Identifying unmet needs in food protein–induced enterocolitis syndrome using search query analysis and a large language model Authors: Thinh H. Nguyen, Mia Vaida, Bernardo Sousa-Pinto, Thinh Doan, Danielle Crethers, Brichelle Peña, Emeline Kao, Wanda Phipatanakul, Erin Syverson, Elena Crestani, Lisa M. Bartnikas Published: 2026 ABSTRACT BackgroundFood protein–induced enterocolitis syndrome (FPIES) is a non–IgE-mediated food allergy with many unanswered clinical and research questions. Existing educational resources, patient-facing education materials, and guidelines may not fully reflect families’ information needs. Digital data and artificial intelligence (AI) offer new ways to identify public concerns in rare diseases.ObjectiveWe developed a framework utilizing online search data and AI-assisted analysis to generate data-informed questions for future patient advice guidance and educational resource development in FPIES, and evaluate these questions with parents/guardians.MethodsWe collected Google Trends data from January 2023 to December 2024 to identify search queries related to FPIES in the United States. Using ChatGPT, query terms and frequencies were analyzed to generate 10 representative questions reflecting public interest. Parents/guardians of children with FPIES enrolled onto the Boston Children’s Hospital FPIES Registry were surveyed to evaluate these questions.ResultsForty-four parents/guardians completed the survey (93.2% mothers; median [interquartile range] child age was 3.0 [2.0, 5.0] years). Nearly half (43.2%) reported more than one FPIES trigger food. Nine of 10 AI-generated questions were rated “important” or “very important” by >90% of participants, confirming their relevance. Questions most frequently reported as difficult to find answers for were managing reactions, outgrowing FPIES, and distinguishing FPIES from other allergies. Similar topics were selected as research priorities.ConclusionA framework built from Google Trends data and AI-supported analysis can generate questions that align closely with parent/guardian priorities, offering a practical approach to informing patient advice guidance and educational resource development, with the ultimate goal of supporting patient-centered guideline development in FPIES. Food protein–induced enterocolitis syndrome (FPIES) is a non–IgE-mediated food allergy characterized by delayed vomiting, lethargy, and hypotension after ingestion of specific food proteins.1 Although often presenting in infancy, FPIES can persist into childhood and occasionally adulthood, leading to significant distress for affected families.2 Diagnosis is clinical because there are no validated biomarkers or standardized tests. Current management primarily involves strict avoidance of the trigger food and supervised oral food challenges to assess tolerance over time. Despite increasing recognition of FPIES in clinical practice, many questions remain unanswered regarding its pathophysiology, optimal diagnostic approach, and evidence-based management.3 Development of clinical practice guidelines traditionally relies on expert consensus and systematic literature reviews.4,5 Accordingly, identification of questions to be addressed by clinical guidelines typically relies on expert consensus and evidence synthesis. However, these methods may not fully capture the lived experiences and unmet information needs of patients and parents/guardians. For rare and emerging conditions such as FPIES, where patient experiences often outpace formal research, this gap can result in patient-facing educational resources and advice materials that do not fully align with the real-world concerns of families.6,7 While inclusion of patients and their families in guideline panels is increasingly being recognized as important, their representativeness is often limited (that is, they are typically not able to account for all diversity of patients and their views).8,9 Recognizing the importance of patient partnerships in creating meaningful practice guidelines, guideline developers including the Joint Task Force on Practice Parameters have begun to proactively include patient stakeholders in the guideline development process to provide critical perspectives regarding the importance of specific outcomes and practical experience with proposed interventions.9 Guidelines that address patient preferences provide opportunities for clinicians and patients to identify where their values align and differ, and facilitate engagement in discussion and shared decision-making, ultimately leading to increased guideline adherence by patients and families.8,9 Online search behaviors represent a powerful, real-time source of information reflecting what patients and parents/guardians seek to understand about their condition. Google Trends, a freely accessible tool, allows researchers to analyze the frequency of specific searches and identify temporal or regional trends in public interest. By quantifying collective search behavior, investigators can infer what topics are most relevant or concerning to the general population. While analysis of large volumes of searches text can be a challenging task, recent advances in artificial intelligence (AI) and natural language processing provide new opportunities to systematically incorporate public perspectives into health research.10 Sousa-Pinto et al11 in 2024 proposed a framework for using large language models to support the development of health guideline questions by analyzing common search queries. Building on this concept, our study extended the use of AI to a public-driven approach by integrating Google Trends data with AI-assisted question generation. We hypothesized that search-derived questions generated through this method would align closely with the priorities and informational gaps identified by families of children with FPIES. To test this, we derived AI-generated questions from online searches at a national level throughout the United States and evaluated these through a survey of parents of children enrolled onto the Boston Children’s Hospital FPIES Registry. To our knowledge, this represents the first study to combine population-level search data, AI-based analysis, and direct patient feedback to validate a framework for identifying unmet needs and informing future patient advice guidance and educational resource development in FPIES, and ultimately leading to the development of balanced and effective guidelines that reflect the patient stakeholder perspective. Methods Google Trends data were collected to identify search queries related to FPIES in the United States from January 2023 to December 2024. The search-derived queries reflect aggregated search behavior at the US national level. For each month in this period, relative search frequency scores and associated related queries were extracted to quantify how often these terms were queried and to characterize patterns of public interest (Fig 1, A). Data were obtained for both the “FPIES” search term and the “food protein–induced enterocolitis syndrome” topic, each categorized under the Web Search section of Google Trends. Using ChatGPT-4 (OpenAI, San Francisco, Calif), we analyzed the collected Google Trends query terms, top related queries, and their frequencies to generate 10 representative questions that capture major themes of public interest and concern regarding FPIES (Table I). We prompted ChatGPT-4 to review all monthly Google Trends top related queries from January 2023 through December 2024. From these inputs, the model identified the most searched topics and emerging themes and distilled them into top 10 guideline questions.Fig 1Workflow and relevance of search-derived questions. A, Overview of workflow integrating Google Trends data and AI analysis to identify unmet information needs in FPIES. Google Trends data from January 2023 to December 2024 were analyzed by ChatGPT to generate 10 representative questions, which were then evaluated in parent/guardian survey. B, Importance ratings of 10 search-derived questions among 44 parents/guardians of children with FPIES.Table IExample of prompt and outputInitial prompt: I am going to ask you to help generate 10 guideline questions that you think we should address for patients based on Google Trends search data. Files are labeled from January 2023 to December 2024. Each file represents a separate Excel spreadsheet.Final output:Most searched topics:•Symptoms (“FPIES symptoms”).•What is FPIES/meaning.•Diagnosis.•Allergy and reaction.•FPIES in babies/infants.•Foods to avoid/food charts.Emerging themes:•Parents searching for basic definitions and understanding.•Questions about diagnosis, symptoms, and management.•Specific concerns about age groups (babies vs adults).•Diet guidance like safe foods and what to avoid.Top 10 guideline questions for patients with FPIES∗1.What is FPIES?2.What are common symptoms of FPIES in babies and children?3.How is FPIES diagnosed, and what tests are used?4.How can I tell if my baby’s symptoms are FPIES or another food allergy?5.What are the most common trigger foods for FPIES?6.What foods are safe for babies and children with FPIES?7.How should I manage FPIES reactions at home, and when should I go to the emergency room?8.Can children outgrow FPIES, and when does that typically happen?9.Is FPIES considered an allergy even if there’s no IgE reaction?10.What support and resources are available for families managing FPIES?∗These top 10 guideline questions address what people are actively searching for. Workflow and relevance of search-derived questions. A, Overview of workflow integrating Google Trends data and AI analysis to identify unmet information needs in FPIES. Google Trends data from January 2023 to December 2024 were analyzed by ChatGPT to generate 10 representative questions, which were then evaluated in parent/guardian survey. B, Importance ratings of 10 search-derived questions among 44 parents/guardians of children with FPIES. Table IExample of prompt and outputInitial prompt: I am going to ask you to help generate 10 guideline questions that you think we should address for patients based on Google Trends search data. Files are labeled from January 2023 to December 2024. Each file represents a separate Excel spreadsheet.Final output:Most searched topics:•Symptoms (“FPIES symptoms”).•What is FPIES/meaning.•Diagnosis.•Allergy and reaction.•FPIES in babies/infants.•Foods to avoid/food charts.Emerging themes:•Parents searching for basic definitions and understanding.•Questions about diagnosis, symptoms, and management.•Specific concerns about age groups (babies vs adults).•Diet guidance like safe foods and what to avoid.Top 10 guideline questions for patients with FPIES∗1.What is FPIES?2.What are common symptoms of FPIES in babies and children?3.How is FPIES diagnosed, and what tests are used?4.How can I tell if my baby’s symptoms are FPIES or another food allergy?5.What are the most common trigger foods for FPIES?6.What foods are safe for babies and children with FPIES?7.How should I manage FPIES reactions at home, and when should I go to the emergency room?8.Can children outgrow FPIES, and when does that typically happen?9.Is FPIES considered an allergy even if there’s no IgE reaction?10.What support and resources are available for families managing FPIES?∗These top 10 guideline questions address what people are actively searching for. Example of prompt and output •Symptoms (“FPIES symptoms”). Symptoms (“FPIES symptoms”). •What is FPIES/meaning. What is FPIES/meaning. •Diagnosis. Diagnosis. •Allergy and reaction. Allergy and reaction. •FPIES in babies/infants. FPIES in babies/infants. •Foods to avoid/food charts. Foods to avoid/food charts. •Parents searching for basic definitions and understanding. Parents searching for basic definitions and understanding. •Questions about diagnosis, symptoms, and management. Questions about diagnosis, symptoms, and management. •Specific concerns about age groups (babies vs adults). Specific concerns about age groups (babies vs adults). •Diet guidance like safe foods and what to avoid. Diet guidance like safe foods and what to avoid. 1.What is FPIES? What is FPIES? 2.What are common symptoms of FPIES in babies and children? What are common symptoms of FPIES in babies and children? 3.How is FPIES diagnosed, and what tests are used? How is FPIES diagnosed, and what tests are used? 4.How can I tell if my baby’s symptoms are FPIES or another food allergy? How can I tell if my baby’s symptoms are FPIES or another food allergy? 5.What are the most common trigger foods for FPIES? What are the most common trigger foods for FPIES? 6.What foods are safe for babies and children with FPIES? What foods are safe for babies and children with FPIES? 7.How should I manage FPIES reactions at home, and when should I go to the emergency room? How should I manage FPIES reactions at home, and when should I go to the emergency room? 8.Can children outgrow FPIES, and when does that typically happen? Can children outgrow FPIES, and when does that typically happen? 9.Is FPIES considered an allergy even if there’s no IgE reaction? Is FPIES considered an allergy even if there’s no IgE reaction? 10.What support and resources are available for families managing FPIES? What support and resources are available for families managing FPIES? These top 10 guideline questions address what people are actively searching for. To evaluate these search-derived questions generated from national-level queries, we conducted a survey using REDCap among parents/guardians of children with FPIES enrolled onto the Boston Children’s Hospital FPIES Registry. FPIES phenotypes were defined for participants on the basis of 2017 international consensus guidelines,5 as follows: “Acute FPIES: Vomiting 1-4 hours after trigger food ingestion, sometimes followed by delayed diarrhea within 24 hours, lethargy, or pallor”; “Chronic FPIES: Intermittent vomiting, sometimes with diarrhea, when trigger food is ingested on a regular basis”; “Atypical FPIES is diagnosed in people with FPIES reactions but who also have IgE sensitization (positive blood allergy testing or skin testing) to the trigger food.” Participants were asked to rate how important each question was to them as parents/guardians managing FPIES on a 5-point Likert scale that ranged from “not at all important” to “very important”; to select the top 3 questions they found most difficult to get answered; and to identify the top 3 topics they thought should be prioritized for future research. Survey responses were analyzed to assess the perceived relevance and importance of the search-derived questions. These data were used to validate the proposed approach as a potential tool for identifying unmet information and research needs among families affected by FPIES. Results Participant characteristics Using the workflow outlined in the Methods and in Fig 1, A, we generated 10 candidate questions about unmet needs in FPIES (Table II) reflecting aggregated national US search behavior, then evaluated them among families in the Boston Children’s Hospital FPIES Registry. The demographics of our participants are shown in Table III. A total of 44 parents/guardians of children with current FPIES completed the survey. Most respondents (93.2%) were mothers. The median (interquartile range) child age was 3.0 (2.0-5.0) years. Nearly half of the children (43.2%) had more than one FPIES food trigger, 19 specific food triggers were represented within the children, and several FPIES phenotypes were represented (all had acute FPIES, with 6.8% chronic FPIES and 22.7% atypical FPIES).Table IITop 10 questions on FPIES generated by AINo.Original questionSimplified question theme1What is FPIES?What is FPIES2What are common symptoms of FPIES in babies and children?Common FPIES symptoms3How is FPIES diagnosed, and what tests are used?FPIES diagnosis and testing4How can I tell if my baby’s symptoms are FPIES or another food allergy?FPIES vs other food allergies5What are the most common trigger foods for FPIES?Common FPIES triggers6What foods are safe for babies and children with FPIES?Safe foods in FPIES7How should I manage FPIES reactions at home, and when should I go to the emergency room?Managing FPIES reactions8Can children outgrow FPIES, and when does that typically happen?Outgrowing FPIES9Is FPIES considered an allergy even if there’s no IgE reaction?Is FPIES allergy10What support and resources are available for families managing FPIES?FPIES support and resourcesTable IIIParent/guardian and child characteristicsCharacteristic [no. of subjects]ValueParent characteristicsRelationship to child [n = 44] Mother41 (93.2) Father2 (4.5) Other1 (2.3)Age (years), median (IQR) [n = 44]38 (35.8, 41.0)Race [n = 44] Asian3 (6.8) Black or African American0 White41 (93.2) Multi-race0Hispanic or Latino [n = 43]0Highest education level [n = 44] High school degree/GED/other high school equivalent4 (9.1) Junior/community college degree1 (2.3) Undergraduate degree11 (25.0) Graduate degree28 (63.6)Annual household income [n = 42] <$50,0004 (9.5) $50,000 to $74,9991 (2.4) $75,000 to $99,9993 (7.1) $100,000 to $149,9992 (4.8) $150,000 to $199,9995 (11.9) ≥$200,00027 (64.3)Child characteristicsAge (years), median (IQR) [n = 40]3 (2.0, 5.0)Race [n = 44] Asian1 (2.3) Black or African American0 White35 (79.5) Multirace8 (18.2) Hispanic or Latino [n = 43]2 (4.7)How long ago diagnosed with FPIES? (years), median (IQR) [n = 43]2 (1.0, 4.0)Type or types of FPIES diagnosed with [n = 44]∗ Acute44 (100.0) Chronic3 (6.8)Child ever diagnosed with atypical FPIES [n = 44] Yes10 (22.7) No26 (59.1) Don’t know or unsure8 (18.2)Foods currently avoided due to FPIES [n = 44]∗ Oat15 (34.1) Egg9 (20.5) Peanut8 (18.2) Cow’s milk7 (15.9) Avocado5 (11.4) Banana5 (11.4) Tree nuts5 (11.4) Rice4 (9.1) Fish3 (6.8) Poultry3 (6.8) Chickpea2 (4.5) Soy2 (4.5) Apple1 (2.3) Beef1 (2.3) Lentil1 (2.3) Papaya1 (2.3) Pea1 (2.3) Wheat1 (2.3) Zucchini1 (2.3)No. of current FPIES food triggers [n = 44] 125 (56.8) 29 (20.5) 38 (18.2) 42 (4.5)Data are presented as nos. (%) unless otherwise indicated.IQR, interquartile range.∗Can select more than one response. Table IITop 10 questions on FPIES generated by AINo.Original questionSimplified question theme1What is FPIES?What is FPIES2What are common symptoms of FPIES in babies and children?Common FPIES symptoms3How is FPIES diagnosed, and what tests are used?FPIES diagnosis and testing4How can I tell if my baby’s symptoms are FPIES or another food allergy?FPIES vs other food allergies5What are the most common trigger foods for FPIES?Common FPIES triggers6What foods are safe for babies and children with FPIES?Safe foods in FPIES7How should I manage FPIES reactions at home, and when should I go to the emergency room?Managing FPIES reactions8Can children outgrow FPIES, and when does that typically happen?Outgrowing FPIES9Is FPIES considered an allergy even if there’s no IgE reaction?Is FPIES allergy10What support and resources are available for families managing FPIES?FPIES support and resources Top 10 questions on FPIES generated by AI Table IIIParent/guardian and child characteristicsCharacteristic [no. of subjects]ValueParent characteristicsRelationship to child [n = 44] Mother41 (93.2) Father2 (4.5) Other1 (2.3)Age (years), median (IQR) [n = 44]38 (35.8, 41.0)Race [n = 44] Asian3 (6.8) Black or African American0 White41 (93.2) Multi-race0Hispanic or Latino [n = 43]0Highest education level [n = 44] High school degree/GED/other high school equivalent4 (9.1) Junior/community college degree1 (2.3) Undergraduate degree11 (25.0) Graduate degree28 (63.6)Annual household income [n = 42] <$50,0004 (9.5) $50,000 to $74,9991 (2.4) $75,000 to $99,9993 (7.1) $100,000 to $149,9992 (4.8) $150,000 to $199,9995 (11.9) ≥$200,00027 (64.3)Child characteristicsAge (years), median (IQR) [n = 40]3 (2.0, 5.0)Race [n = 44] Asian1 (2.3) Black or African American0 White35 (79.5) Multirace8 (18.2) Hispanic or Latino [n = 43]2 (4.7)How long ago diagnosed with FPIES? (years), median (IQR) [n = 43]2 (1.0, 4.0)Type or types of FPIES diagnosed with [n = 44]∗ Acute44 (100.0) Chronic3 (6.8)Child ever diagnosed with atypical FPIES [n = 44] Yes10 (22.7) No26 (59.1) Don’t know or unsure8 (18.2)Foods currently avoided due to FPIES [n = 44]∗ Oat15 (34.1) Egg9 (20.5) Peanut8 (18.2) Cow’s milk7 (15.9) Avocado5 (11.4) Banana5 (11.4) Tree nuts5 (11.4) Rice4 (9.1) Fish3 (6.8) Poultry3 (6.8) Chickpea2 (4.5) Soy2 (4.5) Apple1 (2.3) Beef1 (2.3) Lentil1 (2.3) Papaya1 (2.3) Pea1 (2.3) Wheat1 (2.3) Zucchini1 (2.3)No. of current FPIES food triggers [n = 44] 125 (56.8) 29 (20.5) 38 (18.2) 42 (4.5)Data are presented as nos. (%) unless otherwise indicated.IQR, interquartile range.∗Can select more than one response. Parent/guardian and child characteristics Data are presented as nos. (%) unless otherwise indicated. IQR, interquartile range. Can select more than one response. Relevance of search-derived questions Overall, the US national search-derived questions were rated as highly relevant by the participants surveyed within the Boston Children’s Hospital FPIES Registry. Nine of 10 questions received an importance rating of “important” or “very important” from more than 90% of respondents. The distribution of importance ratings for each question is shown in Fig 1, B. Topics that are the most difficult to find answers for Parents/guardians were also asked to select the 3 questions that were most difficult to get answered in their day-to-day experience seeking information about FPIES. The most frequently selected areas of difficulty were how to manage reactions, when children outgrow FPIES, and how to differentiate FPIES from other allergies, including food allergy phenotypes that may present with overlapping gastrointestinal symptoms. These selections are summarized in Fig 2, A.Fig 2Caregiver-identified knowledge gaps and research priorities. A, Counts of selections for each question as 1 of 3 hardest-to-find answers for among parents/guardians in Boston Children’s Hospital FPIES Registry (n = 44). Each parent/guardian could choose up to 3 items. Bars show total number of times each question was selected, irrespective of within-respondent rank. B, Counts of selections for each question as 1 of 3 priority topics for future research. Each parent/guardian could choose up to 3 items. Bars show total number of times each question was selected, using same method as in (A). Caregiver-identified knowledge gaps and research priorities. A, Counts of selections for each question as 1 of 3 hardest-to-find answers for among parents/guardians in Boston Children’s Hospital FPIES Registry (n = 44). Each parent/guardian could choose up to 3 items. Bars show total number of times each question was selected, irrespective of within-respondent rank. B, Counts of selections for each question as 1 of 3 priority topics for future research. Each parent/guardian could choose up to 3 items. Bars show total number of times each question was selected, using same method as in (A). Research priorities identified by parents/guardians To inform future work that is responsive to public need, parents/guardians selected their top 3 research priorities from the same question set. The leading priorities were when children outgrow FPIES, diagnosis and testing strategies, and differentiating FPIES from other allergies. These priorities are shown in Fig 2, B. Concordance across importance, difficulty, and priority Taken together, 3 themes surfaced consistently across metrics. Questions related to the natural history of outgrowing FPIES, approaches to accurate diagnosis and testing, and clear differentiation from other allergic conditions were rated highly important, frequently identified as difficult to answer in current practice, and prioritized for future research. Discussion In this study, we present a new data-driven framework that combines Google Trends data and AI-supported analysis to identify and validate unmet information needs in FPIES. We used ChatGPT to organize common searches into structured and patient-centered questions that we found to align closely with what parents/guardians identify as their most important and research relevant concerns. This finding suggests that collective search behavior can reflect genuine patient (or parent/guardian) curiosity and unmet needs, particularly in rare conditions where formal research and public education often lag behind clinical realities. Our results support the idea that AI and infodemiology can complement traditional methods of identifying patient preferences and unmet needs and developing patient-centered clinical guidelines.8,12 Current FPIES recommendations are built primarily from expert opinion and small clinical studies. Although this is the best scientific evidence available, it may overlook the day-to-day questions that families struggle with, such as how to manage acute reactions, how to recognize tolerance, and how to distinguish FPIES from other food allergies. Incorporating data from public search behavior appears to introduce a new layer of insight that directly reflects what families are actively seeking to understand about their condition. While most of the generated questions were background questions and guideline recommendations are formulated on the basis of foreground questions,13 it is important to note, first, that the identified questions on FPIES can support the identification of topics that future patient-centered education and guidelines could focus on, and second, that previous studies have noted that Google Trends data can support the identification of foreground questions.11 This work builds on the concept introduced by Sousa-Pinto et al11 in a study that showed that large language models can assist in generating clinical guideline questions from common web searches. Our study differs by involving parents/guardians in evaluating the importance of generated questions. We applied a public-driven approach, beginning with search data that represent what families and parents/guardians are already asking online, then validated these findings directly with affected families through the Boston Children’s Hospital FPIES Registry. This two-step process closes the loop between AI, public data, and parent/guardian feedback, creating a scalable method for identifying relevant and meaningful research questions. The consistent overlap between questions rated as most important, most difficult to answer, and highest priority for research highlights key areas of uncertainty in FPIES. Families repeatedly emphasized their difficulty in managing acute reactions, predicting when their child will outgrow FPIES, and differentiating FPIES from other allergic or gastrointestinal disorders. These topics represent both clinical and emotional pain points for parents/guardians and point to the need for better diagnostic tools, clearer communication, and prospective studies of natural history. The alignment between digital and survey findings supports the reliability of this combined approach and its potential for broader use in other rare conditions. Search-derived questions reflecting public uncertainties and interest may not directly equate with the process and methodology of professional society development bodies. We propose that this novel approach of leveraging digital search behavior and AI-assisted analysis can identify areas of diagnostic, management, and research priorities among public queries, thereby helping inform topic selection and priority setting during development of patient advice guidance and educational materials, as well as in patient-centered guideline development. This approach may serve as an adjunct tool to better understand patient-centered concerns, inform parent-facing educational resources, and better prepare clinicians to engage in compassionate discussions and shared decision-making with their patients. Established guideline methodologies, including evidence synthesis and expert panel deliberation, remain essential for formulating formal society guideline recommendations. However, incorporating patient perspectives and preferences is of growing importance in guideline development, as evidenced by their inclusion in the 2023 atopic dermatitis,14 2023 anaphylaxis,15 and 2025 inborn errors of immunity16 guidelines from the Joint Task Force on Practice Parameters. We acknowledge that some of the AI-generated questions resemble background informational questions rather than foreground questions that form the basis of traditional guidelines. However, particularly in rare conditions such as FPIES, persistent background uncertainty often reflects meaningful informational gaps that may influence educational materials, guideline development, and future research prioritization. We therefore believe the framework has relevance not only to informing guideline topic selection but also to improving patient-facing educational resources and communication needs from their health care providers. Our findings highlight the opportunity for clinicians caring for patients with FPIES to strive to provide open, clear, and compassionate communication to their patients and families.17 The reality that patients and families are going online to search these questions that may appear at surface-level fundamental suggests that the information providers believe they are communicating to their patients may not be meeting their needs, and may also suggest knowledge gaps among clinicians. This study has limitations. Google Trends reports relative search frequency of a sample of data rather than absolute numbers, and public interest can fluctuate with media coverage or awareness campaigns. Future research may assess data collection from other infodemiology data sources, such as social media posts. The survey sample included 44 participants and was drawn from a single tertiary-care registry, which may not represent all families affected by FPIES. We also did not assess participants’ awareness or use of existing FPIES educational resources, including materials from advocacy organizations or emergency action plans, which may influence perceived informational gaps and ratings of question importance. The rating of US nationally derived search questions by the participants from the Boston Children’s Hospital FPIES Registry may reflect temporal or local factors amplified within this specific patient group. However, it was notable that participants strongly agreed with the importance of the nationally derived questions, supporting the utility of this proof-of-concept study in identifying key questions. Our study objective was not to define national research priorities but to evaluate whether search-derived themes align with caregiver priorities within a real-world clinical population. Future studies are needed to evaluate the relevance of search questions in other patient populations within the United States and also internationally, with the dual goals of identifying specific local or group-specific needs and determining the relevance to the broader population. The survey was only available in English, and respondents were predominantly White, non-Hispanic, highly educated, and high income, which may limit the generalizability of our results. These demographic characteristics mirror what is reported in FPIES publications to date,1 suggesting reduced awareness or underdiagnosis in vulnerable and underserved populations. Despite these limitations, the strong concordance between search patterns and parents/guardians’ priorities suggests that this framework can reliably identify meaningful information gaps, which supports the applicability of our findings. By integrating Google Trends data, AI analysis, and parent/guardian feedback, we developed and evaluated a new method to uncover unmet information and research needs in FPIES. The strong alignment between online searches and patient priorities demonstrates that collective online behavior can offer valuable insights into what families truly want to know and rate as important about their condition. This approach creates a bridge between digital data, AI, and participatory research, offering a practical and scalable model for identifying gaps in patient education and research focus. Applying this framework to other rare or emerging diseases could help future patient advice guidance and educational materials, patient-centered guideline development, and research efforts better reflect the real-world concerns of patients and families.Key messages•Current management guidance in rare diseases such as FPIES relies on limited studies and expert consensus and may not fully address the day-to-day questions and information gaps families face.•We present and validate a scalable framework that uses public search behavior and AI-generated questions to identify patient-centered gaps and priorities in FPIES.•Leveraging online search data and AI reveals unmet needs in FPIES, which can be used as a novel tool to inform the development of patient advice guidance and educational resources, research priorities, and patient-centered guidelines. Key messages •Current management guidance in rare diseases such as FPIES relies on limited studies and expert consensus and may not fully address the day-to-day questions and information gaps families face.•We present and validate a scalable framework that uses public search behavior and AI-generated questions to identify patient-centered gaps and priorities in FPIES.•Leveraging online search data and AI reveals unmet needs in FPIES, which can be used as a novel tool to inform the development of patient advice guidance and educational resources, research priorities, and patient-centered guidelines. •Current management guidance in rare diseases such as FPIES relies on limited studies and expert consensus and may not fully address the day-to-day questions and information gaps families face. Current management guidance in rare diseases such as FPIES relies on limited studies and expert consensus and may not fully address the day-to-day questions and information gaps families face. •We present and validate a scalable framework that uses public search behavior and AI-generated questions to identify patient-centered gaps and priorities in FPIES. We present and validate a scalable framework that uses public search behavior and AI-generated questions to identify patient-centered gaps and priorities in FPIES. •Leveraging online search data and AI reveals unmet needs in FPIES, which can be used as a novel tool to inform the development of patient advice guidance and educational resources, research priorities, and patient-centered guidelines. Leveraging online search data and AI reveals unmet needs in FPIES, which can be used as a novel tool to inform the development of patient advice guidance and educational resources, research priorities, and patient-centered guidelines. Disclosure statement Supported by the National Institutes of Health (grant T32AI007512-38 to T.H.N.), the International FPIES Association (to E.C. and L.M.B.), the Boston Children’s Hospital Division of Immunology (to E.C. and L.M.B.), and the Myles & Gertrude Hansen Foundation (to E.C. and L.M.B.). Declaration of generative AI and AI-assisted technologies in the writing process: ChatGPT (OpenAI, San Francisco, Calif) was used to assist in improving clarity, grammar, and structure during the preparation of the report. The authors reviewed all suggestions and are responsible for the final content and interpretation. Disclosure of potential conflict of interest: W. Phipatanakul is a consultant advisory for Teva, Genentech, Novartis, GSK and Regeneron, for asthma-related therapeutics. The rest of the authors declare that they have no relevant conflicts of interest.