A complete guide to sleep-disordered breathing and oral appliances: a domain map running from upper-airway physiology and diagnostic responsibility through the treatment spectrum to the boundaries of the dental role｜證據鏈

A complete guide to sleep-disordered breathing and oral appliances: a domain map running from upper-airway physiology and diagnostic responsibility through the treatment spectrum to the boundaries of the dental role｜證據鏈

F-Units (the fact-unit ledger)

Expand the F1-F20 fact ledger (each entry: source # / confidence / basis / geo / period / caveat)

F1 | “Snoring” and “obstructive sleep apnoea” are not synonyms: people with this condition often snore loudly, but not everyone who snores has it; sleep-disordered breathing is a spectrum whose symptomatic expressions include snoring, raised upper-airway resistance or this condition

• Source #: #01 / #07 | confidence: high | basis: official_statement (MedlinePlus / NIH-NHLBI, verified live HTTP 200 on 2026-08-06) plus peer_reviewed (PMID 40159944) | geo: universal | period: page continuously updated, accessed 2026-08-06 / JBI Evid Synth, 2025
• caveat: MedlinePlus is a statement at patient-education level, not primary research; this entry supports only the distinction between the terms, and supports no categorisation of any individual.

F2 | The core of the mechanism: the fundamental abnormality is the inability of the upper-airway dilating muscles to withstand inspiratory negative pressure; on the collapsing side the factors include abnormal craniofacial anatomy, soft-tissue accumulation in the neck and rostral fluid shift when recumbent; the dilating side, especially the genioglossus, is negatively influenced by sleep onset itself

• Source #: #W03 | confidence: high | basis: peer_reviewed (narrative review, PMID 35609941) | geo: universal | period: J Sleep Res, 2022
• caveat: this paper is a narrative review (PublicationType: Review), not a systematic review or a primary measurement study; the description of the mechanism is a consensus-level synthesis of the field and contains no effect sizes.

F3 | Craniofacial disharmony is an important risk factor for this condition, and a meta-analysis of cephalometric imaging supports the relationship between the two; East Asian populations have more restrictive craniofacial features (cranial base, maxilla, retropalatal space)

• Source #: #W04 (risk factor and meta-analysis) / #08 (East Asian features) | confidence: moderate | basis: peer_reviewed (PMID 27039222 SR/MA; PMID 40110631 SR/MA) | geo: universal
• caveat: many parameters in the meta-analysis of craniofacial measurements show significant heterogeneity; the description of population features is background narrative in the research, and must not be used to infer anything about any individual's facial form and disease risk.

F4 | Global prevalence estimate: among adults aged 30-69, 936 million (95% CI 903-970 million) have mild to severe disease and 425 million (399-450 million) have moderate to severe disease (the zh-Hant original states these as 9.36 and 4.25, and 9.03-9.70 and 3.99-4.50, in units of 10⁸); the thresholds are AHI ≥5 and ≥15 (AASM 2012 scoring criteria)

• Source #: #W01 | confidence: moderate | basis: peer_reviewed (literature-based analysis, PMID 31300334) | geo: universal | period: Lancet Respir Med, 2019
• caveat: reliable data came from only 16 countries and 17 studies; countries without data were estimated by “matching to a similar country”; the presence or absence of symptoms was not specifically analysed; the funding source was a respiratory-device manufacturer (the FUNDING field of the original records ResMed). It is a model estimate rather than a census, and must not be used for any single country's prevalence claim.

F5 | A second set of order-of-magnitude reference points: this condition affects 4-6% of middle-aged men and 2-4% of middle-aged women; another review states that the adult population it describes is affected in up to 11% of cases, with more men diagnosed

• Source #: #11 / #10 | confidence: low | basis: peer_reviewed (PMID 40725707; PMID 41590186) | geo: universal | period: 2025 / 2026
• caveat: both are background narrative in review articles, and their abstracts do not state the original source, the sampling method or the threshold used; the “11%” item is stated in the original for the adult population of one particular country, and this article takes only its order of magnitude, making no cross-country extrapolation. This is among the weaker citations in this article.

F6 | The customary stratification of severity: mild AHI 5-15/h, moderate 15-30/h, severe >30/h; the proportion of the positional type is markedly higher in the mild stratum than in the severe stratum (49.5% / 19.4% / 6.5%)

• Source #: #W14 | confidence: low (for the authority of the cut-offs) / moderate (for the stratum proportions) | basis: peer_reviewed (retrospective chart review, multicentre, PMID 16236865) | geo: universal | period: Chest, 2005
• caveat: this paper is a two-centre retrospective chart review from 2005, not the document that established the cut-offs; it is cited because its results section states that customary stratification verbatim, not because it is the source of the grading standard. The cut-offs themselves are a convention of the field, and interpreting individual values falls to the sleep-medicine specialty.

F7 | Diagnostic responsibility: polysomnography is the standard diagnostic test for this condition in adults; diagnostic testing must go together with a comprehensive sleep evaluation and follow-up; a strong recommendation states that in the absence of polysomnography or home testing, clinical tools, questionnaires and prediction algorithms must not be used to diagnose this condition; if a single home test is negative, inconclusive or technically inadequate, polysomnography should be performed

• Source #: #04 | confidence: high | basis: clinical_guideline (AASM task force, GRADE-graded; PMID 28162150, PublicationType: Practice Guideline) | geo: universal | period: J Clin Sleep Med, 2017
• caveat: the recommendations address adults; this entry is the basis for diagnostic responsibility and constitutes no determination of any individual's suitability for any test.

F8 | The performance of screening tools and its limits: an SR/MA of one validated questionnaire (17 studies, 9,206 people) gives sensitivities of 90% / 94% / 96% for detecting any / moderate-to-severe / severe disease in the sleep-clinic population, with corresponding negative predictive values of 46% / 75% / 90%; the higher-level appraisal of the evidence holds that the accuracy and clinical utility of all potential screening tools are uncertain, and that no RCT compared screening with no screening

• Source #: #W05 (performance) / #W06 (uncertainty) | confidence: moderate (performance) / high (the statement of uncertainty) | basis: peer_reviewed (PMID 26658438 SR/MA; PMID 28118460 USPSTF evidence report and SR) | geo: universal | period: PLoS One, 2015 / JAMA, 2017
• caveat: the sensitivity figures come from sleep-clinic and surgical populations and cannot be extrapolated to the general community population; this article deliberately does not list that questionnaire's items, scores or probability tables, so as not to constitute an instruction for self-diagnosis.

F9 | The limits of home and portable monitoring: an SR/MA of level IV devices (24 studies, 2,068 people) gives an AHI bias range of -14.8 to 10.6 events/h against polysomnography; at the AHI ≥5 threshold, sensitivity 67.5-100% and specificity 25-100%; only 7 (29%) tested in the home setting

• Source #: #W07 | confidence: moderate | basis: peer_reviewed (SR/MA, PMID 29318566) | geo: universal | period: Sleep Breath, 2018
• caveat: this entry concerns “level IV (fewer-channel)” devices and is not equivalent to what the guideline calls a “technically adequate home sleep apnoea testing device”; the two must not be read interchangeably. The authors state that current evidence is insufficient to support stand-alone use.

F10 | The division of responsibility (the core of this article): sleep physicians prescribe oral appliances (primary snoring = STANDARD; intolerance of CPAP or preference for an alternative therapy = STANDARD); a qualified dentist uses a custom, titratable appliance (GUIDELINE); a qualified dentist provides oversight to survey for dental-related side effects or occlusal changes (GUIDELINE); the sleep physician performs follow-up sleep testing (GUIDELINE); both sides instruct patients to return for periodic visits (GUIDELINE)

• Source #: #02 | confidence: high | basis: clinical_guideline (joint AASM/AADSM seven-member task force, approved by the boards of both academies, GRADE-graded; PMID 26094920) | geo: universal | period: J Clin Sleep Med, 2015
• caveat: the indexing discrepancy is disclosed honestly — the PubMed PublicationType of this paper is marked `Meta-Analysis, Systematic Review` and not `Practice Guideline`, although the text describes itself as a Clinical Practice Guideline and assigns STANDARD / GUIDELINE grades clause by clause; after human adjudication it is recorded as clinical_guideline (consistent with point 6 of section 5 of the anchor file). This entry is the 2015 version; whether a newer version exists was not determined in this round.

F11 | The treatment spectrum is plural, not a choice between two: at official patient-education level it lists lifestyle changes, oral appliances, surgery and breathing devices; at guideline level it lists positive airway pressure devices, oral appliances, behavioural treatments, surgery and/or adjunctive treatments, and requires the patient to be included in the decision process; presence or absence and severity must be determined before treatment; this condition should be regarded as a chronic disease requiring long-term, multidisciplinary management

• Source #: #01 / #03 | confidence: high | basis: official_statement plus clinical_guideline (PMID 19960649, PublicationType: Practice Guideline) | geo: universal | period: accessed 2026-08-06 / J Clin Sleep Med, 2009
• caveat: #03 is a 2009 document, and the AASM has since issued separate guidelines on diagnosis (2017) and on PAP (2019); this entry takes only its framework of “the strategy is plural and requires shared decision-making”, and does not cite its individual operational clauses.

F12 | The positive airway pressure side: PAP is used in adults with this condition who have excessive sleepiness (STRONG); it must be based on a diagnosis established by objective sleep testing; after initiation and during treatment there must be follow-up including troubleshooting and monitoring of objective efficacy and usage data

• Source #: #05 | confidence: high | basis: clinical_guideline (AASM task force, GRADE; PMID 30736887, PublicationType: Practice Guideline) | geo: universal | period: J Clin Sleep Med, 2019
• caveat: the recommendation is limited to adults “with excessive sleepiness”; this article does not cite the detail of its other conditional recommendations (such as CONDITIONAL for those with impaired quality of life).

F13 | Where the other pathways sit in the evidence: weight reduction (a 20% fall in BMI associated with a 57% fall in AHI, with a smaller marginal effect thereafter; wide prediction intervals) / positional modification (benefit for the supine-predominant type, but CPAP more effective; a reliable diagnosis of the supine-predominant type should come first) / oral appliance versus positional therapy (non-supine AHI and sleepiness score lower, no significant difference on the other eight measures) / maxillomandibular advancement surgery (45 studies, 518 people; mean AHI change -47.8; success 85.5%, cure 38.5%) / hypoglossal nerve stimulation (30 papers, 26 single-arm, 549 people; one device -20.14 short-term and -15.91 events/h long-term)

• Source #: #W17 / #W15 / #W16 / #W18 / #W19 | confidence: low to moderate (differing item by item) | basis: peer_reviewed (PMID 38908268 / 28012784 / 38849827 / 26606321 / 39401661) | geo: universal | period: 2016-2024
• caveat: 73.5% of the surgical sample were patients in whom previous surgery had failed, a difficult-to-treat population, and this must not be extrapolated into what an ordinary patient should expect from initial treatment; the neurostimulation data are predominantly single-arm (26 of 30 papers), with only 4 RCTs; the authors of the weight-reduction data state that a precise relationship could not be established. Every part of this entry is a population-level result and constitutes no treatment recommendation, body-weight target or determination of surgical suitability for any individual.

F14 | The mechanism of oral appliances and differences between types: mandibular advancement devices increase the area of the airway, bringing the soft palate, tongue and hyoid bone forward and activating the masseter and submental muscles to prevent closure; adjustable and custom-made appliances give better results than fixed and prefabricated ones; monobloc devices give rise to more adverse events, though generally mild and transient

• Source #: #W12 | confidence: moderate | basis: peer_reviewed (systematic review, PMID 26228493; 22 articles from the past 10 years included) | geo: universal | period: Laryngoscope, 2016
• caveat: the PubMed PublicationType of this paper is `Systematic Review`, although its own LEVEL OF EVIDENCE field records NA; the description of the mechanism is a synthesising statement rather than an effect size from imaging measurement.

F15 | The efficacy of oral appliances: effective in comparison with controls (Cochrane, 16 studies, 745 people: reduced daytime sleepiness, improved AHI; systematic review of RCTs in mild-to-moderate disease: custom, titratable appliances effectively reduce AHI and improve sleep-related outcomes; East Asian SR/MA, 12 studies, 382 people: pooled mean AHI reduction of 19.1), but weaker in comparison with CPAP (weaker at reducing AHI, weaker at improving minimum oxygen saturation; all comparative SRs in the umbrella review consistently report CPAP as more efficacious), while patient preference favours oral appliances and symptom scores show no significant difference

• Source #: #06 / #09 / #08 / #07 | confidence: moderate | basis: peer_reviewed (PMID 16437488 / 42108628 / 40110631 / 40159944) | geo: universal | period: 2006 / 2026 / 2025 / 2025
• caveat: the three statements must be set out together, and quoting any one of them alone is prohibited. Cochrane states that all its included studies have the shortcomings of small samples, under-reporting and lack of blinding; the umbrella review's overall GRADE quality is very low to moderate, and most of the reviews it included were published over a decade ago; the concluding sentence of the East Asian SR/MA states explicitly that complete resolution is unlikely. The PubMed PublicationType of #09 (PMID 42108628) is marked only `Journal Article` and not `Systematic Review`, and its character as a systematic review is judged from its title and the self-description of its methods section — a discrepancy between the indexing label and the self-description of the content, disclosed honestly here.

F16 | Weaker efficacy does not mean worse on every endpoint: a network meta-analysis (51 studies, 4,888 people) shows CPAP associated with a 2.5 mmHg fall in systolic blood pressure and oral appliances with a 2.1 mmHg fall, with no statistically significant difference between them; an extra 1 hour of CPAP per night is associated with an additional 1.5 mmHg fall

• Source #: #W11 | confidence: moderate | basis: peer_reviewed (Network Meta-Analysis, PMID 26624827) | geo: universal | period: JAMA, 2015
• caveat: only 1 study compared CPAP and oral appliances directly (a further 3 were three-arm), so the comparison between the two rests mainly on indirect inference; blood pressure is a surrogate endpoint and must not be read as a claim of equivalence for cardiovascular events or mortality.

F17 | Adherence is a difficulty common to both sides: on the CPAP side an SR (82 papers, 1994-2015) gives a non-adherence rate of 34.1% with no significant improvement over twenty years, and behavioural intervention improving adherence by about 1 hour per night on average; on the oral appliance side an SR/MA (32 RCTs) gives an overall mean dropout rate of 0.171 (mean follow-up 4.1 months), with custom appliances worn more hours per night (6.418 vs 5.107) and the dropout rate rising significantly over time

• Source #: #W08 / #W10 | confidence: moderate | basis: peer_reviewed (PMID 27542595 SR; PMID 34865235 SR/MA/meta-regression, PROSPERO CRD42020199866) | geo: universal | period: 2016 / 2022
• caveat: the definitions and time windows differ (the CPAP side is a “non-adherence rate based on 7 hours of sleep per night”, the oral appliance side a “dropout rate with a mean follow-up of 4.1 months”), so the values cannot be compared directly; the GRADE quality on the oral appliance side is very low to moderate.

F18 | The association between adherence and health outcomes: in an IPD meta-analysis (4,186 people), the first MACCE on intention-to-treat analysis was similar between groups (HR 1.01, 95% CI 0.87-1.17), while an on-treatment analysis using a marginal structural model showed good adherence (≥4 hours per day) associated with a reduced risk (HR 0.69, 95% CI 0.52-0.92)

• Source #: #W09 | confidence: moderate (association) / low (causation) | basis: peer_reviewed (IPD Meta-Analysis, PMID 37787793) | geo: universal | period: JAMA, 2023
• caveat: the on-treatment marginal-structural-model analysis is a weighted observational inference, not a randomised comparison; the authors' wording is “associated with”, not causal. It must not be written as “wearing it enough prevents cardiovascular events”. The population is a secondary-prevention population who already have cardiovascular disease (82.1% men) [Fn131], and it must not be extrapolated to primary prevention in the general population.

F19 | The long-term occlusal side effects of oral appliances (demonstrated): on the qualitative side (SR, only 14 papers with ≥4 years of follow-up) upper incisor retroclination, lower incisor proclination, decreased overjet and overbite and change in the total occlusal contact area are reported, with the conclusion that the effect is significant both statistically and clinically; on the quantitative side (SR/MA, 42 studies / 23 pooled, long-term defined as ≥6 months) overbite decreased by 0.87 mm, overjet by 0.86 mm, upper incisors retroclined by 2.58° and lower incisors proclined by -2.67°, accumulating progressively across the year intervals; skeletal changes were not significant, with only a tendency to clockwise rotation of the mandible

• Source #: #13 / #14 / #10 (qualitative supplement) | confidence: high (that the change exists) / moderate (the effect size) | basis: peer_reviewed (PMID 38090465 SR; PMID 39327689 SR/MA; PMID 41590186 Review) | geo: universal | period: 2023 / 2025 / 2026
• caveat: both directions have to be set out: it must not be played down (this is a change significant both statistically and clinically that accumulates with the years, which is why the guideline carries a clause on dentist oversight), and it must not be overstated (the magnitude is a sub-millimetre decrease in overbite and overjet and an incisor inclination of 2-3 degrees, with no significant change at the skeletal level). The PubMed PublicationType of #13 is `Review`; the confidence intervals for the effect sizes are in the footnote spans.

F20 | The association between sleep bruxism and this condition: the evidence is in tension and there is no consensus — a 2025 SR (11 studies, ROBINS-I / GRADE) found that sleep bruxism prevalence is consistently higher in the population with this condition than in the general population, with the mechanistic hypothesis being autonomic arousals and neurotransmitter dysregulation; a 2024 SR/MA (PRISMA 2020, 14 of 2,260 records screened in) found no difference in the odds ratio against controls (OR 1.23, 95% CI 0.47-3.20), with no difference in any severity stratum either

• Source #: #11 / #12 | confidence: low | basis: peer_reviewed (PMID 40725707; PMID 39182463) | geo: universal | period: J Clin Med, 2025 / Sleep Med Rev, 2024
• caveat: both are recent reviews from 2024-2025; this is not a case of a newer version superseding an older one but of the contemporaneous evidence itself lacking consensus. #11 states of itself that inconsistent diagnostic criteria and moderate methodological quality limit the strength of its conclusion and that the causal relationship remains unclear; #12 states of itself that the included studies are of low quality and that the finding of “no correlation” still requires further research, adding the reminder that this should not exempt clinicians from the exact diagnosis of concomitant sleep conditions. Writing it as an assertion in either direction is prohibited.

Compliance note

• This article is health education and a synthesis of medical developments, is general oral-health information, does not solicit medical business, does not constitute medical advertising, and does not constitute diagnostic or treatment advice.
• This article provides no diagnostic instruction: it lists no self-assessment scale, no screening-questionnaire items or score interpretation, and no criterion of the form “how many is too many” or “how many seconds is too long”. Under the international guideline, this condition must not be diagnosed with a questionnaire or a prediction algorithm in the absence of objective sleep testing; diagnosis falls within the responsibility of the sleep-medicine specialty.
• This article gives no amounts, charges or reimbursement information, recommends no institution, practitioner, brand or product, contains no identifiable case, and carries no third party's subjective comment on any treatment. Every device in the text is described by a generic category name (oral appliance / mandibular advancement device / positive airway pressure device), with no brand or model named; the monetary values of the cost-effectiveness research are deliberately not quoted.
• This article contains no medication, dose or procedural instruction, and offers no advice on appliance selection, amount of advancement or adjustment frequency — those decisions have to be assessed by a dentist and coordinated with a sleep-medicine specialist.
• The citations involving weight reduction are population-level statements of association and constitute no weight-loss prescription or body-weight target for anyone.
• All statements about efficacy are research conclusions at population level; in comparison with CPAP the oral appliance is usually less efficacious and complete resolution is unlikely, and this article makes no promise of efficacy of any kind.
• The actual treatment and its effect vary from person to person and must be decided by a dentist after assessment (working together with a sleep-medicine specialist).
• This article is a draft; it has not passed the publication gate, the four language versions are not yet complete, and it is for internal review only.

Source list

Date of access / live verification: 2026-08-06 (all sources were verified live on the same day with curl / the PubMed E-utilities, HTTP 200, and the verbatim spans passed programmatic comparison)

Pre-verified sources from the anchor file (`ida-pillars/anchors/P24-anchors.md`, 15 entries)

Statement from section 0 of the anchor file: under the owner's decision of 2026-08-06 that the whole line runs on the global track, no Taiwanese regulation, national health insurance or health-bureau content has been used as the basis for any medical or institutional claim in this article; local systems are uniformly linked downstream to the corresponding TW canonical card and P12. The single source at the official_statement tier in this article is the public patient-education page of a United States federal government research agency (NIH/NLM), whose content is universal health-education information and contains no institutional claim specific to any country.

WRITER-ADDED SOURCES (20 added by this article, with live verification evidence)

Reason for the additions: the 15 entries in the anchor file cover the five patient questions, but the pillar layer additionally needs (i) the anatomical and physiological mechanism of upper-airway collapse together with the craniofacial axis, (ii) the scale of the field (global prevalence) and the limits of how it is estimated, (iii) the verbatim cut-offs of the severity strata, (iv) where screening tools and home testing sit and their quantified limits, (v) the adherence data for CPAP and for oral appliances respectively, (vi) blood pressure as the counterpoint showing that “a gap in efficacy is not a gap in health outcome”, (vii) where the four treatment pathways of weight reduction, position, surgery and neurostimulation sit in the evidence, (viii) the signals for seeking care (crash risk) and the comorbidity side, and (ix) the international consensus document as higher-level support for the cross-disciplinary division of labour. All 20 entries below were verified live on this machine with curl (HTTP 200) and passed programmatic verbatim comparison, and the anchor file was not modified.

(The clickable URL format for each PubMed source is `https://pubmed.ncbi.nlm.nih.gov//`.)