Cross-sectional studyindependent cross-checkfull text checked I-INTC-13 in-depth review
What is linked to sleep apnoea in children
Risk factors of obstructive sleep apnea syndrome in children.
Overview
Xu Z et al.. J Otolaryngol Head Neck Surg 2020 · PMID 32131901 · DOI 10.1186/s40463-020-0404-1
Enlarged tonsils and adenoids, obesity and male sex were linked to sleep apnoea; breastfeeding was too, probably due to other factors.
Does the study support the specific claim? No · Methodological quality: low
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Identify OSAS risk factors in children
- How was the study done?
Cross-sectional with sleep studies
- How many people, animals or samples were studied?
1578 children
- What was actually found?
Adenotonsillar hypertrophy, obesity, male sex, breastfeeding
- How large was the observed effect?
not reported
- What do the results mean in everyday life?
Formula feeding was not identified as the cause of enlarged tonsils.
- What are the study's limitations?
Confounding; clinic sample
- What does it say about Elawady's specific claim?
Does not support the formula claim
- Do the results apply to Germany and Europe?
Low
- What can explicitly not be concluded from this study?
Causation
Scientific in-depth review
1. Full reference
- Authors
- Xu Z et al.
- Journal
- J Otolaryngol Head Neck Surg
- Year
- 2020
- Identifier
- PMID 32131901 DOI 10.1186/s40463-020-0404-1
- Verification
- verified – PubMed efetch 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
Funded by the Chinese National Key Research and Development Plan (2017YFC0112502), Beijing hospitals authority paediatric centre (XTYB201807) and Capital Health Research and Development special funding; authors declare no competing interests.
2. Research question
Which factors are associated with OSAS in snoring children?
3. Study design and population
- Study design
- Cross-sectional study – Single-centre hospital-based cross-sectional study with polysomnography (part of registered cohort NCT02447614)
- Population
Children aged 2-15 years with snoring undergoing polysomnography at Beijing Children's Hospital sleep centre, mostly referred by ENT or respiratory specialists
- Country
China (Beijing)
- Period
1 March 2016 - 28 February 2017
- Sample size
1578 analysed (1583 eligible, 5 excluded for missing data); 1008 with OSAS (abstract states 1009), 570 without
- Inclusion and exclusion criteria
Snoring children 2-15 y referred for PSG; excluded prior adenotonsillectomy, craniofacial abnormalities, neuromuscular disease, syndromes (Down, Crouzon, Pierre Robin)
4. Methods and endpoints
- Exposure / intervention
Questionnaire risk factors incl. breastfeeding (defined as fully breastfed for >3 months since birth), preterm birth, ETS, allergic rhinitis, parental education; examination-based obesity, neck/waist measures, tonsil grade (Brodsky 3+/4+) and adenoid obstruction (fibro-laryngoscopy)
- Comparator
Snoring children without OSAS (OAHI <1); for breastfeeding, children not fully breastfed >3 months (not specifically formula-fed)
- Endpoints
OSAS defined as obstructive apnea-hypopnea index >=1 on overnight PSG scored blind to clinical findings
- Statistical methods
Chi-square univariate screening (P<0.1) then multivariable logistic regression; ORs with 95% CIs
5. Actual results
- Results
Multivariable ORs for OSAS: breastfeeding 1.72 (95% CI 1.26-2.36), P=0.0007; obesity 1.71 (1.27-2.29); tonsillar hypertrophy 4.16 (3.07-5.63); adenoid hypertrophy 1.41 (1.06-1.88); snoring >=3 months 1.48 (1.00-2.18); male sex 1.27 (1.00-1.61); preterm 1.59 (0.99-2.57, NS). Low maternal education more common among breastfeeding than non-breastfeeding parents (41.65% vs 30.12%, P<0.0001).
- Absolute risks
OSAS prevalence 63.9% (1008/1578) in this referred sample; prevalence by breastfeeding status not reliably extractable from the flattened table
- Authors' conclusion
Snoring duration, male sex, obesity, breastfeeding and adenotonsillar hypertrophy were independent risk factors.
6. Strengths
- Polysomnography-confirmed outcome scored blind to clinical data
- Large clinical sample with broad age range
- Endoscopic adenoid assessment by ENT specialist
7. Weaknesses
- Highly selected referral population (snoring children, 64% OSAS); not generalisable to the general population
- Cross-sectional; exposure recalled by parents years later
- Breastfeeding vs not fully breastfed >3 months, not breastfeeding vs formula; no data on infant cereals
- Likely residual confounding by socioeconomic status (authors' own point)
- Data-driven variable selection (P<0.1) and multiple testing
- Flattened Table 2 percentages internally inconsistent with text (e.g. obesity)
- Outcome is OSAS, not adenotonsillar hypertrophy or surgery
8. Risk of bias
- Orienting tool
- ROBINS-E
- 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
- high
- Rationale
Single-reviewer structured ROBINS-E judgement from full text; not a formal assessment (would require two independent reviewers and protocol).
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | high | Only covariates passing univariate screening were adjusted; socioeconomic status, occupation, access to care not adjusted; breastfeeding linked to lower parental education. |
| Exposure measurement | some concerns | Parent-reported breastfeeding history obtained by interview, possibly years after infancy (children up to 15 y). |
| Selection of participants | high | Hospital sleep-centre referral sample of snoring children; both cases and non-cases already symptomatic; collider-type selection possible. |
| Post-exposure interventions | not applicable | Cross-sectional observational design. |
| Missing data | low | Only 5 of 1583 excluded for missing data. |
| Outcome measurement | low | Standardised PSG scored by staff unaware of clinical findings. |
| Selection of reported result | some concerns | Many candidate factors; stepwise-like selection; hypothesis (breastfeeding protective) contradicted, reported transparently. |
Full-text review (25 October 2026): Added study period, setting, exposure definition (fully breastfed >3 months vs not, not formula), adjusted ORs with CIs (breastfeeding OR 1.72, 1.26-2.36), funding; noted n discrepancy (1008 vs 1009) and flattened-table inconsistency; ROBINS-E domains (overall unchanged: high).
9. Assessment of the statistical interpretation
Breastfeeding OR 1.72 (1.26-2.36) is statistically significant in this sample but likely confounded and selection-biased; it is an association in referred snoring children and does not show that breastfeeding causes OSAS or that formula protects.
10. Relevance to the specific claim
In this sample, breastfeeding (not formula) was associated with OSAS, the main indication for adenotonsillectomy; this contradicts the idea that formula feeding is the reason children need these operations, although confounding is likely.
11. Independent contrary findings
Other studies not reviewed.
12. Transferability to Germany and Europe
Low (Chinese clinical sample).
13. Overall methodological judgement
Limiting evidence only: in referred Chinese snoring children, full breastfeeding >3 months was not protective (positively associated with OSAS). It does not test formula or infant cereals, nor surgical need, so it cannot refute or confirm the claim directly; it shows no support for it.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | low |
| Risk of bias | high |
| Precision of results | not applicable |
| Directness | indirect |
| Reproducibility | unknown |
| Consistency with the overall evidence | unknown |
| Regional transferability | low |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Population-based prospective data on infant feeding (formula, cereals) and adenotonsillar hypertrophy or adenotonsillectomy.
15. Linked fact checks
- Children who eat 'proper food' from infancy do not develop enlarged tonsils, adenoids or appendicitis; formula milk and infant cereals are the reason children need these operations. – Studies contradicting or limiting the claim