The aim of the present investigation was to perform the content and construct validation of the diagnostic criteria for TMD (DC/TMD) Axis I, for children 7 to 11 years old. A Delphi process was used to perfom the content validity of the DC/TMD Axis I. 189 7-11 years old children were assessed with the adapted instrument. Confirmatory factor analysis (CFA) was used to investigate construct validity of the DC/TMD for children. A baseline one-factor model was compared against a two-factor (Model 2) and a seven-factor (Model 3) models based on the original DC/TMD. Root-mean-squared error of approximation (RMSEA), comparative fit index (CFI), chi-square, change in chi-square and Cronbach's Alpha were used to analyze the data. All analysis were performed in STATA© version 13.0.

This research was evaluated and approved by the ethics committee of Universidad CES (File 94, Project Code 584). Subjects signed a written assent and their parents signed written informed consent to participate in the study.

Design: This was a prospective study to perform the content and construct validation of the DC/TMD Axis I for 7-11 year old children.

Population and sample

The target population of the study was children between 7 and 11 years of age of both genders, that were recruited in the specialized orofacial pain clinic at Universidad CES. The sample was recruited by convenience from February to November 2017, after having ready the Delphi process (June 2015-May 2016) and the adaptation of the instrument for children (June 2016- December 2016).

According to Colombian laws and policies for research, participants were evaluated by a psychologist, using a standardized instrument (CES-ERIL), to assure that their reasoning and comprehension abilities, were enough to understand the questions in the instrument.

Examiner calibration to the original DC/TMD Axis I

The investigator who carried out the clinical assessment of participants was first calibrated on the use of the original Spanish translation of the DC/TMD Axis I on adults. The calibration process was performed by three members of INfORM and strictly followed the protocol (https://ubwp.buffalo.edu/rdc-tmdinternational/wp-content/uploads/sites/58/2017/01/DCTMD-Training-and-Calibration-160401.pdf).

The primary statistic for assessing performance (i.e., case classification) was Kappa. Comparisons of interest included pair-wise between the clinical examiner (Suarez N) and the Reference Standard Examiner (Restrepo C, who is a calibrated member of INfORM) and an overall Kappa for all examinations (Kappa=0.82).

Content validity

The original DC/TMD axis I9 was applied to 25 children, who were accompanied by their parents. The children were different from children participating in the study. The exam was performed in the presence of the children´s parents, who confirmed the answers to the symptoms questionnaires. Appreciations from the examiners, children and their parents were recorded, regarding the understanding of the questions in the symptoms questionnaire and the language used to give instructions and ask about pain in the clinical exam.

Clinical information and results from the exploratory study were available to the performers of the adapted DC/TMD Axis I for children, as a basis for the construction of the new instrument.

The content validity of the DC/TMD was performed through a Delphi process described below.

Delphi process

Delphi technique was described by one of its originators as 'a method of eliciting and refining group judgments' and its purpose is to find consensus of experts around a certain situation or research. The Delphi method in this investigation was performed following the process of a previous investigations in TMD.

The subjects of the Delphi study were 3 parents of children with TMD (whose children were evaluated with the original DC/TMD Axis I in the exploratory study), 2 parents of children without TMD (whose children were evaluated with the original DC/TMD in the exploratory study), one pediatrician expert in neurodevelopment; two psychologists, one that works with children with TMD at clinics of Universidad CES and another expert in psychometrics; one physical therapist, two general dental practitioners, two pediatric dentists and one maxillofacial surgeon who were presently involved in TMD patient care and were associated with the Pain clinic at Universidad CES. The group was called in this investigation as "panelists".

The panelists were invited to participate directly and personally by the investigators. By email, the panelists received a detailed explanation about the study and an attached excel sheet that contained the whole questions and items of the clinical examination contained in the original DC/TMD Axis I as rows. In the first round of the Delphi study, panelists were requested to fill in three columns with the following three questions:

For questions contained in the DC/TMD Axis I symptoms questionnaire:

1. How understandable do you think this question is for children 7-11 years old on an 11-point rating scale from 0 "completely not understandable" to 10 "extremely understandable"?
2. Should this question be included in an adapted DC/TMD axis I (0 - no, 1 - yes)?
3. How confident are you with your recommendation about the inclusion of the question on a 11-point rating scale from 0 "not confident at all" to 10 "extremely confident"?
4. Do you have any suggestion on the language and/or number of questions the instrument should include? Why?

For the items included in the DC/TMD Axis I clinical examination

1. How important to you are the following items for TMD diagnoses in 7-11 year-old children on an 11-point rating scale from 0 'completely unimportant' to 10 'extremely important'?
2. Should the item be included in a DC/TMD clinical examination (0 - no, 1 - yes)?
3. How confident are you with your recommendation about the items to be included in the adapted DC/TMD Axis I for children on a 11-point rating scale from 0 'not confident at all' to 10 'extremely confident'? The panelists should fill in the requested information and sent the excel file back by email. In the second round of the Delphi process, panelists received medians and interquartile ranges for the "understandable" and "confidence" ratings of the symptoms questionnaire (question 1 and 3) and for the "importance" and "confidence" ratings of the DC/TMD Axis I clinical examination (questions 1 and 3); added as extra columns in the excel sheet. For question 2, the proportion of positive answers, i.e., the per cent of panelists recommending the inclusion both of questions in the symptoms questionnaire and of items in the DC/TMD Axis I clinical examination, was given. Provided with the findings from the first Delphi round, the panelists answered again the three questions. This process asking panelists to score the questions again while considering the findings from the previous round was continued until a priori set criterion for consensus among the panelists was reached. Criterion measure for a consensus, was set using Cronbach's alpha as a measure of "internal consistency of the panel". An alpha >90 was used to stop the consensus process.

As a result of the Delphi process, an adaption of the early official Spanish translation of the DC/TMD Axis I was performed in order to use the instrument in 7 to 11 years-old children. All children answered the questions in spanish.

In the clinical examination, pressure applied to the muscles and TMJ, the Palpeter® (BUTLER®. Etoy, Switzerland), was used to standardize the pressure to the muscles. To determine the best force to be applied to the masticatory muscles in children, different pressures were tested according to previous literature in children. The pressures used in the original DC/TMD Axis I were evaluated as well.

Each child answered the symptoms questionnaire and was evaluated with the clinical examination form of the adapted DC/TMD Axis I fo 7-11 year-old children. All assessments were obtained by the same calibrated investigator.

Construct validation

It was performed a construct validity using CFA. Factors were initially defined according to the diagnosis tree of the original DC/TMD Axis I9 (Table 1).

Pearson correlation coefficients among the items of the DC/TMD Axis I were obtained. Three fit indices were used to test the best CFA model that could represent the adapted DC/TMD Axis I for children: root-mean-squared error of approximation (RMSEA), comparative fit index (CFI), chi-square, and change in chi-square, according to the change in degrees of freedom between models. Good models have RMSEA values that are 0.08 or less and a CFI greater than 0.90 is considered to be an indicator of a good fitting model. When comparing models, a lower chi-square value indicates a better fit, when an equal number of degrees of freedom is present.

Based upon the hypothetical underlying constructs for the DC/TMD Axis I, three models were developed to represent the best fit for the overall data. Model 1 was a one factor model used as a baseline comparison against the other models. Model 2 was a two factor model with pain disorders and joint disorders treated as latent factors. For Model 3, each factor included in the diagnosis tree of the original DC/TMD Axis I for adults, was treated as latent factor, except for Subluxation, Disc Displacement with Reduction with Intermitent Locking and Disc Displacement with Reduction with Limited Opening.

Inclusion Criteria

- Children otherwise healthy (with and without temporomandibular disorders) were included.

Exclusion Criteria:

• Cleft lip and palate reported by parents and verified its presence and/or sequels by the investigators in the clinical examination.
• Cervical spine malformation.
• Syndromes affecting morphology or facial function.
• Joint diseases.
• Neurological diseases.
• Neuropathies.
• Current or previous treatment of head and neck radiation.
• Fibromyalgia.
• Maxillofacial surgeries.

Children between 7 and 11 years of age of both genders, that were recruited in the specialized orofacial pain clinic at Universidad CES.

Only the data base of the investigation can be shared. Personal data cannot be shared, due to ethics commitee restrictions.

After publication of the article.

By requirement to the e-mail crestrepos@ces.edu.co