Review Reports

Item

Reply

1 - 1- The abstract and the Methods section appear to report different publication windows. The abstract indicates a review period of 2015 to 2025, whereas the Methods section states January 2015 to January 2026. This discrepancy is important and must be clarified, as it affects the reproducibility and credibility of the search strategy.

Thank you for your note; there were some issues between the various versions of the paper; please accept the discrepancy to be fixed with January 2026

2- The abstract and the Methods section appear to report different publication windows. The abstract indicates a review period of 2015 to 2025, whereas the Methods section states January 2015 to January 2026. This discrepancy is important and must be clarified, as it affects the reproducibility and credibility of the search strategy.

We thank the Reviewer for this observation. The methodological protocol was defined a priori by the multidisciplinary working group; however, it was not prospectively registered in a publicly accessible registry.

We have now explicitly stated this in the Methods section and acknowledged the lack of prospective registration as a methodological limitation.

Methods section:

The methodological protocol was defined a priori by the multidisciplinary working group within the broader initiative coordinated by the Italian National Institute of Health (Istituto Superiore di Sanità, ISS).

Discussion:

A further limitation is that the study protocol was not prospectively registered in a publicly accessible systematic review registry. Nevertheless, the methodological ap-proach, eligibility criteria, PICO questions, and procedures for evidence appraisal were defined a priori by the multidisciplinary working group within the framework provided by the Italian National Health Institute [95,96]. The lack of prospective registration therefore does not imply that the review was conducted without a predefined method-ology, but it limits the possibility for external readers to independently verify protocol adherence and to assess whether any methodological decisions were modified during the review process.

3- The Methods section indicates that meta-analysis was conducted only when at least five studies addressed the same intervention/outcome in comparable populations. However, the statistical approach is not fully specified.

We thank the reviewer for the comment. In fact, due to the lack of studies compatible with the original design (i.e. at least five studies addressing the same intervention/outcome in comparable population), the study unfortunately relied on mostly descriptive approach. Therefore we (improperly, we agree about this point) improperly downgraded the section of the text, that was now properly reformulated as follows:

Given the expected clinical and methodological heterogeneity across studies, pooled estimates were calculated using random-effects models. Between-study variance (τ²) was estimated using the restricted maximum-likelihood (REML) method, and statistical het-erogeneity was assessed using Cochran's Q and quantified by the I² statistic. I² values were interpreted as indicating low (<25%), moderate (25–50%), substantial (50–75%), or con-siderable (>75%) heterogeneity [16–18]. All statistical tests were two-sided, and p values <0.05 were considered statistically significant.

5 - The manuscript states that GRADE was used to assess certainty of evidence and that recommendation strength was determined using an Evidence-to-Decision framework. While this is methodologically appropriate, the manuscript does not sufficiently explain how downgrading or upgrading decisions were made for each PICO question, nor how the Evidence-to-Decision judgments were operationalized.

We  again thank the reviewer for this comment that allows us to both improve the quality and the accuracy of our paper. In fact, the section was not fully transparent about the GRADE EtD approach, and it has been accordingly updated providing a more appropriate reference (Alonso-Coello P, Schünemann HJ, Moberg J, Brignardello-Petersen R, Akl EA, Davoli M, Treweek S, Mustafa RA, Rada G, Rosenbaum S, Morelli A, Guyatt GH, Oxman AD; GRADE Working Group. GRADE Evidence to Decision (EtD) frameworks: a systematic and transparent approach to making well informed healthcare choices. 1: Introduction. BMJ 2016; 353: i201’6. doi: 10.1136/bmj.i2016.

)

For each PICO and each critical outcome, the certainty of evidence was initially deter-mined according to study design and subsequently downgraded by one or two levels when serious or very serious concerns, respectively, were identified for risk of bias, in-consistency, indirectness, imprecision, or publication bias. Conversely, where applicable, certainty from non-randomized evidence could be upgraded in the presence of a large magnitude of effect, a dose-response gradient, or residual confounding likely to reduce rather than increase the observed effect. Final certainty was categorized as high, mod-erate, low, or very low.

The certainty of the evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. For each PICO and each critical outcome, the certainty of evidence was initially determined according to study design and subsequently downgraded by one or two levels when serious or very serious concerns, respectively, were identified for risk of bias, inconsistency, indirectness, im-precision, or publication bias. Conversely, where applicable, certainty from non-randomized evidence could be upgraded in the presence of a large magnitude of effect, a dose-response gradient, or residual confounding likely to reduce rather than increase the observed effect. Final certainty was categorized as high, moderate, low, or very low.

The formulation and strength of recommendations followed the GRADE Evi-dence-to-Decision (EtD) framework [21] and were not determined solely by the certainty of evidence. For each intervention within each PICO question, the multidisciplinary panel systematically considered the priority of the health problem, desirable and undesirable effects, certainty and directness of the available evidence, patients' values and prefer-ences, resource requirements and cost-effectiveness, equity, acceptability, and feasibility. Recommendations were subsequently formulated through structured discussion and consensus among panel members. Accordingly, recommendation strength was not as-signed through a direct correspondence with the GRADE certainty rating. Strong rec-ommendations were issued only when the overall EtD assessment clearly supported or opposed an intervention. Thus, moderate-certainty or, where justified by the overall EtD assessment, low-certainty evidence could support a strong recommendation when the expected benefits clearly outweighed potential harms or burdens, particularly when the consequences of delayed or inappropriate diagnosis were considered clinically relevant. Conversely, conditional recommendations were preferred when uncertainty remained regarding the magnitude of benefit, applicability, resource requirements, or directness of the supporting evidence. Recommendations relying predominantly on indirect or limited evidence were therefore generally formulated as conditional or consensus-based.

In several areas, the evidence base appears to rely heavily on observational studies, systematic reviews, guidelines, and expert consensus, with few direct comparative studies and very limited randomized evidence. This is especially relevant for diagnostic strategy recommendations, where evidence is often indirect. The manuscript should ensure that recommendation strength is proportionate to certainty of evidence.

We agree with the Reviewer that the strength of recommendations should remain proportionate to the certainty and directness of the supporting evidence. We therefore re-examined the recommendations across all PICO questions, with particular attention to diagnostic strategies supported predominantly by observational or indirect evidence. Where the available evidence was limited, indirect, or characterized by relevant uncertainty, recommendation strength was revised or explicitly framed as conditional/consensus-based. Strong recommendations were retained only where the panel considered that the overall Evidence-to-Decision assessment clearly supported the intervention despite the absence of high-certainty comparative evidence; the rationale for these judgments has now been made explicit in the corresponding EtD tables and in the Methods section.

In particular, two PICOs benefited from that revision, as follows:

PICO

“In children and adolescents with chronic urticaria, is a targeted diagnostic approach based on clinical history, compared to an extensive diagnostic work-up with non-targeted allergological and autoimmune tests, more effective in improving diagnostic appropriateness, including reducing costs, increasing correct diagnoses, and reducing unnecessary tests”

Summary analysis was revised as follows:

Strength        

Conditional recommendation in favour of a targeted diagnostic approach

Balance of benefits/risks

A targeted diagnostic strategy is expected to improve diagnostic appropriateness, facilitate the identification of clinically relevant autoimmune comorbidities, and reduce unnecessary investigations, healthcare costs, and patient burden. These expected benefits are considered to outweigh the small risk of missing uncommon conditions, provided that careful clinical assessment and appropriate follow-up are ensure.

Values and preferences

Families generally value diagnostic pathways that avoid unnecessary testing while identifying clinically relevant conditions. Appropriate counselling improves acceptance of a selective diagnostic strategy.

Impact on resources /  implementation

Implementation is straightforward through standardized history-taking, focused first-line investigations, and selective second-level testing. This approach may reduce unnecessary laboratory investigations and promote more appropriate use of healthcare resources.

PICO: “In children and adolescents with moderate to severe atopic dermatitis, psychological support in-tegrated into disease management is more effective than medical management alone in improving quality of life and reducing anxiety and depression”

Strength

Conditional recommendation; structured educational interventions are supported by the available evidence, while psychological support should be considered according to individual clinical needs.

Balance of benefits/risks

Structured educational programmes improve disease severity, treatment adherence, and quality of life. Selected patients may derive additional benefit from psychological assessment and support. Overall, the expected benefits outweigh the additional resource requirements.

Values and preferences

Patients, families, and clinicians place high value on interventions that improve disease control, quality of life, and psychosocial well-being.

Impact on resources / implementation

Implementation requires multidisciplinary teams and access to educational and psychological services. Availability of pediatric psychologists may limit implementation in some healthcare settings, and resource requirements should therefore be considered according to local availability and individual clinical needs.

Moreover, following minor amendments were done:

Table 2:

from: Moderate to strong in favour of specialist referral

to: Conditional recommendation in favour of specialist referral

Comment

Reply

Typographical corrections: missing letters (”Evidence. ncreasing evidence suggests ”...line 642), misspelt words (i.e. “In children and adolescents with sus-pected food allergy and atopic dermatitis, is an al-lergy consultation recommended over the regular follow-up with a primary care paediatrician to initiate specific therapy and improve quality of life?” lines 344, 345, 346).

Thank you for your suggestions: in fact, most of typos were due to cut&paste of the content from previous sections of the main text, with the inappropriate results you properly identified. WE hope that the paper has been properly fixed.

Renumber the last table; it is numbered as table15, even if the previous one is also 15.

Thank you, we fixed accordingly as Table 16. Moreover, we’ve properly cited across the main text tables 12 to 16.

Remove text duplication: " The following inclusion criteria were used” appears twice consecutively on lines 157-158

Fixed, thank you

Conflicts of interest -- A large network of Italian societies (SIMPE, SIP, SIAIP, etc.) and the Italian National Institute of Health (ISS) are cited. However, it is unclear how conflicts of interest (COI) were addressed throughout the question voting process.

We totally agree, and the COI section was amended as follows:

All members of the multidisciplinary panel participated in their individual expert capacity and not as representatives of their respective scientific societies or institutions. Potential financial and non-financial conflicts of interest relevant to the topics addressed by the guideline were considered throughout the recommendation-development process. Panel members were required to disclose any relevant interests that could potentially influence their judgment regarding individual PICO questions. Had a relevant conflict arisen, the concerned member would have been excluded from the corresponding discussion and voting process. No conflicts of interest requiring restriction or recusal from discussion or voting were identified during the guideline development process. In the end, all the authors declare no conflicts of interest relevant to the present work. No financial or non-financial conflicts requiring restriction or recusal from the formulation or voting of individ-ual recommendations were identified among panel members.

The abstract is unusually long; please verify that it meets the journal's requirements.

We amended the abstract providing a more concise version, thank you

Background. Appropriate diagnostic testing is essential to quality pediatric allergy care, affecting diagnostic accuracy, treatment decisions, resource use, and patient outcomes. This study aimed to develop evidence-based recommendations for specialist consultations and diagnostic investigations in children with food allergies, eosinophilic gastrointestinal diseases, urticaria, angioedema, and atopic dermatitis.

Methods. A multidisciplinary panel convened for the Italian National Institute of Health formulated and prioritized clinical questions using the PICO framework and structured expert consensus. A systematic review of the literature from 2015 to 2026 informed the recommendations, with certainty of evidence assessed using GRADE methodology.

Results. In food allergy, sensitization tests alone are insufficient for diagnosis, while oral food challenges remain central to diagnostic confirmation. Endoscopic biopsies are essential for diagnosing eosinophilic gastrointestinal diseases. In chronic urticaria, specialist consultation is recommended, whereas routine allergy testing should be avoided without a suggestive clinical history. In angioedema, distinguishing histaminergic from bradykinin-mediated forms is critical, with targeted laboratory testing for hereditary or acquired conditions. In atopic dermatitis, standardized severity assessment should guide management and consideration of biologic therapies, while patient and family education represents a key component of care.

Conclusions. These recommendations provide a framework for standardized pediatric allergy diagnostic pathways, aiming to improve quality and equity of care while reducing unnecessary investigations and optimizing healthcare resource use.

The conclusions are quite brief and do not reflect the research findings.

We thank the REviewer for this comment. We have therefore implemented some further sections in order to make discussion ore consistent with research findings.

Discussion was revised as follows:

This work provides evidence-based recommendations for the appropriateness of diagnostic prescriptions in pediatric allergology, focusing on food allergy, eosinophilic gastrointestinal disorders, chronic urticaria, and angioedema. The primary objective was to support evidence-informed clinical decision-making while promoting more rational use of healthcare resources, reducing unnecessary investigations, and improving equity of access to specialist care.

Overall, the available evidence supports a selective and structured diagnostic ap-proach across the clinical conditions considered. Rather than promoting extensive di-agnostic testing, the present recommendations emphasize the central role of careful history taking and physical examination, followed by investigations selected according to the clinical presentation and their expected contribution to diagnosis or management [28,89,99,100]. At the same time, the findings identify specific clinical settings in which specialist assessment and second-level investigations remain essential and should not be delayed. Therefore, diagnostic appropriateness should not be interpreted simply as a reduction in testing, but rather as the appropriate selection and timing of investigations according to the suspected condition and individual clinical context [32,89,101].

For food allergy, the evidence confirms the pivotal role of specialist multidisciplinary assessment and indicates that sensitization tests alone are insufficient to establish a di-agnosis. Skin prick tests and serum-specific IgE should be interpreted within the clinical context, while oral food challenge remains central to diagnostic confirmation when in-dicated [30,33,102]. This approach is particularly relevant because inappropriate inter-pretation of sensitization may result in overdiagnosis and unnecessary elimination diets, with potential nutritional and psychosocial consequences [103,104]. Specialist assessment may therefore improve diagnostic precision and support individualized management while limiting unnecessary dietary restrictions.

A similar principle applies to eosinophilic gastrointestinal disorders, although the diagnostic pathway differs substantially. In this setting, endoscopic assessment with histological evaluation has a central role in confirming the diagnosis, whereas allergo-logical investigations should be regarded as complementary rather than diagnostic in themselves [34,35,43,105,106]. The recommendations therefore emphasize the need to distinguish investigations required for diagnostic confirmation from those that may subsequently contribute to the characterization and management of the individual patient [43,105].

In CSU, the available evidence supports a conservative diagnostic strategy based primarily on clinical evaluation. Routine allergological investigations provide limited diagnostic benefit in the absence of a suggestive clinical history, as allergen sensitization does not necessarily identify the underlying cause of CSU [46,53–56,58]. Accordingly, skin prick tests and serum-specific IgE should be reserved for selected patients in whom an allergic trigger is clinically suspected. Likewise, investigation of autoimmune comorbidities should follow a targeted approach, with particular consideration given to thyroid dysfunction and coeliac disease and further investigations guided by symptoms, personal or family history of autoimmunity, or refractory disease [102,107]. This strategy may improve diagnostic appropriateness while limiting false-positive findings, unnec-essary investigations, and inappropriate management. Conversely, specialist referral assumes greater importance in patients with persistent or treatment-resistant urticaria, facilitating appropriate reassessment and access to advanced therapeutic options when indicated. Recurrent angioedema represents another condition in which a structured diagnostic pathway is particularly important. The distinction between hista-mine-mediated and bradykinin-mediated forms should be based on clinical character-istics and targeted investigations, particularly when hereditary angioedema is suspected. Early recognition of hereditary forms through appropriate laboratory assessment is es-sential for establishing disease-specific management and preventing potentially severe complications.

In the context of atopic dermatitis, the formulated recommendations similarly em-phasize an integrated and individualized approach, particularly in children with mod-erate-to-severe disease [68,72,95,98,108]. Standardized assessment of disease severity is important for defining subsequent management and identifying patients who may re-quire advanced treatment, including biological therapies. Specialist allergological as-sessment and targeted testing should be considered according to the individual clinical presentation rather than routinely applied to all patients. Importantly, management extends beyond diagnostic and pharmacological interventions: structured educational programmes represent an important component of care, while psychological assessment and support should be considered according to the emotional and psychosocial burden experienced by individual patients and their families.

Taken together, the findings identify a common principle across the different areas of pediatric allergology addressed by this work: diagnostic appropriateness requires a balance between avoiding indiscriminate testing and ensuring timely access to inves-tigations that can meaningfully confirm a diagnosis, identify clinically relevant comor-bidities, or modify patient management. Their integration into Diagnostic and Thera-peutic Care Pathways (PDTAs) may contribute to harmonizing clinical practice across different healthcare settings and improving collaboration between primary care and specialist services. Continuous education of healthcare professionals will be essential to facilitate implementation of these recommendations in routine clinical practice.

Limitations. Several methodological limitations should be acknowledged. First, the overall certainty of evidence varied across the different PICO questions and was fre-quently limited by the predominance of observational studies, with relatively few randomized controlled trials directly addressing diagnostic strategies. Second, many recommendations were necessarily supported by indirect evidence derived from diag-nostic accuracy studies, systematic reviews, and international clinical guidelines rather than direct comparative studies evaluating alternative diagnostic pathways. A further limitation is that the study protocol was not prospectively registered in a publicly ac-cessible systematic review registry. Nevertheless, the methodological approach, eligi-bility criteria, PICO questions, and procedures for evidence appraisal were defined a priori by the multidisciplinary working group within the framework provided by the Italian National Health Institute [109,110]. The lack of prospective registration therefore does not imply that the review was conducted without a predefined methodology, but it limits the possibility for external readers to independently verify protocol adherence and to assess whether any methodological decisions were modified during the review process. Finally, the rapid evolution of pediatric allergology, particularly regarding biologic therapies and molecular diagnostics, will require periodic updating of these recom-mendations as new evidence becomes available.

The authors must clarify the use of R packages and the RR and OR calculation algorithms; the authors should provide more detail. However, no meta-analysis is included in the results section (all results are provided narratively).

Thank you for your comment. The Methods section was accordingly revised:

In order to ensure a minimum degree of robustness and interpretability of pooled estimates, whenever at least five studies evaluated the same intervention in comparable populations and reported the same outcome, a quantitative synthesis (meta-analysis) was performed. For dichotomous outcomes, pooled effect estimates from randomized con-trolled trials were expressed as Relative Risks (RRs) with their corresponding 95% confidence intervals (95%CIs), whereas Odds Ratios (ORs) with 95%CIs were calculated for observational studies. For continuous outcomes, pooled estimates were expressed as either the Mean Difference (MD) or the Standardized Mean Difference (SMD), depending on whether the included studies used the same or different measurement scales.

Given the expected clinical and methodological heterogeneity across studies, pooled estimates were calculated using random-effects models. Between-study variance (τ²) was estimated using the restricted maximum-likelihood (REML) method, and statistical het-erogeneity was assessed using Cochran's Q and quantified by the I² statistic. I² values were interpreted as indicating low (<25%), moderate (25–50%), substantial (50–75%), or con-siderable (>75%) heterogeneity [16–18]. All statistical tests were two-sided, and p values <0.05 were considered statistically significant.

Meta-analyses were undertaken only when sufficient quantitative information was available, including the sample size, mean value, and standard deviation for each comparison group. When standard deviations were not directly reported but could be derived from other summary statistics (e.g., confidence intervals), they were estimated according to standard statistical procedures [19]. When essential data required for quantitative synthesis were unavailable, the corresponding authors were contacted by e-mail to request the missing information.

The meta-analysis was conducted by means of R (version 4.3.1) [20] and Rstudio (version 2023.06.0 Build 421; Rstudio, PBC; Boston, MA, USA) software by means of the packages meta (version 7.0) and fmsb (version 0.7.5).

We thank the Reviewer for this careful observation. We acknowledge that the original presentation of the PRISMA diagrams could generate some confusion regarding the relationship between the two identification pathways and the breakdown of the evidence reported in the final inclusion boxes. In particular, the original figures did not make sufficiently clear that the numbers reported in the final boxes classified the included records according to study/document type, irrespective of their source of identification (database searches or expert consultation). We have therefore carefully rechecked the numerical consistency of all PRISMA flow diagrams and revised the corresponding labels and figure captions to make the identification and classification processes more transparent. The revised captions now explicitly report the number of documents retained from each identification pathway and clarify that the subsequent breakdown refers to the type of evidence included in the final synthesis. The revised captions read as follows:

Figure 1. Flow-chart of included studies on the topic of IgE-mediated and non-IgE mediated food allergy and eosinophilic gastrointestinal disease. A total of 28 documents were included, com-prising 15 records identified through database searches and 13 identified through expert consulta-tion. Irrespective of their source of identification, the final evidence base comprised 14 observa-tional studies, 5 systematic reviews/meta-analyses, and 9 guidelines or consensus documents.

Figure 2. Flow-chart of included studies on urticaria and angioedema. A total of 19 documents were included, comprising 9 records identified through database searches and 10 identified through expert consultation. Irrespective of their source of identification, the final evidence base comprised 7 observational studies, 1 randomized controlled trial, 6 systematic re-views/meta-analyses, and 5 guidelines or consensus documents.

Figure 3. Flow chart of studies included into the analyses on atopic dermatitis. A total of 34 docu-ments were included in the final evidence synthesis, comprising 9 observational studies, 9 ran-domized controlled trials, 9 systematic reviews/meta-analyses, and 7 guidelines or consensus documents. Irrespective of their source of identification, documents were classified in the final box according to study/document type.

The authors are encouraged to better explain on paper how the recommendations are formulated based on the value of the evidence for the following:

Table 2. ⨁⨁◯◯ = Low certainty of evidence and interpretation – Moderate to strong in favour of specialist referral

Table 4. ⨁⨁◯◯ = Low certainty of evidence and interpretation – Conditional recommendation in favour of specialist referral

Table 5. ⨁⨁⨁◯ = Moderate certainty of evidence and interpretation – Strong recommendation in favour of endoscopic evaluation with biopsy

Table 8. ⨁⨁⨁◯ = Moderate certainty of evidence and interpretation – Strong recommendation against routine allergological investigations. Targeted testing is recommended only when supported by the clinical history

Table 9. ⨁⨁⨁◯ = Moderate certainty of evidence and interpretation – Strong recommendation in favour of a targeted diagnostic approach

We thank the Reviewer for this important observation. We agree that the relationship between certainty of evidence and recommendation strength was not sufficiently explained in the original manuscript. We have therefore expanded the Methods section to clarify that recommendation strength was determined through the GRADE Evidence-to-Decision framework and was not assigned through a direct correspondence with the GRADE certainty rating. In addition to certainty and directness of evidence, the panel considered the balance between desirable and undesirable effects, consequences of delayed or inappropriate diagnosis, patients' values and preferences, resource requirements, equity, acceptability, and feasibility.

Following the Reviewer's comment, we also reassessed the strength of the recommendations specifically identified in Tables 2, 4, 5, 8, and 9. Where the available evidence was considered insufficiently direct to support a strong recommendation, the wording was revised accordingly. In particular, the recommendation in Table 9 has been changed from strong to conditional, reflecting the predominantly indirect evidence supporting the comparative advantages of a targeted diagnostic strategy. The recommendation in Table 2 has been revised from “moderate to strong” to “conditional. Table 4 already provided a conditional recommendation based on low-certainty evidence and was therefore retained unchanged.

Conversely, the strong recommendations in Tables 5 and 8 were retained despite moderate certainty of evidence because the overall EtD assessment provided a clear direction of recommendation. For Table 5, endoscopic evaluation with biopsy represents a central component of diagnostic confirmation, and its expected diagnostic benefit was considered to outweigh the procedural burden. For Table 8, the available evidence consistently indicates limited diagnostic value of routine allergological testing in the absence of a suggestive clinical history, while a targeted strategy reduces unnecessary testing and the potential consequences of clinically irrelevant positive results. We have clarified these considerations in the corresponding EtD sections to make the rationale underlying recommendation strength more explicit.

Eventually, regarding Table 9 the main text was largely revised as follows:

(See following Table)

Quality of evidence (GRADE)

Strength

Balance of benefits/risks

Values and preferences

Impact on resources /

implementation

⨁⨁⨁◯

Conditional recommendation in favour of a targeted diagnostic approach

A targeted diagnostic strategy is expected to improve diagnostic appropriateness, facilitate the identification of clinically relevant autoimmune comorbidities, and reduce unnecessary investigations, healthcare costs, and patient burden. These expected benefits are considered to outweigh the small risk of missing uncommon conditions, provided that careful clinical assessment and appropriate follow-up are ensured.

Families generally value diagnostic pathways that avoid unnecessary testing while identifying clinically relevant conditions. Appropriate counselling improves acceptance of a selective diagnostic strategy.

Implementation is straightforward through standardized history-taking, focused first-line investigations, and selective second-level testing. This approach may reduce unnecessary laboratory investigations and promote more appropriate use of healthcare resources.