The Pediatric Screening–Priority Evaluation Dysphagia: Validation of a new tool for screening swallowing disorders in infants and children

Abstract

Aim

To assess the psychometric properties of the Pediatric Screening–Priority Evaluation Dysphagia (PS–PED), a novel 14‐item, non‐invasive tool for identifying dysphagia risk in infants and children.

Method

Internal consistency and interrater reliability were evaluated using Cronbach's alpha and intraclass correlation coefficient (ICC). Concurrent validity was assessed by correlating PS–PED scores with the Penetration Aspiration Scale (PAS). Analysis of variance examined score differences across neurological and neuromuscular conditions, congenital and musculoskeletal abnormalities, cardiovascular disease, and genetic syndromes. Receiver operating characteristic (ROC) curves determined cut‐offs for optimal sensitivity and specificity.

Results

The PS–PED was administered to 117 children (59 males and 58 females; mean age 6 years 8 months, SD 4 years 4 months), showing good internal consistency (Cronbach's alpha = 0.716) and strong interrater reliability (ICC = 1). A positive correlation (0.765) was found with the PAS, with significant score differences among diagnostic groups. ROC analysis established cut‐offs for two dysphagia risk levels.

Interpretation

The PS–PED is a reliable, valid screening tool for dysphagia, facilitating early identification in infants and children across various medical conditions.

The Pediatric Screening‐Priority Evaluation Dysphagia (PS‐PED) is a novel, non‐invasive, 14‐item screening tool designed to identify swallowing disorders in children. It encompasses three key domains: clinical history, health status, and feeding condition. Based on two defined thresholds, PS‐PED enables the identification of children at low risk or at risk for dysphagia, supporting early detection and prioritization of clinical management.

This original article is commented on by Edney and Sanchez on pages 10–11 of this issue.

Abbreviations

AUC

area under the curve

ICC

intraclass correlation coefficient

PAS

Penetration Aspiration Scale

PS–PED

Pediatric Screening–Priority Evaluation Dysphagia

ROC

receiver operating characteristic

VFSS

videofluoroscopic swallowing study

What this paper adds.

• The Pediatric Screening–Priority Evaluation Dysphagia (PS–PED) is a valid and reliable tool for identifying risk of dysphagia in infants and children.

• Two levels of risk for dysphagia (low and at‐risk) were identified according to PS–PED scores.

• The PS–PED allows prioritization of dysphagia management and rehabilitation according to the level of risk.

Dysphagia, or problems with swallowing, poses significant challenges for children, impacting their nutritional intake, growth, and overall quality of life. ¹ Understanding the prevalence and incidence of dysphagia in childhood is crucial for health care professionals to provide timely intervention and support. While exact figures vary across studies and populations, research consistently indicates that dysphagia is not uncommon among children. The prevalence of paediatric dysphagia varies significantly, and is influenced by factors such as the definition of dysphagia applied, the characteristics of the study population, and the methodologies employed. ² It is assumed that the incidence of swallowing disorders is increasing because of the improved survival rates of children with complex and medically fragile conditions; ³ , ⁴ for example, the prevalence of swallowing difficulties was estimated to be 13.5% in craniofacial microsomia, ⁵ 86% in laryngeal cleft, type ¹ , ⁶ and 42.9% in congenital heart disease. ⁷ Screening for dysphagia in childhood is imperative to enable identification of at‐risk individuals early and initiate appropriate management strategies. Early detection allows health care providers to implement interventions aimed at minimizing complications such as malnutrition, dehydration, aspiration pneumonia, and impaired growth and development. ⁸ Moreover, timely screening enables tailored treatment plans to address the underlying causes of dysphagia, which may include anatomical abnormalities, neuromuscular disorders, or developmental delays. ⁹ , ¹⁰ , ¹¹ , ¹²

A variety of assessment tools are available to evaluate swallowing function in paediatric populations, each serving a specific purpose and offering unique advantages. ¹³ Clinical assessment methods involve a thorough evaluation of a child's medical history, physical examination, observation of feeding behaviours, and assessment of oral motor skills, tailored to the child's age and developmental milestones. ⁸ These tools provide valuable insights into the nature and severity of dysphagia, guiding clinical decision‐making regarding further diagnostic investigations and therapeutic interventions. Instrumental assessments, such as videofluoroscopic swallowing study (VFSS) and fiberoptic endoscopic evaluation of swallowing, offer detailed visualization of the swallowing process and allow for objective measurement of swallowing function. ¹⁴ These modalities are particularly useful for identifying aspiration, assessing the effectiveness of swallowing manoeuvres, and guiding recommendations for dietary modifications and feeding techniques. While instrumental assessments provide valuable diagnostic information, they often require specialized equipment and expertise, limiting their availability in certain health care settings. Furthermore, despite instrumental evaluations remaining the preferred methods for evaluating swallowing dysfunction, ¹⁵ they are limited in terms of texture and do not always reflect the reality of actual meals, especially in a paediatric context.

The Pediatric Screening–Priority Evaluation Dysphagia (PS–PED) is a paediatric screening tool, developed in 2023 ¹⁶ and used to identify risk of dysphagia in children and infants. In keeping with recommendations made in recent studies ¹⁷ , ¹⁸ and in the white paper by the European Society for Swallowing Disorders, ¹⁹ the PS–PED was developed to be easy to use including by non‐specialist personnel (e.g. nurses, other rehabilitation professionals who have no or little experience in swallowing disorder evaluation), to be rapid and easy to administer, and to avoid health risk. It does not involve food administration and involves little financial commitment. The PS–PED was developed and used in a pilot study ¹⁶ with a limited sample of children, but has not yet been validated in a large sample of children. Preliminary results show good internal consistency (Cronbach's alpha = 0.716), very good reproducibility (intraclass correlation coefficient [ICC] = 1), and a positive linear correlation (Pearson's correlation coefficient of 0.765 with p < 0.01) observed with the VFSS. The PS–PED has already been used in clinical studies for children with Down syndrome ²⁰ and CHARGE syndrome, ²¹ although with a provisional cut‐off, it provides useful information regarding the risk of dysphagia and targeted rehabilitation.

Based on the promising preliminary results ¹⁶ and considering that the PS–PED has already been developed and used in clinical practice, there is the necessity to validate this screening tool on a larger sample of children, namely neurological and neuromuscular conditions, congenital and musculoskeletal abnormalities, cardiovascular disease, and genetic syndrome. This validation aims to demonstrate the reliability, validity, discriminative ability, and assessment capacity of the PS–PED.

METHOD

Study design

This is an observational cross‐sectional study conducted at the Bambino Gesù Children Hospital, between December 2022 and August 2023. This study was approved by the Bambino Gesù Children's Hospital Ethics Committee—Rome, Italy (protocol number 2352_OPBG_2021). All parents or caregivers of the participants signed written informed consent. Confidentiality of the provided information was maintained, and all data were encoded for analysis and only used for research purposes.

Tools

The PS–PED, developed by Cerchiari et al., ¹⁶ consists of 14 items grouped into three domains: clinical history, health status, and feeding condition. Each item is structured to elicit a binary ‘Yes’ or ‘No’ response based on clinical information. All data are collected from medical records, with the option to consult parents or caregivers if any information is missing. Scoring is straightforward: a score of 0 denotes the absence and 1 the presence of each statement. The total number of affirmative responses provides a score indicative of dysphagia risk, with higher scores reflecting a greater likelihood. Details of the PS–PED tool are presented in Table 1.

TABLE 1.

Description of the Pediatric Screening–Priority Evaluation Dysphagia.

Domain	Number of items	Items	Response score
			Yes	No
Clinical history	4	1. Neurological diagnosis	1	0
		2. Epilepsy medications	1	0
		3. Heart disease	1	0
		4. Structural anomalies of the digestive and respiratory systems	1	0
Health status	7	5. Tracheal tube	1	0
		6. Decreased alertness	1	0
		7. Malnutrition and/or poor growth	1	0
		8. Recurrent respiratory tract infections	1	0
		9. Use of the suction machine/aspirator	1	0
		10. Lack of head control and/or postural instability	1	0
		11. Gastrointestinal diseases (gag reflex, vomit, constipation, GERD)	1	0
Feeding conditions	3	12. Parenteral/enteral nutrition (nasogastric tube, gastrostomy tube, etc.)	1	0
		13. Feeding with consistency and unsuitable food for the child's development stage	1	0
		14. Prolonged mealtime (over 50 minutes)	1	0

Abbreviation: GERD, gastroesophageal reflux disease.

The Penetration Aspiration Scale (PAS) ²² serves as a classification system for the results obtained from VFSS, also known as modified bolus studies. VFSS offers dynamic and sequential visualization of the various stages of swallowing. ²³ , ²⁴ The PAS, a widely utilized method in both clinical practice and research settings, is employed to characterize and quantify the severity of bolus entry into the airway. ²⁵ , ²⁶ With eight levels, the PAS delineates the extent of bolus penetration and aspiration within the airway. ²⁷ Given the grave respiratory implications of penetration and aspiration events, all VFSS results in this study were categorized based on PAS scores and subsequently divided into three levels. Specifically, PAS 0 encompassed score 1 findings (indicating no entry of material into the airway), PAS 1 included scores 2 through 5 (denoting penetration beyond the laryngeal aditus, up to the true vocal cords), and PAS 2 comprised scores 6 to 8, which indicate aspiration into the subglottic space and trachea. Table 2 provides an overview of the PAS scores and their corresponding categorization.

TABLE 2.

Penetration Aspiration Scale (PAS) and division into different classes.

Score	Material	PAS
1	No penetration or aspiration	0
2	Penetration, contrast remains above the vocal folds, not ejected	1
3	Penetration, contrast remains above the vocal folds, not ejected
4	Penetration, contrast contacts vocal folds, subsequently ejected
5	Penetration, contrast contacts vocal folds, not ejected
6	Aspiration (contrast below vocal folds), subsequently ejected (at least into larynx)	2
7	Aspiration (contrast below vocal folds), not ejected despite effort
8	Aspiration (contrast below vocal folds), no effort made to eject

Study participants

All children undergoing VFSS because of feeding and swallowing disorders for clinical purposes were included in the study. Participants were recruited from the Radiology Department at Bambino Gesù Children's Hospital, ensuring an adequate sample size to meet the minimum ratio of five to eight patients per item. ¹⁰ , ²⁸

To be eligible for participation, patients had to meet the following criteria: age between 0 years and 18 years and scheduled for a VFSS between December 2022 and August 2023. No additional specific inclusion or exclusion criteria were applied. Considering the heterogeneity of health conditions who were referred to the Dysphagia Unit of the Bambino Gesù Children's Hospital IRCCS, the research group decided to use internationally recognized classifications, ²⁹ , ³⁰ , ³¹ and in this respect were classified as follows: (1) neurological and neuromuscular conditions; (2) congenital and musculoskeletal abnormalities; (3) cardiovascular disease; (4) genetic syndromes; (5) other.

The PS–PED was administered by an experienced speech and language pathologist, while the VFSS procedure and the PAS score were performed by a radiologist with a senior speech and language pathologist.

Procedures and data analysis

The PS–PED scoring process involved five raters to guarantee both reliability and methodological precision. Initially, three independent speech therapists conducted the screening assessments separately, each meticulously recording the date and time of their evaluations. This thorough documentation ensured that every screening session was accurately logged and could be referenced later, thereby supporting the study's reproducibility.

Subsequently, a separate evaluation was performed by two additional experts: a radiologist and a speech therapist specializing in deglutology. These specialists were tasked with analysing the VFSS exam and assigning scores using the PAS. To prevent any potential bias, they were blinded to the previously determined PS–PED score. Their independent assessment ensured that the PAS scoring was conducted objectively and without influence from the initial screening results.

This two‐phase approach—independent screening followed by blinded expert evaluation—was implemented to enhance the accuracy and reliability of the findings by minimizing subjective bias and ensuring that each evaluation phase was executed with the necessary expertise and rigorous documentation.

Three primary aspects were examined to evaluate the psychometric properties of the PS–PED: internal consistency reliability, interrater reliability, and criterion validity. Internal consistency was assessed using Cronbach's alpha, with a criterion of an alpha coefficient less than 0.70 considered satisfactory for scale homogeneity, as suggested by Kline. ³² Interrater reliability was assessed by examining the agreement between two independent raters administering the PS–PED to the same child. The ICC was used to measure interrater reliability, with a value of 0.70 deemed acceptable, while values of 0.80 and 0.90 were considered good and very good respectively. While three speech therapists independently conducted the PS–PED screening, only two raters' scores were included in the ICC calculation to maintain a standard two‐rater comparison. To evaluate concurrent validity, two additional independent raters, a radiologist and a speech therapist specializing in deglutology, assessed the VFSS examination and PAS scores while remaining blinded to the PS–PED results. ³³

Additionally, the Pearson correlation coefficient was used to assess the criterion validity of the PS–PED by comparing its scores with those of the PAS subscales. Pearson correlation coefficients range from −1 to +1, indicating the strength and direction of a linear relationship. While typical concurrent/convergent validity studies often use lower correlation thresholds (around 0.40–0.50) depending on the construct being measured, ³⁴ we imposed a higher correlation coefficient threshold (≥0.70), given that the PAS serves as the criterion standard for swallowing impairment assessment. This was done to ensure a strong level of agreement between the PS–PED and the established reference measure.

Receiver operating characteristic (ROC) curves and the coordinates of the curve were analysed to identify cut‐off points with optimal sensitivity and specificity, presented as percentages. ³⁵ The cut‐off values were determined using a combination of statistical methods, including Youden's index, which identifies the threshold that maximizes the sum of sensitivity and specificity while minimizing false positives and false negatives. ROC curves are particularly advantageous when a tool needs to evaluate multiple cut‐off points, as they allow visualization of sensitivity and specificity across a range of values.

Two independent ROC analyses were carried out to derive PS–PED cut‐off scores corresponding to two clinically relevant dysphagia risk levels. First, a ‘rule‐out’ ROC contrasted children with no airway invasion (PAS 0) against those with any penetration or aspiration (PAS 1 + 2) to establish the threshold that most accurately identifies children at low risk for dysphagia. Second, a complementary ‘rule‐in’ ROC contrasted children with aspiration (PAS 2) against those with no or only penetration (PAS 0 + 1) to define the threshold that best identifies cases requiring further evaluation. In each analysis the optimal cut‐off was selected by maximizing Youden's index (J = sensitivity + specificity −1). The resulting lower and upper thresholds were adopted to dichotomize PS–PED scores into low‐risk and at‐risk categories, with priority given to high sensitivity in the rule‐out context and to balance accuracy in the rule‐in context. ³⁶

Through these two ROC analyses, we established two distinct cut‐off points, allowing for the classification of children into low‐ and at‐risk categories based on their PS–PED scores. The final PS–PED cut‐off scores were selected by prioritizing high sensitivity, ensuring that at‐risk individuals were correctly identified while maintaining an acceptable level of specificity to avoid excessive false positives.

To further assess diagnostic accuracy, we calculated the area under the curve (AUC) with confidence intervals, providing an empirical basis for evaluating the discriminative power of the PS–PED. AUC values were interpreted as follows: AUC 0.90 or higher (excellent), 0.80–0.89 (good), 0.70–0.79 (fair), and less than 0.70 (poor). ³⁷ Additionally, a one‐way analysis of variance test was conducted to evaluate the PS–PED's performance across different diagnostic categories, further validating the tool's classification accuracy.

RESULTS

Based on the inclusion criteria, approximately 300 children in a year underwent VFSS, resulting in the recruitment of 117 participants for our study, corresponding to a recruitment ratio of 1:2.57. Among these 117 children (59 males and 58 females) with various health conditions, the mean age was 6 years 8 months (SD 4 years 4 months). A summary of demographic characteristics is provided in Table 3.

TABLE 3.

Demographic characteristics of the participants.

Characteristics	n	%
Sex
Male	59	50.43%
Female	58	49.57%
Age
Range, years:months	0:4–17:0
Mean (SD), years:months	6:8 (4:4)
Health conditions
Neurological and neuromuscular conditions	57	48.72
2Congenital and musculoskeletal abnormalities	16	13.67
3Cardiovascular disease	9	7.69
4Genetic syndromes	32	27.36
5Other	3	2.56

For criterion validity, data revealed a positive linear correlation between PS–PED and PAS scores, with a Pearson's correlation coefficient of 0.765 (p < 0.01). For reliability analysis, the internal consistency of the PS–PED, as measured by Cronbach's alpha, was found to be 0.716, demonstrating satisfactory internal consistency of the instrument. While for interrater reliability ICC was 1.00, obtaining a mean (SD) score of 6.13 (2.93) for both administrations.

The analysis conducted using ROC curves determined the optimal cut‐off values for classifying dysphagia risk categories using the PS–PED in comparison to PAS. The ROC analysis contrasting PAS 0 versus PAS 1 + 2 (low‐risk level) yielded an AUC of 0.964 (Figure 1). Maximizing Youden's index identified PS–PED 5 or less as the optimal rule‐out threshold (sensitivity = 98%, specificity = 71%; J = 0.69). For the complementary analysis PAS 2 versus PAS 0 + 1 (at‐risk level), the AUC was 0.937 (Figure 2); the optimal rule‐in threshold was PS–PED 6 or more (sensitivity = 95%, specificity = 80%; J = 0.75). All diagnostic‐accuracy metrics for the two cut‐offs are summarized in Table 4.

FIGURE 1.

Receiver operating characteristic (ROC) curve to evaluate low risk of dysphagia (PAS 0 vs. PAS 1 + 2). Abbreviation: PAS, Penetration Aspiration Scale.

FIGURE 2.

Receiver operating characteristic (ROC) curve to evaluate at risk of dysphagia (PAS 2 vs. PAS 0 + 1). Abbreviation: PAS, Penetration Aspiration Scale.

TABLE 4.

Cut‐off points of PS–PED according to PAS.

	PAS score
	PAS 0 vs. PAS 1 + 2 (low‐risk)		PAS 2 vs. PAS 0 (at‐risk)
Area (95% CI)	96.4% (92.4%–100%)		93.7% (89.7%–97.7%)
PS–PED score	Sensitivity	Specificity	Sensitivity	Specificity
4	100%	56%	100%	51%
5	98%	71%	97%	65%
6	96%	98%	95%	80%
7	76%	97%	77%	92%
8	49%	99%	51%	96%

Optimal cut‐off points	PAS	Sensitivity	Specificity	Significance
≤ 5	Low‐risk	98%	71%	< 0.01
≥ 6	At‐risk	95%	80%	< 0.01

Selected cut‐offs for sensitivity and specificity is shown in bold type.

Abbreviations: CI, confidence interval; PAS, Penetration Aspiration Scale; PS–PED, Pediatric Screening–Priority Evaluation Dysphagia.

The use of the one‐way analysis of variance test establishes an F value of 9.87, reaching significance with a p‐value less than 0.01. This indicates a statistically significant difference among the mean scores across different health conditions. To better understand how PS–PED behaves across the different diagnostic groups, the average scores have been reported in a box plot (Figure 3).

FIGURE 3.

Box plot of Pediatric Screening–Priority Evaluation Dysphagia (PS‐PED) scores according to different diagnostic groups.

DISCUSSION

Screening tools play a pivotal role in preventive medicine, aiming to reduce morbidity and mortality by detecting diseases early. Current guidelines for dysphagia emphasize the importance of health care professionals conducting screenings to identify high‐risk individuals who may require dysphagia speech therapy assessment.

Analyses confirm that the PS–PED is a reliable and valid screening tool for identifying risk of swallowing disorders in infants and children. Utilizing screening and assessment tools with robust psychometric properties is essential in research, enhancing clinical decision‐making and guiding appropriate treatment choices based on valid and reliable assessment scores. Effective screening tools for dysphagia should be rapid (15–20 minutes), cost effective, reliable, minimally invasive, and pose minimal risk to patients. ¹⁷

The study's results indicate that the PS–PED reliably represents the clinical condition of dysphagia in a diverse group of children. Interrater reliability showed perfect concordance (ICC 1.00), while internal consistency revealed a Cronbach coefficient alpha of 0.716, consistent with previous validation studies. ¹⁶ Although Cronbach's alpha is a conventional indicator of internal consistency, its application to the PS–PED should be interpreted in light of the tool's design, which includes a range of items reflecting clinical history, health status, and feeding conditions. In this context, a lower Cronbach's alpha would not necessarily indicate a flaw in the instrument. Nevertheless, the observed Cronbach's alpha value was acceptable, supporting the overall coherence of the measure and providing a useful benchmark for comparison with other paediatric dysphagia screening tools. With regard to interrater reliability, information was gathered from medical records, with items addressing clear and unambiguous conditions such as clinical history, health status, and feeding conditions. For instance, item 5 enquires about the presence of a tracheal cannula, providing a clear and indisputable indication. Moreover, the yes‐or‐no response format minimizes ambiguity, likely contributing to perfect concordance between raters.

Pearson's correlation demonstrated a positive linear correlation of 0.77 (p < 0.01) with the PAS, slightly lower than the preliminary study's findings (0.82). ¹⁶ The PS–PED score positively correlates with the PAS score, indicating that higher scores on both scales correspond to an increased risk of dysphagia.

Analysis through ROC curves established cut‐off values for identifying dysphagia risk categories using the PS–PED in comparison to the PAS. The PS–PED demonstrated a high sensitivity of 98% and specificity of 71% for identifying low‐risk dysphagia (PAS 0 vs. PAS 1 + 2) and a sensitivity of 95% and specificity of 80% for identifying at‐risk dysphagia (PAS 2 vs. PAS 0 + 1). These results underscore the PS–PED's potential as a screening tool in paediatric settings, particularly when used by non‐specialist personnel to identify children at risk of dysphagia. The primary objective of a screening tool is to ensure that children with dysphagia are identified early and referred for further evaluation. Given the serious health consequences of undiagnosed dysphagia, such as aspiration pneumonia, malnutrition, and growth failure, prioritizing sensitivity is critical in this context. As highlighted in previous studies, ¹⁵ , ¹⁶ screening for dysphagia in children presents unique challenges, but the risks of false negatives (missed cases) far outweigh those of false positives in this population. A high sensitivity (98% for low‐risk and 95% for at‐risk dysphagia) ensures that nearly all at‐risk children are identified, reducing the likelihood of delays in diagnosis and treatment. To further assess the performance of the PS–PED, we calculated Youden's index to quantify the balance between sensitivity and specificity. The cut‐off for low‐risk (PS–PED ≤5) yielded a Youden's index of 0.69, while the cut‐off for at‐risk (PS–PED ≥6) yielded a Youden's index of 0.75. These values indicate that PS–PED effectively stratifies dysphagia risk, with a slightly better performance in identifying at‐risk cases. The moderate specificity (71% for low‐risk and 80% for at‐risk dysphagia) reflects the expected trade‐off in a screening tool, where false positives are acceptable to minimize the risk of false negatives.

Importantly, false positives in a screening context do not necessarily lead to unnecessary invasive procedures. Follow‐up assessments can be conducted non‐invasively by a dysphagia expert through clinical bedside evaluations, which are both cost effective and accessible. ¹⁶ The balance between strong sensitivity and acceptable specificity ensures that clinicians can confidently identify children requiring further assessment, minimizing misclassification while avoiding harm to patients. ²⁹ Furthermore, an effective screening tool provides safety and peace of mind for both clinicians and caregivers, ensuring that children at risk receive timely and appropriate intervention. However, a score below the cut‐off does not eliminate the risk of dysphagia and should not preclude referral to or acceptance into dysphagia services if suspicion of dysphagia/aspiration remains.

These findings reinforce the PS–PED's role as a reliable first‐line screening tool in paediatric dysphagia assessment. Future research could explore refinements to the cut‐off values to further optimize specificity while maintaining high sensitivity, particularly in diverse clinical populations.

Similarly to the preliminary study, we tried to explore how PS–PED works in different health conditions and diagnostic groups. Swallowing and feeding difficulties in childhood can arise from specific medical conditions or complex clinical scenarios. Diagnostic categories linked to dysphagia encompass neurological disorders, anatomical abnormalities, genetic conditions, and cardiovascular diseases. However, sensory, motor, and behavioural factors may complicate paediatric screening processes ³⁸ , ³⁹ and therefore a screening tool that does not require direct collaboration from the child, such as the PS–PED, is a notable advantage. Our findings revealed that group with neurological and neuromuscular conditions showed a higher risk of dysphagia; in fact dysphagia is prominent in children with histories of preterm birth, and neurological and neuromuscular diagnoses, and is associated with medical, health, and neurodevelopmental problems. ⁴⁰ , ⁴¹ In children with developmental disabilities, dysphagia can reach up to 80%. ⁴² , ⁴³ Dysphagia frequency in children with stroke varied widely from 24.2% to 88.6%, ⁴⁴ it is 47.2% for children with neuromuscular conditions, ⁴⁵ while for children with cerebral palsy prevalence estimates are 50.4% (95% confidence interval 36.0–64.8). ⁴⁶ In our sample, we found that children with a diagnoses of both neurological and neuromuscular conditions represent the population with the highest risk of dysphagia, according to PS–PED mean scores. For other diagnostic groups, we found high variability in terms of scores, probably due to the heterogeneity of clinical conditions and age. Further studies should investigate these specific aspects and evaluate how PS–PED specifically performs in each diagnostic group with a larger sample.

Despite the encouraging findings, this study has certain limitations that merit acknowledgement. First, as a single‐centre study, the findings are drawn from a single institution. Although the research team is based at a major paediatric hospital in Europe, conducting a multicentre study would provide a broader assessment of the PS–PED tool's feasibility. Such a study could evaluate the tool across diverse hospital settings with variations in departmental organization, dysphagia awareness, and available human resources. Additionally, the interrater reliability of the tool was based on scores from experienced speech therapists specializing in dysphagia, which may have influenced the ICC results and the observed positive concordance. Future studies could involve professionals with varying levels of experience to further validate the tool's reliability. Since the PS–PED is intended to aid in intervention prioritization, assessing its usability among health care professionals with diverse roles, including occupational therapists, physiotherapists, nurses, and physicians, would be valuable. Regarding the determination of cut‐off values, these were derived from a heterogeneous clinical population, while it may be advantageous to calculate specific cut‐offs for different health conditions. Given that dysphagia risk varies across clinical populations (see Figure 3), tailored cut‐off points could improve screening accuracy. For instance, children with progressive conditions (e.g. neuromuscular or neurodegenerative diseases) may benefit from a lower cut‐off (higher sensitivity) to detect early swallowing impairments before severe complications arise. Conversely, children with transient or fluctuating symptoms (e.g. postsurgical recovery, preterm birth‐related feeding difficulties) may require a higher cut‐off (higher specificity) to avoid unnecessary follow‐ups for cases likely to resolve spontaneously. Additionally, different disease‐specific dysphagia patterns could influence the predictive value of the PS–PED, warranting targeted ROC analyses for distinct clinical groups. Future research should explore population‐specific cut‐offs to optimize the tool's performance across different health conditions, ensuring greater clinical applicability and precision in risk stratification. Finally, we did not assess whether certain items might carry more weight in the overall score, such as whether the presence of a tracheostomy might be a stronger predictive factor than other items. Future research could explore the impact of individual items on the total score, particularly across various paediatric populations.

In conclusion, this study demonstrated the PS–PED as a valid and reliable tool for assessing the risk of dysphagia in infants and children.

CONFLICTS OF INTEREST STATEMENT

The authors declare no conflict of interest.

Cerchiari A, Biondo G, Fassari FM, Pizza F, Giordani C, Falbo M, et al. The Pediatric Screening–Priority Evaluation Dysphagia: Validation of a new tool for screening swallowing disorders in infants and children. Dev Med Child Neurol. 2026;68:72–81. 10.1111/dmcn.16390

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.