methodological points justify our position that no available instrument has undergone adequate internal or external validation.First, the NTST validation was retrospective, whereas validation of diagnostic questionnaires should be prospective.Second, no pilot study (typically 15-60 participants) was described.Third, although the population was Brazilian, no formal translation procedureindependent forward translation, reconciliation, and back-translationwas reported.Fourth, internal consistency was not assessed by Cronbach's alpha or any equivalent measure.Fifth, in the intra-observer analysis, only 36 of 100 patients were reassessed and half had undergone frenotomy between evaluations, introducing intervention bias incompatible with reliability testing.Finally, the interobserver analysis was performed indirectly through photographs and videos from the primary examiner's archive, using the Intraclass Correlation Coefficient rather than Cohen's or Fleiss's kappa, the standard for categorical agreement. 4-9These observations were applied uniformly to every instrument reviewed.