Validation and Cultural Adaptation of an Arabic Version of Pediatric Eating Assessment Tool (Pedi-EAT-10_Arabic)

Oropharyngeal dysphagia has been recently addressed more in the pediatric population [1]. According to research, around 1% of children in the general community suffer from swallowing problems [2]. But this incidence is substantially higher in special populations of the pediatric age group (e.g. children with cerebral palsy, traumatic brain injury, and airway malformations [3, 4].

There is a high incidence of parental reporting of abnormal swallowing and feeding function up to 50% of the parents of general pediatric population and the negative impacts on children [3, 5‐7]. These negative impacts include dehydration, malnutrition which consequently affects both development and psychological well-being. Pulmonary complications, namely pneumonia, pulmonary abscess, and death are also common consequences of oropharyngeal dysphagia [4, 8, 9].

Despite dysphagia was reported in healthy children, it is more common in pediatric patients with a history of prematurity, neuromuscular disorders, cardiopulmonary disorders, upper aero-digestive tract congenital anomalies, and gastrointestinal tract disorders [10].

Even though early recognition of pediatric dysphagia is important to ensure the appropriate precautions and interventions. This disorder is frequently missed in clinical practice [9]. Pediatric swallowing disorders are assessed by both non-instrumental and instrumental techniques [5]. The two most commonly used instrumental evaluations of swallowing for the pediatric population are the video fluoroscopic swallowing study (VFSS) [11, 12] and FEES [13, 14]. It is not feasible to perform these procedures on all pediatric patients at risk for dysphagia because of time, cost, and need for specialized equipment [15].

The non-instrumental assessment has evolved into an important aspect of the clinical evaluation of dysphagia to determine the need for instrumental detailed evaluation of swallowing [16, 17].

Though over referral to instrumental assessment is of concern for an adult patient, it is more concerning for the pediatric patient and further investigation is required to improve the sensitivity and specificity of a swallow screen. Extra consideration needs to be made in the frequency of radiation exposure, the risks of exposure in an infant or young child, and the ability of a child to complete the instrumental assessment. Some children may refuse to complete the swallow study due to the presence of barium in the food. Others will not tolerate placement of a scope [18]. In an effort to preserve personal protective equipment (PPE) and limit exposure of patients and clinicians to SARS-CoV-2 [19], following the guidance from the Centers for Medicare & Medicaid Services (CMS), ASHA recommended delaying non-essential endoscopies in those with unknown transmission risk [20].

The validity and reliability of the Pedi-EAT 10 have been proven in patients with a wide range of dysphagia etiologies. It has high internal consistency and test–retest reliability. This test can be used to determine the severity of dysphagia as well as track treatment progress and therapeutic effects. It's a five-point ordinal scale ranging from 0 (no problem) up to 4 (very difficult or severe problem) [15].

The use of Pedi-EAT-10 as a screening tool to determine whether a child's swallowing function requires additional instrumental evaluation has been proved. It also has the advantage of capturing dysphagia symptoms in everyday situations rather than a single point of time assessment in the clinic which can miss a lot of important information about the child’s typical swallowing function [15].

The purpose of the current study is to develop an Arabic version of Pedi-EAT 10 questionnaire, and to test its validity and reliability. Additionally, the study aimed at assessing its ability to predict the further need of instrumental assessment of swallowing by correlating its results with penetration aspiration scale and residue formation, in order to be used as a valid and reliable sensitive non-invasive screening tool deciding the need to further instrumental pediatric dysphagia assessment.

The research was conducted in three phases as shown in flow chart (Fig. 1).

The translated Arabic version (Pedi-EAT10 _Arabic) is presented in Fig. 2

Pedi-EAT 10_Arabic is a practical, simple to administer dysphagia caregiver symptom-specific outcome tool. It is a valid reliable non-instrumental assessment of pediatric patients at high risk of penetration-aspiration [15]. Considering the proved advantages of parent reported measures in pediatric population [26], Pedi-Eat 10 was originally adapted from EAT 10 [27],which has a translated valid Arabic version for adults [28]. Up to our knowledge, no current valid Arabic tool for pediatric age group, so this study aimed to provide an Arabic version of the Pedi-EAT10 questionnaire, validated to be used for research and clinic in Arabic speaking countries. This translated version, after transcultural adaptation and validation has proven to be a discriminant, valid and reliable tool to screen for dysphagia and assess the need of further instrumental assessment of swallowing. To provide equivalence between the Arabic and the English version of the Pedi-EAT 10, a rigorous translation and Cross-cultural adaptation processes was followed. Proof of Correctness and equivalence between the two questionnaires was provided by the high internal consistency of the translated questionnaire, by its consistent construct validity and the excellent test–retest reliability observed in results.

The mean age of the studied population was 2.48 ± 1.79 years, min–max (0.75—10.08) while the mean age for healthy children in this study was 2.81 ± 1.79 years, min–max (0.75—9.83) of which 54.5% were males. The mean age of the healthy children in the original PEDI-EAT 10 was 4.4 (± 1.5) years (min ¼2, max ¼8), of which 51% were male [15]. So our study had wider age range recruited both younger and elder than original study with lower mean age.

The pediatric dysphagia patients showed different etiologies predominantly neurogenic followed by GIT patients, at the expense of airway structural problems which can affect generalization to other patients. However, the original Pedi-EAT 10 was validated by comparing healthy to children with cerebral palsy (CP) [15]. Another study done in 2017 showed that Pedi-EAT 10 as a valid tool in predicting aspiration in children with esophageal atresia [29].

The total content validity of the translated Arabic version was of 0.96 so it was very close to the CVI of original version which was 0.91 [15].

Pedi-EAT 10_Arabic was proved to be reliable by high internal consistency (Cronbach's alpha of 0.986), mildly higher than original Cronbach alpha for test retest was 0.87 [15]. Item to item and item to total correlation coefficient ranged from 0.546 to 0.94 denoting significant moderate to strong correlation. Adequate test–retest reliability for all questionnaire items with intra class correlation coefficient ranged from 0.97 to 1 (excellent) and total ICC = 0.998, which is similar to the results of the original English version, with total ICC 0.983 at 95% confidence interval [15].

In this cross-sectional observational study, discriminant validity of the Pedi-EAT 10_Arabic scores was established on a group of 202 pediatric patients with dysphagia and 202 healthy, median of dysphagia group was median ± IQR = 27(21–34) versus 0 (0–2) for the healthy group with statistically significant difference between both groups. In the original version. The mean PEDI-EAT-10 score of the normal cohort was significantly lower than the mean PEDI-EAT-10 scores of the CP children who had complaints about swallowing difficulty (z ¼ 10.01, P < 0.001) [15].

The construct (convergent) validity was proofed by symptom-specific questionnaire Pedi-EAT 10_Arabic and the characteristics of pharyngeal dysphagia identified using FEES in dysphagic patients was described mainly in terms of PAS score and pharyngeal residue [29, 30]. A strong positive correlation between Pedi-EAT 10_Arabic item score and PAS score with Spearman correlation coefficient r = 0.803 and P < 0.001. Spearman correlation showed significant correlation between residue score and Pedi-EAT 10_Arabic total score with r = 0.63 and P > 0.001. Mean Pedi-EAT 10_Arabic scores were significantly higher among those aspirating than patients not aspirating. These similar results were reported by the original Pedi-EAT 10_Arabic tool which was proven to be a consistent scale for determining the risk of penetration/ aspiration in children with CP tested with VFSS and is used as a discriminative tool for identification of penetration/ aspiration risk in children and higher scores indicate a greater risk for penetration/ aspiration [15].

Regarding the predictive validity of the Pedi-EAT 10_Arabic, the cut off point for detection of aspiration with higher sensitivity than specificity was more than 28 with 89% sensitivity and 87.5% specificity with positive predictive value (PPV) 84.5% and negative predictive value (NPV) 90.5%. In a study done in 2018, the sensitivity of Pedi-EAT-10 score greater than 12 in predicting aspiration in children with neurogenic impairment was 77% and the specificity was 54%, has a PPV of 69% and a NPV of 64% [31]. While in another study sensitivity and specificity of Pedi-EAT 10 to predict aspiration in children with esophageal atresia with cut off score 7 was 88% and 77% and the PPV and NPV were 22% and 11%, respectively [29]. Possible explanation for the gap in cut off value for aspiration and the discriminatory capacity can be cultural variation as the mean of Pedi-EAT 10_Arabic was 3.82 ± 3.53 for healthy children and 26.70 ± 7.70 for children with dysphagia, the mean score of Pedi-EAT 10 tool for healthy children and children with cerebral palsy was 0.26 ± 1.83 and 19.5 ± 11, respectively, so even healthy controls had higher scores. Another explanation is the different etiologies: esophageal versus oropharyngeal esophageal dysphagia patients will have Pedi-EAT 10_Arabic scores similar to those with oropharyngeal dysphagia, but lower PAS scores than oropharyngeal dysphagia patients. CP children mostly with oropharyngeal dysphagia were around half of the patients and most of the other patients showed mainly esophageal dysphagia, but pharyngeal dysphagia was also not absent in these patients. The results of original Pedi-EAT 10 score of 4 or higher was classified abnormal with high sensitivity and specificity in predicting penetration/ aspiration on instrumental swallowing study [15].

Since post swallow pharyngeal pooling may be a risk factor for tracheal aspiration [30] was estimated by FEES. A significant correlation was found in our study between residue score and Arabic Pedi-EAT 10 total score. These results were similar to those reported by residue assessment by FEES but in the adult form of the questionnaire EAT 10 [32].

Our study has several limitations: exclusion of illiterate patients was made to decrease bias of the language complexity, especially the fact that it is written in formal Arabic language not in Egyptian dialect.

While ensuring the patients included in the study did not undergo major therapeutic interventions during the 2-week test retest interval, some had to change their mode of feeding from oral to NGT when aspiration was proved by instrumental assessment so these were excluded from the test retest assessment.

Lastly Pedi-EAT 10_Arabic was not tested in our study for responsiveness, i.e., the ability of the caregiver reported outcome measurement to reflect change after intervention, though this psychometric property has been demonstrated in other studies [27, 31, 33]. The ability of questionnaire to reflect after intervention which can be tested in future studies.

Arabic Pedi-EAT 10_Arabic is a reliable and valid care giver reported outcome measure for Arabic speaking caregivers of pediatric dysphagia patients. It has a strong correlation with objective findings on ISA, including penetration, aspiration and residue. As a potential screening tool for the need of further instrumental assessment of swallowing. Its discriminatory capacity of pharyngeal dysphagia was found to be high as described in terms of aspiration and post swallow residue.