The oral health functioning impairment instrument for people with Down syndrome is probably the only validated instrument to assess the oral health independently of clinical examination. The Brazilian Portuguese version of the scale exhibited acceptable reliability and validity. The validation and adaptation of instruments related to oral health is important to ensure the comparability and usefulness of research results [7, 10, 14, 15].
To achieve validation it is necessary to follow some standard recommendations. The basic recommendation for this process is in this order: translation, committee approach and back translation [7, 10, 14–16].
Translation can be performed by two or more professionals. The inclusion of the third professional as fulfilled in this study was designed to assess the clinical importance of the items phrasing that might be of great importance for eventually distinguishing cases in the clinical range [10, 16].
The back translation process is important to identify and correct discrepancies that may occur in the translation process. In this study two back translation versions were obtained as suggested by Beaton et al. . The participation of the original instrument’s author increased the assertiveness of the final version in Portuguese language .
Questionnaire’s internal consistency was almost equal to those obtained in the original validation which presented values of 0.52-0.79. These results are in fact acceptable as consistency of approximately 0.70 is ideally, over 0.90 it suggests redundancy and under 0.50 it suggests poor internal consistency . Test-retest reliability was considered good and showed similar results with the original study, where it achieved ICC values of 0.64 to 0.84 .
Construct validity was verified by comparing the instrument scores with clinical examination for malocclusion because it’s a usual abnormality in people with DS [1, 3–5]. The disharmony between the bones of the face has a high prevalence in this group [3–5]. Although malocclusions are rarely life threatening, they can cause pain, infection, respiratory complications, and problems with mastication and speech [2–5].
It was expected that parafunction subscale presented association with malocclusion what did not happen [3, 4]. In regard of this result, it can be observed that almost all of the six items of parafunctional subscale deal with habits concerning feeding process, which do not have an association with malocclusion. Only the items concerning protruding tongue and grinding teeth could present relationship with malocclusion . The value of 0.49 Cronbach's alpha of the subscale parafunction shows that this is a subscale should not be used alone. This way maybe the parafunction subscale needs to be correlated with other clinical measures to assure its validity. That only makes sense if used in the complete instrument, which showed a value of 0.80, being quite acceptable.
It was expected that communication subscale was correlated with malocclusion as the deformities in the overall oral cavity lead DS people to present problems in their speech development. Due to malformation of the nasal bones, muscle hypotonia and the large tongue, the DS people usually keeps his mouth open and the tongue between his lips [3, 4]. Regarding this results it can be observed that half of the six items concerning communication subscale deal with speech capability.
Eating capability of DS people is in fact affected by the reduced production of saliva, the large tongue, the small oral cavity, oral hypotonia, abnormal tongue movement and uncoordinated sucking and swallowing. Besides, other overall health problems will affect their eating capacity, as digestive dismotility [3, 4]. Thus the eating subscale may be influenced by the overall health more than by the malocclusion severity.
Symptoms subscale presented a significant correlation with malocclusion although the index is based on aesthetic features. It must be noticed that items of this subscale concerns about pain, bleeding gums, bead breath and the role perception about the oral health. This way, symptoms subscale was in fact expected to present correlation with malocclusion, instead its’ aesthetical feature.
Otherwise, there was a significant correlation between the instrument total assumed as a scale with malocclusion, proving the construct validity despite the results for the subscales in separate.
In the original validation study for the English version, all of the subscales presented no correlation with the clinical outcomes observed, caries experience and periodontal status. The instrument assumed as a role scale was not assessed in the Canadian study . This suggests that subscales may not present construct validity in separate, but only when analyzed as a single measure.
Educational level was selected to be a second variable aiming to verify the construct validity as social support influences the overall health of people more than economic status . The current study considered years of schooling (caregivers) as proxies for socioeconomic status in the assessment of independent negative impacts in the instrument. Low educational level may leads to reduced income, unemployment and poor occupational status. These conditions influence the health behaviours and self-rated oral health [1, 3, 4, 10]. The educational level did not presented association with people with DS oral health indexes in another study . Results pointed out that caregivers’ schooling was correlated with the scale and the subscales communication and symptoms. It is possible that those aspects are in fact correlated with educational level [3, 4, 6]. In conclusion, the construct validity of the overall scale seems to be proved but for subscales it must be observed with careful.
Probably the weak correlations occurred because the instrument has very subjective questions and also for being a proxy measure, which not always represents the real oral health status of the individual with Down syndrome. Moreover, measures with fewer items tend to be more sensitive to this analysis. The authors of the original instrument have not found high values in the respective correlations .
Discriminant validity proved to be valid. The results proved that the instrument as a scale and the subscales behaves differently within DS and non-DS individuals. As in the original study, the Brazilian version of the scale discriminates the specifically problems of DS people and the result was similar .
There are several limitations in this study that should be pointed out. First the sample selection could arise some doubts. Convenience sample increases the possibility of bias and may lead the sample to be similar in many aspects. The DS people in the role population may present different characteristics. There was a greater range of groups’ age. Thus the caregivers of the oldest people with DS were not their parents but other people closest in age, as their brothers. They may not notice about their health conditions or may have a worst expectancy compared with the youngest caregivers as the DS overall heath seems to be improving as their life expectancy [2, 3]. Other limitation of this study is that the questionnaire is a proxy measure, which not always represents the real oral health condition .
The oral health conditions affect quality of life of all types of individuals. Thus, check symptoms, functional limitations, emotional and social wellbeing related to oral cavity should be considered when evaluating the patient and the population overall health [6, 8, 10]. This scale has an important role in the evaluation of oral conditions of individuals with DS, identifying, through the report of the caregivers, the impact of oral diseases, mainly of malocclusion on quality of life in this part of the population.