A complete guide to verifying oral-health claims: an evidence map of popular practices, self-treatment and common myths｜證據鏈

A complete guide to verifying oral-health claims: an evidence map of popular practices, self-treatment and common myths｜證據鏈

F-Units (fact ledger; each entry: source # / confidence / basis / geo / period / caveat)

F1 | The key element of evidence-based medicine is the hierarchical system of categorising evidence; different types of study answer different kinds of clinical question

• Source #: #W9 | confidence: moderate | basis: peer_reviewed (PMID 31797840) | geo: universal | period: J Contemp Dent Pract, 2019
• caveat: that paper is a short appraisal with limited abstract content (the whole abstract is four sentences), and its nature is that of a critical appraisal of “level-of-evidence” classification systems (the title itself is A Critical Appraisal on the "Level-of-evidence" Classification Systems) — what this article takes from it is the consensus background sentence stated in its opening (which in the original carries a citation marker), not an endorsement of hierarchical systems by it, nor an authoritative definition of any particular hierarchy (such as OCEBM or GRADE); the verification framework of this article does not rest its weight on this single entry, and each of the five questions has its own independent source (#W5 / #W1 / #W2 / #W6).

F2 | Industry-sponsored drug and device studies more often have efficacy results favourable to the sponsor (RR 1.27, 95% CI 1.17–1.37) and favourable conclusions (RR 1.34, 95% CI 1.19–1.51); within industry-sponsored studies there is less agreement between results and conclusions; this bias cannot be explained by standard risk-of-bias assessments

• Source #: #W5 | confidence: high | basis: peer_reviewed (Cochrane methodology review, 75 papers included, PMID 28207928) | geo: universal | period: Cochrane Database Syst Rev, 2017
• caveat: this is a shift in the overall distribution across fields (drugs and devices), not specific to dentistry, and not “industry-sponsored studies are all untrustworthy”; the efficacy-results entry is moderate-quality evidence and the conclusions entry low-quality evidence (as labelled by the original); it must not be used to accuse any individual study or author.

F3 | The sugar industry is influencing dental research, oral health policy and professional organisations; the current treatment-dominated, interventionist approach in high-income countries is not tackling the underlying causes of disease; clearer and more transparent conflict-of-interest policies are needed

• Source #: #W6 | confidence: moderate | basis: peer_reviewed (the second paper in the Lancet oral health Series, PMID 31327370) | geo: universal | period: Lancet, 2019
• caveat: it is a Series commentary / review rather than original research; it is a structural commentary on a system of care and must not be read as an assessment of any individual practitioner or institution; it is a statement made in 2019.

F4 | Oil pulling: a meta-analysis of 25 RCTs and 1,184 people shows a “probable benefit” for gingival health, but chlorhexidine remained superior in reducing plaque, and the overall quality of the evidence was very low; another systematic review of 31 studies shows moderate reductions in microbial load and improved gingival health, though findings varied compared with chlorhexidine, and its conclusion was that traditional practices “may offer benefits similar to conventional methods”, with a request for cautious interpretation

• Source #: #05 (meta-analysis) / #06 (systematic review) | confidence: low (both papers state their own limits on evidence quality / methodology) | basis: peer_reviewed (PMID 37635453 / 41670379) | geo: universal | period: 2024 / 2026
• caveat: the “short-term non-inferiority” sentence in the abstract of #06 comes from its hypothesis section, not its conclusion section (Fn20); this article has corrected the way it is cited against the original, and an author's hypothesis must not be written as an author's conclusion. The “more than half” in #05 is at trial level, not participant level (Fn14a) — the original reports no participant-level distribution, this article does not convert it into a number of people, and the claim that holds is “in more than half of the included trials the participants had no reported oral health issues”, with reservation required when extrapolating to people with pre-existing disease. The conclusion sentence of #06 is an umbrella conclusion covering several classes of practice including miswak, herbal preparations and oil pulling (Fn20b / Fn21), and the one that obtained the stronger result was miswak (Fn21a); it must not spill over into a report card for oil pulling. Neither paper positions oil pulling as a replacement for routine oral hygiene.

F5 | Charcoal and charcoal-based dentifrices: of 118 candidate articles, 13 involved brushing with raw charcoal or soot and none met the inclusion criteria; 3 reported deleterious outcomes (increased caries, enamel abrasion, non-quantified negative impact), while within the same batch 2 offered non-specific caries reductions and 1 indicated no adverse effects (Fn23a / Fn23b, findings in the other direction recorded alongside); internet advertisements contain unsubstantiated therapeutic claims (antibacterial / antifungal / antiviral / oral detoxification); the conclusion is that the available clinical and laboratory data are insufficient to substantiate the safety and efficacy claims

• Source #: #07 | confidence: moderate (the conclusion is clear, but it is a literature review rather than a systematic review) | basis: peer_reviewed (PMID 28599961) | geo: universal | period: J Am Dent Assoc, 2017
• caveat: it is “neither safety nor efficacy substantiated”, not “efficacy in doubt but safe”; none of those 13 studies (including the 3 with harm, the 2 with caries reduction and the 1 with no adverse effect) met the inclusion criteria (Fn22), so results in all three directions are equally unusable as evidence, and the same qualification applies where this article lists deleterious outcomes in §6; the search closed at February 2017 (Fn25a) and this article has not searched for any newer systematic review of human safety since, so it must not be implied that more complete safety research has become available in recent years; this entry is also the citable basis in this article for the “oral detoxification” claim.

F6 | Baking soda (sodium bicarbonate) dentifrice: a literature review states that it is effective and safe for removing tooth stains and whitening, and records its acid-buffering capacity, antibacterial action at high concentrations and relatively lower abrasivity; a separate in-vitro study assigned specimens to six groups (three regular toothpastes, one reference slurry, two whitening toothpastes), the relative dentine abrasion values across all six groups combined ranging from 26 to 166, and that study also records that the abrasion value of the whitening toothpaste containing hydrogen peroxide was not high compared with the regular toothpaste, and that abrasion values vary with the mechanism and composition of the whitening agents

• Source #: #W3 (efficacy and safety) / #W11 (abrasion range) | confidence: low | basis: peer_reviewed (PMID 29056186 / 35793939) | geo: universal | period: J Am Dent Assoc, 2017 (supplement 148(11S):S20-S26) / Dent Mater J, 2022
• caveat: three reasons for labelling confidence low: (1) #W3 is a literature review, not a systematic review or meta-analysis; (2) #W3 was published in a supplement issue and its PubMed record carries no conflict-of-interest statement — under the framework of §1-5 of this article, that is a bibliographic fact to be taken into the weighting (this entry is a verifiable bibliographic description, not an accusation against the authors or the journal); (3) “relatively lower abrasivity” is a relative comparison, not an absolute value. #W11 is an in-vitro study (bovine dentine specimens, 10,000 brushing strokes, n=8) and must not be equated directly with the result of long-term use in a human mouth; its range of 26–166 is the union across six tested toothpaste groups (regular and whitening), and must not be hung on the single class of “whitening toothpaste” (Fn31 / Fn31a); the result of that study pointing the other way (the abrasion value of the whitening toothpaste containing hydrogen peroxide was not high compared with the regular toothpaste, Fn31b) is presented alongside in §2-2 and §6. This article names no product.

F7 | The direct-to-consumer orthodontic model: its definition is precisely “without routine in-person clinical supervision”; a thematic analysis of posts (165 screened in from 721) derives seven themes with negative sentiment predominating, and the authors note that reliance on self-monitoring and peer advice may delay recognition of complications; a Spanish cross-sectional questionnaire of 101 people shows that the majority's expectations were not met and that they would not recommend it, that cost was the primary motivation, and that many users attended for a dental consultation because of complications

• Source #: #08 (post analysis) / #09 (questionnaire) | confidence: low (both are non-controlled studies of self-selected samples) | basis: peer_reviewed (PMID 42405800 / 40217832) | geo: universal | period: 2026 / 2025
• caveat: neither study can yield a complication rate, and neither is a controlled trial of effectiveness against supervised orthodontics; #08 is a self-selected sample from a social platform (negative experiences are more likely to be posted) and #09 is 101 people in a single country; this article therefore provides no rate of any kind, makes no comparison of effectiveness, and names no brand.

F8 | The case of closing an anterior diastema with an elastic band: a 9-year-old boy developed severe acute periodontitis of the maxillary central incisors because the band migrated apically, and the two incisors were finally extracted; the author recommends that elastic gap bands without fixed orthodontic appliances should not be used to treat an anterior diastema

• Source #: #W4 | confidence: low (a single case report) | basis: peer_reviewed (Case Reports, PMID 30477782) | geo: universal | period: Am J Orthod Dentofacial Orthop, 2018
• caveat: a single case report; it does not represent an incidence rate; the reason for including it is that self-treatment behaviour of this kind is inherently hard to study prospectively, so a case report is the direct evidence retrievable in the literature.

F9 | The evidence level on which correct first aid for an avulsed permanent tooth stands: it is one of the most serious dental injuries, and prompt and correct emergency management is essential to obtaining the optimal outcome; a systematic review records that the key steps include the replantation time window, the appropriate storage medium and correct handling technique, and that knowledge of this kind is generally inadequate across most populations

• Source #: #04 (IADT consensus guidelines) / #10 (systematic review, 19 studies, 5,752 participants, 10 countries) | confidence: high | basis: clinical_guideline + peer_reviewed (PMID 32460393 / 41583179) | geo: universal | period: 2020 / 2025
• caveat: #04 states of itself that no warranty is given that adherence to the guidelines will produce favourable outcomes (Fn46, a negative self-disclosure by the source); the awareness proportions in #10 (23% of parents to over 80% of dental professionals; only 2–37% of non-dental populations willing to attempt) are a pooling across 10 countries, and the distribution differs between countries, so they must not be applied to any particular country; under red line 1 of the non-conflict rule this entry deliberately does not restate the specific content of the three key steps (the figures of the time window, the order of preference for storage media, the handling technique), because that answer belongs to canonical card KM-DENTAL-30 and this layer only links down to it — the same treatment principle as F12's avoidance of the material-performance figures of #12; this article states only the level-of-evidence fact that “the time window and the storage medium are key variables in the literature”.

F10 | The case of self-treatment with self-fabricated prostheses cemented with super glue: the author explicitly warns that patients must not attempt self-treatment for aesthetics with self-fabricated prostheses, because severe and irreversible hard and soft tissue adverse reactions may occur

• Source #: #11 | confidence: low (a single case report) | basis: peer_reviewed (Case Reports, PMID 16836177) | geo: universal | period: J Oral Implantol, 2006
• caveat: the original is a preventive warning (`may occur`), not an observed outcome — that abstract does not record the actual tissue outcome for that patient, so throughout this article (including the risk passage of §2-4) the wording is “may occur”, and it must not be upgraded into “has been recorded as severe and irreversible”; the object of attribution is the behaviour of “self-treating with self-fabricated prostheses”, not the adhesive on its own; a single case report from 2006, which must not be used as a claim about incidence or statistical significance; this article's search obtained no systematic review or cohort study designed around “treating teeth oneself with super glue” or “using an over-the-counter temporary filling material oneself”, and those two evidence gaps are left honestly blank, with no judgement made as to whether either works or does not work.

F11 | Amalgam fillings: the official patient-information page states that when a filling is in good condition and there is no decay beneath it, removal or replacement is not recommended, because removal causes unnecessary loss of healthy tooth structure and exposure to a temporary increase in mercury vapour; the association's position confirms that mercury bound in the alloy poses a negligible health risk under normal conditions of use and is safe for the general population (excluding people allergic to amalgam components and those with severe renal impairment)

• Source #: #01 (official page) / #02 (association policy and position statements) | confidence: high | basis: official_statement + clinical_guideline (https://www.fda.gov/medical-devices/dental-devices/dental-amalgam , the page is marked Content current as of 02/18/2021 / PMID 42200317) | geo: universal | period: 2021 (the page's current-as-of date) / J Dent Res, 2026
• caveat: the PubMed PublicationType of #02 was measured on this machine as `Journal Article` only (not labelled Consensus Statement) — the wording on this point in §0 of the anchor file P23-anchors.md does not match what was measured on this machine, so this article corrects it against the measurement and describes its nature as “the association's policy and position statements” instead (its authority comes from the organisation's signature on the document itself, not from the PubMed label); #01 is the patient-information page of a single country's competent authority, and this article takes from it only the clinical recommendations and the group list that apply universally, quoting no country's regulations or reimbursement content.

F12 | The other axis of amalgam: concerns exist about toxicity to human health and the environment, and the Minamata Convention on Mercury recommends a phase-down of amalgam use in dentistry; the association supports a responsible phase-down strategy at the same time as confirming safety

• Source #: #12 (background section of a Cochrane overview) / #02 (association position) | confidence: high | basis: clinical_guideline + peer_reviewed (PMID 42444634 / 42200317) | geo: universal | period: Cochrane Database Syst Rev, 2026 / J Dent Res, 2026
• caveat: “safe” and “phase down” are two different axes and coexist without contradiction: the former is the clinical risk to an individual patient, the latter the environmental and occupational exposure policy across the life cycle of the material; neither extreme — “amalgam is entirely harmless” or “amalgam must be taken out” — is a permissible simplification. This article deliberately does not cite the material-performance comparison figures of #12 (failure rates, follow-up durations and so on) — that set of figures belongs to KM-DENTAL-12 / KM-DENTAL-16 and domain article P04, and under red line 1 of the non-conflict rule this layer only links down to them.

F13 | Removing dental amalgam with a high-speed drill generates particulate that volatilises significant amounts of mercury vapour, at levels that frequently exceed the safety thresholds of several jurisdictions; the authors characterise this as a significant, under-recognised source of localised exposure

• Source #: #W10 | confidence: low (a single study, measuring exposure rather than health outcomes) | basis: peer_reviewed (PMID 31346345) | geo: universal | period: J Occup Med Toxicol, 2019
• caveat: what this entry measures is occupational exposure among dental staff, not health outcomes in patients, and it must never be used to claim that “a patient having amalgam removed will be poisoned”; the sole respect in which it points the same way as the official position is that “the act of removal itself generates mercury vapour” (consistent with Fn56), and this article uses it only under that restriction. A single study, with no meta-analysis and no independent replication. Under the framework of §1-5 of this article, the bibliographic facts of its provenance are disclosed alongside (the same yardstick as the disclosure of #W3 in F6): all four authors' affiliations as registered on PubMed are private dental practices, with no academic or occupational-health institution named; the funding section of the PMC full text (PMC6637613) states `Funding for the study was provided by the authors.` (self-funded), while its CoiStatement records that the authors declare no competing interests. This is a verifiable bibliographic description, not an accusation against the authors.

F14 | The place of the focal-infection theory in the history of science: the “era of focal infection” from the 19th to the early 20th century led to widespread extraction and limited the development of endodontics, and the theory was eventually refuted because it rested only on anecdotal evidence; the modern consensus is that the relationship between oral and systemic health is more complex than the classical theory, and interdisciplinary teams have established standardised scoring and referral procedures

• Source #: #14 (narrative review) / #03 (interdisciplinary consensus) | confidence: moderate | basis: clinical_guideline + peer_reviewed (PMID 35888650 / 38044216, the latter measured as containing the PublicationType `Consensus Statement`) | geo: universal | period: Medicina, 2022 / Int Dent J, 2024
• caveat: the consensus in #03 was formulated by an interdisciplinary working group in a single country (the Czech Republic), and its specific scoring algorithm must not be extrapolated as an international standard; this article takes from it only the qualitative point that “modern practice is a standardised procedure rather than wholesale rejection or wholesale acceptance”; #14 is a narrative review, not a systematic review.

F15 | The direction of root canal treatment and systemic inflammatory burden: a narrative review judges that convincing evidence supports successful root canal treatment having a beneficial impact on systemic health by reducing the inflammatory burden, thereby dismissing the misconception of the focal infection theory; in a systematic review and meta-analysis of 20 clinical studies, 18 reported a reduction in inflammatory markers after treatment, and the meta-analysis showed a significant reduction in hs-CRP at 6 months after treatment (the 1-month time point did not reach significance), with no significant change in the other markers

• Source #: #14 / #13 (PROSPERO CRD42024574082) | confidence: low (the GRADE self-assessment of #13 is low) | basis: peer_reviewed (PMID 35888650 / 42026989) | geo: universal | period: 2022 / 2026
• caveat: this article obtained no cohort study or systematic review directly investigating the association between root canal treatment and cancer incidence, and therefore nowhere claims that “it has been established not to cause cancer”, provides no related epidemiological figures, and presents only the chain of argument “the historical theory has been refuted + the modern mechanistic evidence points the other way”, claiming no exhaustive exclusion of an association; the hs-CRP result of #13 carries a timing qualifier (significant at 6 months after treatment, not at 1 month; Fn73 / Fn73a), the time point has been added in both the main text and the FAQ, and it must not be written as an unconditional “significant reduction after treatment”; the low certainty of #13 is mainly due to heterogeneity among the primary studies and inadequate control of confounders such as marginal periodontitis and smoking (as stated by the original); #14 also records that bacteraemia associated with untreated apical periodontitis may negatively affect systemic health (Fn68a), a direction opposite to “treated teeth are harmful”.

F16 | NICO: current evidence is insufficient to support it as a distinct, validated clinical entity; 41 studies show recurrent methodological limitations (absence of control groups, non-standardised criteria, limited histological evidence, interpretive bias), the proposed mechanisms lack independent validation, and diagnostic tools remain unvalidated; another systematic review (29 studies) could identify no gold standard diagnostic means, found all studies to be observational in design, and rated all of them poor quality; the aetiology section of that review also records one convergent exception — hereditary coagulopathies, identified as potential risk factors in 5 studies

• Source #: #W1 (scoping review, 41 studies) / #W2 (systematic review, 29 studies) | confidence: high (with respect to the single conclusion of “insufficient evidence”) | basis: peer_reviewed (PMID 41849956 / 33893686) | geo: universal | period: Arch Oral Biol, 2026 / Oral Dis, 2022
• caveat: “the evidence is insufficient to support this diagnostic entity” is not the same as “the patient's pain does not exist” — #W1 states plainly that many people so diagnosed may be more appropriately classified as having persistent idiopathic facial pain or painful trigeminal neuropathy (Fn79); the remission range reported in #W2 (as low as 66%, as high as all cases) comes from observational studies without control groups (Fn4), and the follow-up periods of the individual studies span 2 months to 18 years (Fn85b), the follow-up period being a necessary condition for reading that range, which must not be omitted, and the range must not be listed beside or compared with controlled effectiveness data; this article cites the range only to show how divergent it is; the statement on aetiology must be written together with its exception (hereditary coagulopathies, Fn85a), or it would be more absolute than the source itself; this article gives no diagnostic or treatment advice, and the full framework for chronic orofacial pain belongs to domain article P09.

F17 | Evidence on overdiagnosis: in the South Korean pseudo-patient study (196 private clinics, 58 pseudo-patients, August to December 2018), only 33.2% (65/196) of interactions were diagnosed as free of caries, while 20.9% (41/196) were diagnosed with five or more carious teeth; in a questionnaire of 42 final-year Spanish dental students, 87.8% proposed an overtreatment in the scenario of “repairing a defective existing restoration”, and a tendency to perform unnecessary complementary tests was observed; the same abstract also carries figures pointing the other way — the caries treatment proposals accorded with available evidence in the majority of scenarios, management at different caries stages was correct in 51.2–92.7% of cases, and 58.8% correctly identified an early carious lesion

• Source #: #15 / #W7 | confidence: low (single country, single design; the subjects of #W7 are students, not practising clinicians) | basis: peer_reviewed (PMID 39633350 / 34886311) | geo: universal | period: BMC Oral Health, 2024 / Int J Environ Res Public Health, 2021
• caveat: both are quantitative evidence under a particular country and a particular study design, and neither may be inferred to any individual dentist, any institution or any other country; the oral examination in #15 used visual and tactile methods only; the sample of #W7 is a questionnaire of 42 students (52 sent out) and does not represent clinical reality in any country; the citation of #W7 must present its positive figures alongside (Fn99a / Fn99b / Fn99c), because taking only the 87.8% and the “unnecessary tests” would make this article more negative than the source itself; this article's search obtained no comparable pseudo-patient study from another country against which #15 could be set, so it makes no cross-national comparison and provides no “global average overdiagnosis rate”. The extrapolating tone of “and worldwide” in the original authors' conclusion is their own statement, which this article records faithfully without adopting it as a claim of its own.

F18 | Part of overdiagnosis originates in technical factors of image interpretation: optical effects such as the Mach band effect and triangular-shaped radiolucencies can produce false positives; a meta-analysis shows that effects of this kind led to false-positive diagnoses of caries or fractures in approximately 13% of observations (60/464), and that the prevalence of non-carious TSR on maxillary molars was 26.44% (270/1021)

• Source #: #16 (PROSPERO CRD420251083823) | confidence: low | basis: peer_reviewed (PMID 41331196) | geo: universal | period: Oral Radiol, 2026
• caveat: only 5 of 640 reports were included, heterogeneity was high (I² > 90%), and the certainty was rated low to very low (as stated by the original); the use of this entry in this article is to turn the question constructively from “is the clinician honest” to “what is the diagnosis based on”, and it must not be turned round to question the overall validity of radiographic diagnosis; the clinical recommendation of the original is to interpret with caution and always correlate with a thorough clinical examination (Fn97).

F19 | Behaviour around oral information on social platforms: in a questionnaire run by the faculty of dentistry at Damascus University (301 collected, 291 analysed), 50.2% said they sometimes feel that advice on social platforms is marketing for a company, a product or a doctor; 89.7% visit a dentist when they have a medical problem, while 10.3% follow the advice on social platforms

• Source #: #W8 | confidence: low | basis: peer_reviewed (PMID 38516284) | geo: universal | period: Cureus, 2024
• caveat: a self-selected sample recruited through social media and run by a single institution, with an educational distribution skewed high (the sample were people with secondary education and above, undergraduates and postgraduates), which must never be extrapolated into a proportion for any other country or for the world; the original states only that the institution running the study was the faculty of dentistry at Damascus University (Fn11), and does not describe the sample by country — “Syria” is general geographical knowledge inferred by the reader from the institution, not a statement of the source, and the main text of this article has been rewritten accordingly; within this article this entry is used only to show that “commercial motive” is already a question the public asks spontaneously, and it is not a claim about the prevalence of any behaviour.

Compliance note

• This article is health education and a compilation of medical news, is general oral-health information, does not solicit medical business, does not constitute medical advertising, and does not constitute diagnosis or treatment advice.
• This article provides no amount, charge or reimbursement information; every passage touching on cost states only the items it is composed of and the variables involved.
• This article does not recommend, rank or disparage any institution, clinician, brand or product, contains no identifiable individual case, and carries no third party's subjective account of a treatment. The case reports cited are all de-identified records in published literature.
• The research on “overdiagnosis” described here is at population level throughout and does not constitute an accusation against any individual clinician, institution or country; this article also provides no checklist for judging a clinician's integrity.
• Every statement at the level of ingredients, materials and practices is a compilation of the literature, not a claim about product performance; oral hygiene products may not claim medical efficacy and cannot replace diagnosis and treatment.
• This article contains no medication, dose or procedural instruction; where emergency management is cited, only the principle-level content already stated in the guidelines is presented, and actual management must be carried out by a professional.
• The account of the “verification framework” is intended to help readers understand the evidence structure of medical information, and does not encourage substituting one's own judgement for professional assessment; the main risk this article points to repeatedly is delay in seeking care.
• All the data here are research results at population level. The actual method of treatment and its effect vary from person to person and must be assessed by a dentist before any decision.
• 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 retrieval / live testing: 2026-08-06 (all sources were tested live on the same day with curl / the PubMed E-utilities, HTTP 200; every verbatim span passed a programmatic Python `span in text` comparison)

Sources already verified in the anchor file (`ida-pillars/anchors/P23-anchors.md`)

Statement in §0 of the anchor file: under the owner's decision of 2026-08-06 that the whole line is global, Taiwanese regulations / national health insurance / health-bureau content is not used as the basis for any medical or system claim in this article; the law tier for this domain is honestly left empty (0 entries), and local systems and costs are always linked down to the corresponding TW canonical cards and to domain article P12.

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

Reason for the additions: the anchor file covers seven patient questions, but the scope of this article contains four further blocks not covered by it — (1) the methodological basis required by the verification framework itself (hierarchy of evidence, control groups, interests); (2) the NICO cavitation thesis (explicitly listed in the work order as within scope, and entirely absent from the anchor file); (3) baking soda whitening (explicitly listed in the work order; the anchor file covers only activated charcoal); and (4) reinforcement for the geographical limits of overdiagnosis (§5 of the anchor file states that it has evidence from South Korea only) together with a direct case on the orthodontic side of DIY self-treatment. All 11 entries below were tested live on this machine with curl (HTTP 200) and passed a programmatic verbatim comparison; the anchor file was not modified.