Antibiotics treat bacterial infections; they don't work on viruses.

1. Pass Title matches the evidence

   Title 'Antibiotics treat bacterial infections; they don't work on viruses' matches finding False and the cited guidance and trial review.

2. Pass No obvious contradictions

   Finding False and conclusion agree; the bacterial-coinfection qualification is a caveat, not a contradiction.

3. Pass Plausible justification

   Conclusion explains that mechanism guidance, trial review, pathology and guideline evidence converge on the rating.

4. Pass Sources are relevant

   CDC guidance, a common-cold trial review, influenza pathology and an IDSA influenza guideline are on topic.

5. Pass Common questions answered

   Three objections (flu improvement, bacterial pneumonia, precautionary use) each receive a direct response.

Coverage: Read the full article payload; sources judged by name/title/description/role only, per First Look's rules; no source snapshots read.

1. Concern (minor) Context is complete

   Note in e2 that the Cochrane review found some benefit for acute purulent rhinitis, and explain how that fits the bacterial-coinfection qualification.

   The article includes the key qualification (bacterial coinfection can warrant antibiotics) and the harm side (adverse effects, stewardship). One omission: e2 says the review covered 'the common cold and acute purulent rhinitis' but reports only the common-cold result. The article does not address the review's acute purulent rhinitis finding, which (to my knowledge) suggested some benefit alongside adverse effects and advised against routine use. That is relevant because CDC says colds do not need antibiotics 'even if the mucus is thick, yellow or green', and a reader may wonder about the apparent tension. Mentioning it would actually reinforce the coinfection qualification. This rests on my own knowledge because the PubMed page was unreadable, so it is phrased as something the article should address. Impression unchanged, so Minor. Confidence: Medium.

   About evidence:common-cold-trial-review, source:kenealy-arroll2013 The finding

2. Pass Central claims are accurate

   One-sentence summary: antibiotics do not act on viruses or benefit uncomplicated viral illness, but they can be appropriate for a coexisting bacterial infection. Claim ratings: (1) Finding/answer 'Antibiotics don't act on viruses; they help only when a bacterial infection is also present' - True. CDC source: 'Antibiotics DO NOT work on viruses' and 'Antibiotics will NOT make you feel better if you have a virus'; this is also a well-established, uncontested fact. (2) e1 CDC summary ('do not work on viruses such as those that cause colds, flu, or most sore throats') - True, matches CDC text nearly verbatim. (3) e2 Cochrane review found no benefit for routine treatment of colds and increased adverse effects in adults - Unverifiable from the source (PubMed returned a cookie wall); consistent with my knowledge of the review. (4) e3 Morens et al. found secondary bacterial pneumonia predominant in many pandemic deaths - Unverifiable from source (403), but the cited title itself states 'Predominant Role of Bacterial Pneumonia as a Cause of Death in Pandemic Influenza'. (5) e4 IDSA guideline separately addresses suspected bacterial coinfection - Mostly True/Unverifiable: the readable abstract covers testing, antiviral treatment and outbreak management, not coinfection; the full guideline was not available in the snapshot. Central claims are verified by a primary public-health source and are uncontested. Confidence: High for central claims; Medium for e2-e4 details.

3. Pass Claims are supported and labeled

   Each major claim maps to a cited evidence item: mechanism to e1 (CDC), lack of benefit in colds to e2 (Cochrane review, DOI given), bacterial complications to e3 (Morens 2008), clinical separation of antiviral and coinfection care to e4 (IDSA). Inference is labeled as such: e3 'supports treating a bacterial complication, not the virus itself'; the conclusion separates population evidence from individual decisions ('individual treatment decisions require clinical assessment'). No speculation is presented as fact. Confidence: High.

4. Pass Sourcing is strong and transparent

   Sources are primary or high-quality secondary: CDC official guidance, a Cochrane systematic review of RCTs (DOI 10.1002/14651858.CD000247.pub3, PMID 23733381), a peer-reviewed pathology review (PMID 18808358), and the IDSA clinical practice guideline (CDC Stacks 80092; readable record confirms Uyeki et al., Clin Infect Dis 68(6):e1-e47). Each evidence item lists its sourceIds, so it is clear which source supports which claim. No partisan, anonymous, or circular sourcing. Two links could not be fetched server-side (PubMed cookie wall, Europe PMC 403), which limits automated verification but does not reduce source quality; identifiers allow retrieval. Confidence: High.

5. Pass Framing is fair and neutral

   The claim 'Antibiotics directly treat viral infections' is represented plainly, and objections ('Some people with flu improve after taking antibiotics', 'It is safer to take antibiotics just in case') are answered with evidence and without ridicule. Emphasis is balanced: the article stresses the legitimate coinfection exception rather than overstating the myth ('identifying that situation requires clinical judgment rather than assuming every viral illness does or does not need treatment'). Confidence: High.

6. Pass Uncertainty is stated

   High confidence is appropriate for a mechanism fact that is uncontested. Limitations are specific and stated per item (e.g., e2 'Findings about uncomplicated common colds do not rule out a separately diagnosed bacterial infection'; e3 'does not show that antibiotics benefit uncomplicated influenza'; e4 'Recommendations may change'). The conclusion separates the population-level claim from individual diagnosis. No contested evidence is presented as settled. Confidence: High.

Coverage: Read all four cited sources via get_source_snapshot. CDC 'Antibiotic Do's and Don'ts' (live, full text) and the CDC Stacks record for the IDSA 2018 influenza guideline (live; landing page with citation and abstract only, not the full guideline text) were readable. Kenealy & Arroll 2013 (PubMed) returned only a cookie wall, and Morens et al. 2008 (Europe PMC) returned HTTP 403 with no Wayback capture, so the specific statements in evidence e2 and e3 and the coinfection content of e4 could not be checked against source text; they were assessed against general knowledge only and rated Unverifiable from sources. Images and changelog were not evaluated.

1. Concern (minor) Confidence matches the evidence

   Reword the summary and context: only e1 and e2 bear on direct antiviral action; e3 and e4 support the coinfection qualification, not that point.

   High confidence is warranted for the mechanism claim. But the summary says all four lines 'converge: antibiotics target bacteria, not viruses', and context says the conclusion 'is strong because evidence from different methods, institutions, countries, and study periods converges'. As stated, e3 (fatal pandemic pathology) and e4 (influenza guideline) do not test whether antibiotics act on viruses; e3's own text says it 'supports treating a bacterial complication, not the virus itself'. The direct-action point therefore rests on e1 and e2 only. The verdict is unaffected, but the convergence language overstates how many lines speak to the central point. I would keep High confidence and fix the wording. Minor.

   About conclusion, evidence:influenza-pathology-coinfection, evidence:influenza-clinical-guideline The finding

2. Concern (material) Strongest objections answered

   Add an objection addressing claims that some antibiotics (e.g., azithromycin, doxycycline) have antiviral effects, as promoted for COVID-19, and the trial evidence on them.

   The three objections answered ('people improve after taking antibiotics', 'bacterial pneumonia after influenza', 'safer to take antibiotics just in case') are reasonable but are not the strongest form of the claim. The article does not address whether certain antibiotics, such as azithromycin or doxycycline, have antiviral or immunomodulatory activity in laboratory studies, a version of the claim that circulated widely during COVID-19, when azithromycin was promoted as a treatment. That is the case where a reader is most likely to encounter 'antibiotics treat viral infections' with apparent scientific backing, and the article's evidence (CDC mechanism guidance; common-cold trials from 2013 and earlier) does not speak to it. Answering it with randomized trial evidence on azithromycin in viral illness would close the gap. This rests on my own knowledge, so it is raised as something the article should address. It would change how strongly a reader who has heard that version believes the verdict, so Material.

   About conclusion The finding

3. Pass Rates the real claim

   Claim: 'Antibiotics directly treat viral infections.' As it circulates, the belief is that antibiotics will help a cold, flu, or other viral illness. The word 'directly' makes the claim slightly narrower, but the article does not rely on that narrowing: its answer and conclusion also address benefit ('do not benefit uncomplicated viral illness'), supported by the common-cold trial review (e2). The conclusion stays within scope and explicitly carves out bacterial coinfection. I would rate the same claim. Confidence: High.

4. Pass Rating follows from the evidence

   Before reading the verdict, from e1 ('antibiotics ... do not work on viruses') and e2 ('no benefit supporting routine antibiotic treatment of the common cold') I would choose False. e3 and e4 show that bacterial complications are a separate target, which supports the qualification rather than the claim. The article's 'False' and answer 'Antibiotics don't act on viruses; they help only when a bacterial infection is also present' match. Misleading would not fit: the claim's core is wrong, not a true fact framed badly. Confidence: High.

5. Pass Evidence is independent

   As stated, the items use different methods (stewardship guidance, systematic review of RCTs, historical pathology, clinical guideline) from different institutions (CDC, Cochrane ARI Group, NIAID, IDSA) and years (2008-2026), and cite distinct sources. e1 and e4 are both US public-health/clinical guidance and may draw on overlapping literature, but nothing stated shows they rest on the same dataset or study, and e2 is a separate trial synthesis. The thinness of independent support for the central point is a calibration issue, reported under confidence-calibrated, not a dependence issue. Confidence: Medium-High.

Coverage: Examined the full article payload: claim, finding, answer, confidence, title, summary, context, conclusion, all four evidence items with limitations, three objections, and five method steps. Per the Skeptic's instructions, no source snapshots were read and evidence was taken as stated.

1. Concern (minor) Each section does its job

   Replace context with one sentence of framing (why the belief is common) and drop the hand-written convergence note; make the summary one sentence.

   Rule: context is essential framing only, and the author must not write an independence note (the page generates it); summary is one sentence. Context reads: 'Fact check - the direct effect of antibiotics on viral infections. No single item is complete proof by itself. The conclusion is strong because evidence from different methods, institutions, countries, and study periods converges.' The first sentence is a label, and the rest is a hand-written independence note duplicating what the page generates, rather than framing. The summary is two sentences. The conclusion synthesizes and states the qualification and what remains individual (passes); editorialNote states scope, caveats, and review status (passes); keyPoints state what the evidence establishes (passes). Minor: content in the wrong field.

   About conclusion The finding

2. Concern (minor) Readers can verify it

   Rewrite method step 3 to name the sources reviewed instead of 'the supplied stable links', and add the IDSA guideline's DOI to its source.

   Rule: method steps include the sources reviewed; key claims have primary anchors such as a DOI. Sources have meaningful role labels ('Systematic review of treatment trials', 'Historical pathology evidence') and descriptions; Kenealy and Morens carry PMIDs and a DOI. Method steps 1, 2, 4, 5 cover claim definition, independent evidence, and limits. Method step 3, 'Used the supplied stable links and recorded DOI ... for the common-cold review', does not say which sources were reviewed and refers to an unseen supplier of links, reading like process residue. The IDSA guideline is anchored only by a repository id ('CDC Stacks 80092') rather than its journal DOI. Minor: one vague method element and one weaker anchor; the key claims remain retrievable.

   About method:3, source:idsa-influenza-guideline The finding

3. Pass Title states the finding

   Title: 'Antibiotics treat bacterial infections; they don't work on viruses.' It leads with a positive statement, names the subject, and resolves the claim for a title-only reader. The negative second clause states an absence (no antiviral action), for which no accurate positive wording exists, so the 'lead with what is true' rule is met. No clickbait or loaded words. seoTitle 'Do Antibiotics Treat Viral Infections? No-With an Important Exception' uses search wording and states the finding ('No'), as the rule allows.

4. Pass Verdict fields agree and stand alone

   Read as: FactHub rates the claim "Antibiotics directly treat viral infections." False (High confidence): Antibiotics don't act on viruses; they help only when a bacterial infection is also present. The claim is one declarative sentence stated as the claim, not its correction; the answer stands alone and says what is true. Title, summary ('antibiotics target bacteria, not viruses'), and conclusion ('Antibiotics do not directly act on viruses ... Confidence: High') agree with the False label. No stale wording from the earlier free-text finding.

5. Pass Limitations are specific

   Each limitation is specific to its item: e1 'Guidance summarizes established evidence rather than reporting a new controlled experiment'; e2 'trials varied in antibiotics, populations, outcome definitions, and age' and do 'not rule out a separately diagnosed bacterial infection'; e3 'concerns severe and fatal historical pandemic cases ... does not show that antibiotics benefit uncomplicated influenza'; e4 'cannot replace examination, testing, or judgment'. None is boilerplate.

6. Pass Tone is measured and respectful

   Language is measured ('converge', 'Confidence: High for the direct-treatment claim'). Objection responses are short and evidence-based, e.g. 'Improvement can coincide with the natural course of a viral illness, or treatment may have addressed a bacterial coinfection', with no ridicule or sarcasm. No unjustified absolutes; 'proves' appears only in restating an objection and is answered with 'It proves the importance of recognizing a bacterial complication', which is warranted.

7. Pass Consistent house style

   American spelling throughout; years and dates consistent; names (CDC, IDSA, Kenealy and Arroll, Morens) used consistently, and the full institution names appear in each evidence item's institutions and in the source list. Technical terms ('systematic review', 'coinfection', 'stewardship') are clear from context. No typos, placeholders, or notes to editors found in reader-visible text.

Coverage: Examined only the article payload: title, seoTitle, description, claim, finding, answer, confidence, summary, keyPoints, context, conclusion, four evidence items and limitations, three objections, five method steps, editorialNote, four sources, image caption and alt text. No sources or web were used.