Ecological studyindependent cross-checkfull text checked I-LEG-12 in-depth review
Food allergy deaths in the UK
Food anaphylaxis in the United Kingdom: analysis of national data, 1998-2018.
Overview
Baseggio Conrado A et al.. BMJ 2021 · PMID 33597169 · DOI 10.1136/bmj.n251
Peanut/tree nut caused at least 46% of food-anaphylaxis deaths; cow's milk is now the top single cause in school children.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
UK trends
- How was the study done?
National data
- How many people, animals or samples were studied?
Whole UK
- What was actually found?
Peanut/nut ≥46% of deaths
- How large was the observed effect?
Case fatality 0.19%
- What do the results mean in everyday life?
Peanut matters for allergic people.
- What are the study's limitations?
Coding
- What does it say about Elawady's specific claim?
Partly supports
- Do the results apply to Germany and Europe?
High
- What can explicitly not be concluded from this study?
Cannot show risk for non-allergic people.
Scientific in-depth review
1. Full reference
- Authors
- Baseggio Conrado A et al.
- Journal
- BMJ
- Year
- 2021
- Identifier
- PMID 33597169 DOI 10.1136/bmj.n251
- Verification
- verified – PubMed efetch 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
UK Medical Research Council clinician scientist award (MR/K010468/1) and UK Food Standards Agency (FS101222); funders had no role. Authors report personal fees from DBV Technologies, Aimmune, Allergenis, ILSI Europe, Prota Therapeutics, FSA and others outside the submitted work.
2. Research question
Trends in food anaphylaxis admissions and deaths.
3. Study design and population
- Study design
- Ecological study – Descriptive time-trend analysis of national routinely collected data (hospital admissions databases of all four UK nations, UK Fatal Anaphylaxis Registry, adrenaline autoinjector prescriptions)
- Population
Whole UK population (England, Scotland, Wales, Northern Ireland)
- Country
United Kingdom
- Period
1998-2018 (fatality trigger analysis also includes 35 deaths from 1992-1998)
- Sample size
255,913 allergy admissions, of which 101,891 anaphylaxis admissions; 30,700 (30.1%) coded food-triggered; 152 deaths with high suspicion of food anaphylaxis (120 confirmed by coroner)
- Inclusion and exclusion criteria
Admissions with primary ICD-10 anaphylaxis codes (T78.0, T78.2, T88.6) plus T78.1/T78.4 for allergy; ED attendances without admission excluded; deaths assessed for probability of food anaphylaxis using inquest reports and media, cross-checked with ONS where possible
4. Methods and endpoints
- Exposure / intervention
Food vs non-food triggers; age, sex, nation, calendar year
- Comparator
Earlier years; non-food anaphylaxis; age groups
- Endpoints
Admission rates per 100,000, fatality rates, case fatality (deaths/admissions), triggers of fatal reactions, autoinjector prescriptions
- Statistical methods
Age-standardised rates; Poisson regression rate ratios (annual change) with 95% CIs for 1998-2018, 2009-18, 2014-18
5. Actual results
- Results
Food anaphylaxis admissions rose from 1.23 to 4.04 per 100,000/year (RR 1.057, 95% CI 1.055-1.059); in children <15 y from 2.1 to 9.2 (RR 1.066, 1.063-1.070). Food anaphylaxis deaths 0.009 per 100,000 population in 1998 vs 0.008 in 2018; case fatality fell from 0.70% to 0.19% confirmed (RR 0.931, 0.904-0.959) or 0.30% suspected (RR 0.970, 0.945-0.996). Triggers 1992-2018: at least 86/187 (46%) peanut or tree nut; food unidentified in more than a quarter; cow's milk 17/66 (26%) of deaths in school-aged children and 5% in adults; proportion due to peanut/tree nut declining since 1992 while cow's milk increased. Autoinjector prescriptions +336% (RR 1.113).
- Absolute risks
Population fatality rate from food anaphylaxis 0.008 per 100,000 per year (2018); case fatality 0.19% (confirmed) to 0.30% (suspected) of food anaphylaxis admissions in 2018
- Authors' conclusion
Hospital admissions for food anaphylaxis increased 1998-2018 but the case fatality rate decreased; cow's milk is now the most common single cause of fatal anaphylaxis in school-aged children.
6. Strengths
- National coverage of all four UK nations over 20 years
- Consistent methods over time allow trend estimation
- Deaths verified through inquest reports and registry
7. Weaknesses
- Administrative coding errors (trigger and diagnosis); ED-only cases excluded
- Admissions are a poor surrogate for anaphylaxis incidence (authors)
- Fatality ascertainment relies partly on media/inquests; Scotland/NI deaths not cross-checked
- Trigger unknown in more than a quarter of deaths, so the peanut/tree-nut share is a minimum
8. Risk of bias
- Orienting tool
- none applicable (national surveillance / routine-data trend analysis)
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
No standard RoB tool fits; structured single-reviewer judgement from full text, not formal. Trend conclusions robust; trigger proportions are minimum estimates.
| Domain | Judgement | Rationale |
|---|---|---|
| Case ascertainment (admissions) | some concerns | ICD-10 coding with miscoding of trigger and severity; threshold to admit may have changed (NICE 2011 guidance). |
| Case ascertainment (deaths) | some concerns | Registry using inquests and media; probable under- or misclassification (e.g. asthma deaths). |
| Trigger attribution | some concerns | Food unidentified in >25% of fatal cases. |
| Analytic approach | low | Standardised rates, Poisson rate ratios, consistent methods over time. |
Full-text review (25 October 2026): Added absolute population death rate (0.008/100,000 in 2018), confirmed vs suspected case-fatality RRs, unknown trigger in >25% of deaths, declining nut share since 1992, coding limitations, full funding/COI; structured validity domains (overall unchanged).
9. Assessment of the statistical interpretation
Precise national trend estimates; trigger proportions descriptive with substantial unknowns. UK data show peanut/tree nut as the leading combined trigger (at least 46%), not more than 90%.
10. Relevance to the specific claim
Confirms peanut/tree-nut importance; adds milk nuance (C-LEG-11).
11. Independent contrary findings
None.
12. Transferability to Germany and Europe
High for Europe (UK national data); the >90% figure in the claim refers to a US registry and is not replicated in UK data.
13. Overall methodological judgement
Partly supports the claim (peanut/tree nut are the leading combined trigger of food-anaphylaxis deaths) but contradicts the >90% share for the UK and shows a declining nut share and very low absolute death rates.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
None.
15. Linked fact checks
- Peanut leads all food-anaphylaxis fatalities; peanut and tree nuts caused more than 90% of food-anaphylaxis deaths in the US registry. – Studies with mixed or unclear results