Food hypersensitivity: an examination of factors influencing symptoms and temporal changes in the prevalence of sensitization in an adult sample

Food hypersensitivity (FHS), encompassing food allergy and intolerance, is prevalent in adults, with estimates ranging from 2% to 37% in Europe and about 19% in the US. While oral and skin symptoms predominate, a notable proportion of adults with self-reported FHS experience respiratory or cardiovascular symptoms, and many report at least one severe reaction. Predicting severity is challenging due to variability in food allergens and host factors, and prior evidence on predictors (e.g., specific IgE levels) is inconsistent. Respiratory comorbidities, especially asthma, have been linked to more severe, potentially life-threatening reactions. Although changes in sensitization with age have been documented in children, less is known about adult-onset FHS and changes in food sensitization over time in adults. This study investigates factors associated with severe self-reported food reactions, estimates the age of onset of FHS, and examines temporal changes in the prevalence of sensitization to common foods in an adult cohort.

Prior studies show substantial variation in FHS prevalence and causative foods across regions. Severe manifestations can occur even among those with previously mild reactions, complicating risk prediction. Some research links wheezing and asthma with respiratory symptoms during food reactions and identifies asthma as a major risk factor for food-induced anaphylaxis and fatalities. Evidence about the predictive value of specific IgE levels for reaction severity is mixed. Limited adult longitudinal data suggest possible declines in sensitization to certain foods (e.g., peanut, soy) despite stable self-reported FHS, but these analyses have often been restricted to a small number of foods or specific populations, leaving gaps regarding broader sensitization patterns and adult-onset cases.

Design and setting: Analysis of data from the European Community Respiratory Health Survey (ECRHS), a multicenter community-based cohort. Initial recruitment (ECRHS I, 1991-1993) enrolled random population samples aged 20-44 years from 55 centers in 19 countries, with clinical assessments in a random subsample plus a symptomatic subsample. Follow-ups with similar protocols occurred in ECRHS II (1998-2002) and ECRHS III (2010-2014). This study restricted analyses to the random sample who participated in ECRHS II and III. Ethics approvals were obtained locally and informed consent was provided. Participants: After excluding centers not asking FHS questions (n=871) and non-responders to FHS items (n=168) at ECRHS III, 4865 participants aged 38-67 years (52.3% female) were included for self-reported FHS analyses. Subsamples: 1673 participants in six countries had food-specific serum IgE measured at ECRHS III; 1612 had food-specific IgE measured at both ECRHS II and III. Measures: Self-reported FHS was assessed at ECRHS III with questions on lifetime reproducible illness/trouble caused by specific foods, the foods causing the main problems (up to three), symptom types (pre-defined list), latency from ingestion to reaction, and age at first episode. In Switzerland, FHS was defined by a positive response to ever having illness/trouble and a reported reaction to at least one individual food due to omission of the repeatability question. Laboratory assessments: Serum specific IgE to five food mix groups covering 25 core foods was measured using the Pharmacia CAP System; samples screening positive (≥0.35 kU/L) were further tested for individual foods. Serum specific IgE to inhalant allergens (house dust mite, cat, Timothy grass) was assessed at ECRHS II and III; birch sensitization was assessed by skin prick test at ECRHS III (wheal ≥3 mm). Definitions: - Food hypersensitivity (FHS): positive responses indicating reproducible adverse reactions to food. - Severe food reaction: symptoms within 4 hours of ingestion meeting criteria aligned with anaphylaxis guidelines, including combinations of skin-mucosal symptoms with respiratory compromise, hypotension, or severe gastrointestinal symptoms; or respiratory compromise/hypotension/laryngeal involvement without skin symptoms; or requiring an emergency injection. - Mild reaction: other reported symptoms post-ingestion. - Sensitization: specific IgE >0.35 kU/L to food or inhalant allergens; positive birch SPT as above. - Asthma/nasal/skin allergies: positive lifetime self-report. Statistical analysis: Univariate comparisons used chi-square, t-test, or Wilcoxon rank-sum tests. Multiple logistic regression (adjusting for sex and country) identified factors associated with severe versus mild reactions among those reporting symptoms. Age of onset was described overall and by food. Among those reporting FHS, the proportion sensitized to the implicated food was compared between severe and mild symptom groups (chi-square). Temporal changes in sensitization prevalence between ECRHS II and III were tested with McNemar’s test overall and by birth cohort; changes in median IgE levels to food mix groups were assessed by Wilcoxon signed-rank test. Analyses used R 3.6.3 with α=0.05.

- Prevalence: Self-reported FHS (any food) at ECRHS III was 13.5% (655/4865); 9.4% reported FHS to at least one of 25 core foods. Prevalence was higher in females (16.9%) than males (9.7%). Commonly implicated foods included cow’s milk (2.6%), hazelnut (2.1%), apple (2.1%), kiwi (1.9%), and shrimp/lobster (1.7%). - Severity: Among 611 with symptom details, 26.4% reported severe reactions (breathlessness 13.3%, emergency injection 9.5%, skin-mucosal plus GI 9.0%, faintness 7.4%). - Age of onset: Of 596 reporting age at first reaction to any food, 76.5% had onset at/after age 15. Across 1033 food-specific reactions, ~81% began at ≥15 years (78.0% for core foods). Median age of first episode varied by food (e.g., fish median 20 years [IQR 10–37.2]; peanut median 40 [IQR 27.5–45]). - Risk factors for severe reactions: In multivariable analysis, self-reported asthma was associated with severe reactions (OR 2.12, 95% CI 1.13–3.44). Earlier age at first FHS was associated with higher odds of severe reactions (per year younger: OR 1.02, 95% CI 1.01–1.03). Associations for nasal and skin allergies attenuated after adjustment. - Sensitization patterns: In the ECRHS III subsample (n=1673), sensitization to ≥1 food mix was 14.8%. Among reported food-specific reactions (n=381), 16.5% had detectable IgE to the implicated food. The prevalence of sensitization to the relevant food was lower in those with severe reactions vs mild (6/54 vs 39/183), but not statistically significant (p=0.138). - Temporal changes (n=1612 with IgE at both II and III): Overall sensitization to ≥1 food mix did not change significantly over ~10 years (14.45% to 15.88%, p=0.085). By birth cohort, 1954–1963 showed an increase (13.33% to 16.02%, p=0.019). Sensitization increased for food mix group epcx3 (banana, kiwi, apple, peach, melon) from 9.31% to 11.97% (p<0.0005), driven notably by melon (0.70% to 2.59%, p<0.0005). No significant changes in other food mixes. Inhalant allergen sensitization decreased (house dust mite, grass, cat; all p≤0.021).

This multicenter adult cohort found that FHS is commonly self-reported and that about one-quarter of affected adults experienced symptoms compatible with severe reactions, addressing the knowledge gap on severity distribution in community samples. The association between asthma and severe reactions reinforces prior observations that respiratory comorbidity elevates risk of serious outcomes during food reactions, likely via respiratory compromise. The link between earlier FHS onset and greater severity may reflect a higher likelihood of IgE-mediated allergy among earlier-onset cases and/or longer exposure time to experience severe episodes; inclusion of intolerance in FHS definitions, more common in later life, may also dilute severity among later-onset cases. Adult-onset FHS was frequent: most first reactions occurred at or after 15 years. Food-specific onset patterns (e.g., later onset for peanut and kiwi) could reflect cross-reactivity with aeroallergens (e.g., Bet v 1 homologs), often associated with milder oral allergy syndrome. Despite slight increases in some food-mix IgE medians and a rise in epcx3 sensitization (especially melon), overall food sensitization prevalence remained stable over a decade, contrasting with prior reports limited to specific countries or earlier survey comparisons. Concurrent declines in inhalant sensitization underscore differing temporal dynamics between aeroallergen and food sensitization. The relatively low concordance between self-reported severe reactions and food-specific IgE positivity highlights that self-reported FHS may overestimate IgE-mediated allergy and that biomarkers at survey times may not capture the immunologic status at the time of reactions.

In this adult population sample, 13.5% reported FHS, and approximately one-quarter of these reported ever having severe reactions. Asthma and earlier age at first FHS were associated with higher odds of severe symptoms. Most FHS began at or after age 15, indicating frequent adult-onset reactions. Over a 10-year period, the overall prevalence of sensitization to foods remained largely unchanged, though sensitization to a fruit-related mix (epcx3) and melon increased. Future research should: - Standardize severity definitions and symptom capture to improve comparability. - Use larger, well-powered cohorts with longitudinal, food-specific IgE and component-resolved diagnostics to clarify mechanisms and predictors of severity, especially in adult-onset FHS. - Examine the role of aeroallergen cross-reactivity in adult-onset fruit and nut reactions. - Integrate clinical verification (e.g., oral food challenges) to better distinguish intolerance from IgE-mediated allergy.

- Reliance on self-reported FHS and symptom proxies may introduce misclassification and recall bias. - Biomarkers were measured at survey times and may not reflect immunologic status during reaction episodes. - Self-reported FHS likely overestimates proven IgE-mediated food allergy, contributing to low observed sensitization prevalence among those with severe symptoms. - Multiple testing raises the possibility of chance findings (e.g., cohort effects). - Limited sample size for food-specific analyses reduces power to detect differences. - Some country-level variations and center exclusions may affect generalizability.