Vaccines and autism: the claim, the tests and the institutions (1998 to 2026)

As of 2026-10-02. Times are UTC; every event here is known to the day, month or year only, so dots are hollow or drawn as bars. Publication dates are issue or online dates as given in the cited record. Events marked "single" or "reported" rest on one source or on reporting rather than a document read for this dig. Panel B enlarges 2025-2026, when the US institutional position changed.

Shape = kind of source: machine-recorded data a primary document an official statement a person's account a scholar's account a news report Style = status: 2+ sources agree one source sources differ our inference / time not reported the same event at another source's time Every dot is a link to its main source; the table lists all sources.

Panel A: 1998 to 2020 (one tick per two years)

Click a dot to open its source. Scroll sideways on a phone.

1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 Time, UTC The 1998 paper and the UK Studies Courts and compensation Health agencies Media and statements 1. 28 Feb 1998: Lancet publishes Wakefield et al. (12 children) (reported, document). RETRACTED in 2010. Parents linked onset to MMR in 8 of 12 [abstract].1 6. 2003-04: MMR coverage at age 2 in England bottoms out at 79.9% (reported, data). Down from 91.8% in 1995-96 [read].6 8. 6 Mar 2004: 10 of the 13 authors retract the paper's interpretation (reported, document). Wakefield, Linnell and Harvey are not among the signatories [author list read].8 12. 20 Jun 2008: HPA reports measles endemic again in the UK (reported, official). Attributed to almost a decade of low MMR coverage [read, as quoted by Eurosurveillance].12 15. 2 Feb 2010: The Lancet retracts the 1998 paper (reported, document). Announced days after the GMC's findings of fact of 28 Jan 2010; the editor called the paper's statements 'utterly false' [Guardian, read]. The notice itself (6 Feb 2010 issue) was not read.15 17. 24 May 2010: GMC strikes off Wakefield and Walker-Smith (reported, official). Murch not guilty of serious professional misconduct [judgment para 1; BMJ].17 21. 7 Mar 2012: High Court quashes Walker-Smith's striking-off (reported, official). The judge found the panel's reasoning inadequate and its determination on Walker-Smith could not stand (paras 186-187) [read].21 2. Jun 1999: Taylor et al. find no clustering of autism after MMR (498 cases) (reported, analysis). Regression within 2 months: relative incidence 0.92 [abstract].2 5. Nov 2002: Madsen et al., 537,303 Danish children, RR 0.92 (reported, analysis). Adjusted RR 0.92 (0.68 to 1.24) for autistic disorder, 0.83 (0.65 to 1.07) for other autistic-spectrum disorders [read in full].5 7. Feb 2004: DeStefano et al. (CDC), age at MMR and autism (reported, analysis). Whole-sample OR 1.49 for vaccination before 36 months, attributed to early-intervention requirements [abstract].7 13. Sep 2008: Hornig et al. find no persistent measles RNA in the gut (4% vs 8%) (reported, analysis). Three blinded labs, including the original one [read].13 24. 3 Oct 2014: Hooker's reanalysis of the CDC data retracted (reported, document). Undeclared competing interests; concerns about methods and statistics [read].24 25. Apr 2015: Jain et al.: no increase in children with an affected older sibling (reported, analysis). 95,727 children [abstract].25 26. 24 Apr 2017: Mawson et al. survey reports OR 4.2 for autism (reported, analysis). Self-selected online survey; 3 unvaccinated cases [read].26 27. Mar 2019: Hviid et al., 657,461 Danish children, HR 0.93 (reported, analysis). 6,517 autism diagnoses, HR 0.93 (0.85 to 1.02); no increase in subgroups, no clustering after vaccination [read in full].27 11. Nov 2007: US government concedes the Poling compensation claim (reported, official). CNN: the program 'concluded' the vaccinations 'significantly aggravated' an underlying illness; NPR: 'possible' they 'may have exacerbated' a mitochondrial disorder. Both read; the decision itself was not.11 14. 12 Feb 2009: special masters deny the first three autism test cases (Cedillo, Hazlehurst, Snyder) (reported, official). Denials confirmed in the Federal Circuit opinions 10-5004 and 09-5128 (footnote 1 of the first), which also say the Court of Federal Claims affirmed Hazlehurst and Snyder and that the Petitioners' Steering Committee proposed the test-case approach [opened]; the special masters' decisions themselves were not opened.14 16. 12 Mar 2010: special masters' decisions in three further test cases (King, Mead, Dwyer) (reported, official). Listed in the court's decision list with this date; the decisions were not opened and their outcome is not stated here.16 18. 27 Aug 2010: Federal Circuit affirms Cedillo (reported, official). Hazlehurst had been affirmed on 13 May 2010; Snyder was not appealed (Federal Circuit 10-5004, footnote 1) [opened].18 20. 22 Feb 2011: Bruesewitz v. Wyeth: design-defect suits preempted; no autism finding (reported, official). Autism appears only in the dissent's footnote 25 [read].20 22. 22 Aug 2012: Maryland revokes Mark Geier's licence (reported, official). The Board found he gave Lupron to autistic children outside its approved use without adequate examination, and used consent forms it called misleading and in one case blatantly false (223 Md. App. 404, quoting the Board's order) [read].22 3. 9 Jul 1999: AAP and US Public Health Service ask for thimerosal to be removed as soon as possible (reported, official). The statement says there were 'no data or evidence of any harm' at the exposure in use, and that some children could exceed a federal methylmercury guideline [read directly].3 4. 7-8 Jun 2000: Simpsonwood meeting reviews preliminary thimerosal data (reported, official). Autism (transcript p. 44): 'a slight, but not significant, increase' at the highest exposure [read].4 9. 2004: IOM: the evidence favors rejecting a causal link (MMR and thimerosal) (reported, official). Leaves a very small susceptible subset unexcluded, with no evidence for it [read].9 10. Jun 2005: 'Deadly Immunity' published by Salon and Rolling Stone (reported, media). Corrected five times; Salon removed it in Jan 2011 [read].10 19. 5 Jan 2011: BMJ publishes Deer's records comparison; its editors call the paper fraudulent (reported, media). Deer: 3 of 9 'regressive autism' children had no autism diagnosis in the records [read via fetch tool]. The fraud judgment is the editorial's (Godlee et al.).19 23. 27 Aug 2014: CDC co-author William Thompson says a subgroup result was omitted from DeStefano 2004 (single, witness). Read in a third-party copy; the CDC's response addresses the same point [read].23

Panel B: June 2025 to October 2026 (one tick per two months)

Click a dot to open its source. Scroll sideways on a phone.

01 Jun 00:00 01 Aug 00:00 01 Oct 00:00 01 Dec 00:00 31 Jan 00:00 02 Apr 00:00 02 Jun 00:00 02 Aug 00:00 02 Oct 00:00 Time, UTC Studies Courts and compensation Health agencies Media and statements 28. Jul 2025: Danish aluminum cohort (1,224,176 children): no increase (reported, analysis). Autism HR 0.93 per mg [press release].28 30. 11 Aug 2025: the journal's editor says she sees no reason for retraction (reported, media). Reported by Reuters (11 Aug) and Live Science (13 Aug).30 36. 18 May 2026: expression of concern on Hooker & Miller 2020 (reported, document). The article is under investigation [read].36 35. 16 Mar 2026: federal court stays the new schedule and the new ACIP's votes (AAP v. Kennedy) (reported, official). On appeal to the First Circuit; the order was not read.35 29. 1 Aug 2025: HHS Secretary calls for the aluminum paper's retraction (reported, official). Reported by Reuters and Live Science.29 31. 9 Sep 2025: unpublished Henry Ford draft entered into a Senate hearing record (reported, official). Its autism result: adjusted HR 0.62 (0.10 to 3.69), 23 vs 1 cases [read].31 33. 19-20 Nov 2025: CDC page changed: 'vaccines do not cause autism' called not evidence-based (reported, official). Kennedy said he ordered the change (STAT, 21 Nov 2025).33 34. 11 Dec 2025: WHO safety committee: no evidence of a causal relationship (reported, official). 20 of 31 primary studies and all 5 meta-analyses null; 11 studies (9 from one US group) suggested an association, rated very low strength and high risk of bias [page read]. Adoption, not evidence.34 37. 22 Jul 2026: CDC page revised again (reported, official). Adds that scientists have not identified the root causes of autism [read via fetch tool].37 38. 10 Aug 2026: Executive Order 14420 orders a childhood-schedule reassessment (reported, official). Does not mention autism [read via fetch tool].38 32. 26 Sep 2025: Henry Ford Health says the draft was shelved for data and methods problems (reported, document). [read via fetch tool]32 39. 29 Sep 2026: survey: 68% of US adults say MMR does not cause autism, 13% say it does (reported, data). Fielded 4-17 Aug 2026; statistically unchanged since 2022 [read via fetch tool].39

Events (numbers match the diagram)

#When (UTC)WhatStatusSources
128 Feb 1998 (date only)28 Feb 1998: Lancet publishes Wakefield et al. (12 children)
RETRACTED in 2010. Parents linked onset to MMR in 8 of 12 [abstract].		reported documentsrc-wakefield-1998
2Jun 1999 (month only)Jun 1999: Taylor et al. find no clustering of autism after MMR (498 cases)
Regression within 2 months: relative incidence 0.92 [abstract].		reported analysissrc-taylor-1999
309 Jul 1999 (date only)9 Jul 1999: AAP and US Public Health Service ask for thimerosal to be removed as soon as possible
The statement says there were 'no data or evidence of any harm' at the exposure in use, and that some children could exceed a federal methylmercury guideline [read directly].		reported officialsrc-mmwr-1999
407 Jun 2000 00:00 to 08 Jun 2000 23:597-8 Jun 2000: Simpsonwood meeting reviews preliminary thimerosal data
Autism (transcript p. 44): 'a slight, but not significant, increase' at the highest exposure [read].		reported officialsrc-simpsonwood-2000
5Nov 2002 (month only)Nov 2002: Madsen et al., 537,303 Danish children, RR 0.92
Adjusted RR 0.92 (0.68 to 1.24) for autistic disorder, 0.83 (0.65 to 1.07) for other autistic-spectrum disorders [read in full].		reported analysissrc-madsen-2002
601 Apr 2003 00:00 to 31 Mar 2004 23:592003-04: MMR coverage at age 2 in England bottoms out at 79.9%
Down from 91.8% in 1995-96 [read].		reported datasrc-nhs-immunisation-2004-05
7Feb 2004 (month only)Feb 2004: DeStefano et al. (CDC), age at MMR and autism
Whole-sample OR 1.49 for vaccination before 36 months, attributed to early-intervention requirements [abstract].		reported analysissrc-destefano-2004
806 Mar 2004 (date only)6 Mar 2004: 10 of the 13 authors retract the paper's interpretation
Wakefield, Linnell and Harvey are not among the signatories [author list read].		reported documentsrc-murch-2004
92004 (year only)2004: IOM: the evidence favors rejecting a causal link (MMR and thimerosal)
Leaves a very small susceptible subset unexcluded, with no evidence for it [read].		reported officialsrc-iom-2004
10Jun 2005 (month only)Jun 2005: 'Deadly Immunity' published by Salon and Rolling Stone
Corrected five times; Salon removed it in Jan 2011 [read].		reported mediasrc-salon-2011
11Nov 2007 (month only)Nov 2007: US government concedes the Poling compensation claim
CNN: the program 'concluded' the vaccinations 'significantly aggravated' an underlying illness; NPR: 'possible' they 'may have exacerbated' a mitochondrial disorder. Both read; the decision itself was not.		reported officialsrc-poling-cnn-2008
1220 Jun 2008 (date only)20 Jun 2008: HPA reports measles endemic again in the UK
Attributed to almost a decade of low MMR coverage [read, as quoted by Eurosurveillance].		reported officialsrc-eurosurv-2008
13Sep 2008 (month only)Sep 2008: Hornig et al. find no persistent measles RNA in the gut (4% vs 8%)
Three blinded labs, including the original one [read].		reported analysissrc-hornig-2008
1412 Feb 2009 (date only)12 Feb 2009: special masters deny the first three autism test cases (Cedillo, Hazlehurst, Snyder)
Denials confirmed in the Federal Circuit opinions 10-5004 and 09-5128 (footnote 1 of the first), which also say the Court of Federal Claims affirmed Hazlehurst and Snyder and that the Petitioners' Steering Committee proposed the test-case approach [opened]; the special masters' decisions themselves were not opened.		reported officialsrc-oap-decisions, src-cafc-cedillo-2010, src-cafc-hazlehurst-2010
1502 Feb 2010 (date only)2 Feb 2010: The Lancet retracts the 1998 paper
Announced days after the GMC's findings of fact of 28 Jan 2010; the editor called the paper's statements 'utterly false' [Guardian, read]. The notice itself (6 Feb 2010 issue) was not read.		reported documentsrc-guardian-lancet-2010, src-lancet-retraction-2010
1612 Mar 2010 (date only)12 Mar 2010: special masters' decisions in three further test cases (King, Mead, Dwyer)
Listed in the court's decision list with this date; the decisions were not opened and their outcome is not stated here.		reported officialsrc-oap-decisions
1724 May 2010 (date only)24 May 2010: GMC strikes off Wakefield and Walker-Smith
Murch not guilty of serious professional misconduct [judgment para 1; BMJ].		reported officialsrc-bmj-gmc-2010, src-walker-smith-2012
1827 Aug 2010 (date only)27 Aug 2010: Federal Circuit affirms Cedillo
Hazlehurst had been affirmed on 13 May 2010; Snyder was not appealed (Federal Circuit 10-5004, footnote 1) [opened].		reported officialsrc-oap-decisions, src-cafc-cedillo-2010
1905 Jan 2011 (date only)5 Jan 2011: BMJ publishes Deer's records comparison; its editors call the paper fraudulent
Deer: 3 of 9 'regressive autism' children had no autism diagnosis in the records [read via fetch tool]. The fraud judgment is the editorial's (Godlee et al.).		reported mediasrc-deer-2011, src-godlee-2011
2022 Feb 2011 (date only)22 Feb 2011: Bruesewitz v. Wyeth: design-defect suits preempted; no autism finding
Autism appears only in the dissent's footnote 25 [read].		reported officialsrc-bruesewitz
2107 Mar 2012 (date only)7 Mar 2012: High Court quashes Walker-Smith's striking-off
The judge found the panel's reasoning inadequate and its determination on Walker-Smith could not stand (paras 186-187) [read].		reported officialsrc-walker-smith-2012
2222 Aug 2012 (date only)22 Aug 2012: Maryland revokes Mark Geier's licence
The Board found he gave Lupron to autistic children outside its approved use without adequate examination, and used consent forms it called misleading and in one case blatantly false (223 Md. App. 404, quoting the Board's order) [read].		reported officialsrc-geier-md-2015-licence, src-geier-md-2015, src-geier-md-2019
2327 Aug 2014 (date only)27 Aug 2014: CDC co-author William Thompson says a subgroup result was omitted from DeStefano 2004
Read in a third-party copy; the CDC's response addresses the same point [read].		single witnesssrc-thompson-2014, src-cdc-2004-study-statement
2403 Oct 2014 (date only)3 Oct 2014: Hooker's reanalysis of the CDC data retracted
Undeclared competing interests; concerns about methods and statistics [read].		reported documentsrc-hooker-2014-retraction
25Apr 2015 (month only)Apr 2015: Jain et al.: no increase in children with an affected older sibling
95,727 children [abstract].		reported analysissrc-jain-2015
2624 Apr 2017 (date only)24 Apr 2017: Mawson et al. survey reports OR 4.2 for autism
Self-selected online survey; 3 unvaccinated cases [read].		reported analysissrc-mawson-2017
27Mar 2019 (month only)Mar 2019: Hviid et al., 657,461 Danish children, HR 0.93
6,517 autism diagnoses, HR 0.93 (0.85 to 1.02); no increase in subgroups, no clustering after vaccination [read in full].		reported analysissrc-hviid-2019
28Jul 2025 (month only)Jul 2025: Danish aluminum cohort (1,224,176 children): no increase
Autism HR 0.93 per mg [press release].		reported analysissrc-andersson-2025
2901 Aug 2025 (date only)1 Aug 2025: HHS Secretary calls for the aluminum paper's retraction
Reported by Reuters and Live Science.		reported officialsrc-livescience-2025
3011 Aug 2025 (date only)11 Aug 2025: the journal's editor says she sees no reason for retraction
Reported by Reuters (11 Aug) and Live Science (13 Aug).		reported mediasrc-livescience-2025
3109 Sep 2025 (date only)9 Sep 2025: unpublished Henry Ford draft entered into a Senate hearing record
Its autism result: adjusted HR 0.62 (0.10 to 3.69), 23 vs 1 cases [read].		reported officialsrc-henry-ford-draft
3226 Sep 2025 (date only)26 Sep 2025: Henry Ford Health says the draft was shelved for data and methods problems
[read via fetch tool]		reported documentsrc-henry-ford-statement
3320 Nov 2025 (date only)19-20 Nov 2025: CDC page changed: 'vaccines do not cause autism' called not evidence-based
Kennedy said he ordered the change (STAT, 21 Nov 2025).		reported officialsrc-stat-cdc-page-2025
3411 Dec 2025 (date only)11 Dec 2025: WHO safety committee: no evidence of a causal relationship
20 of 31 primary studies and all 5 meta-analyses null; 11 studies (9 from one US group) suggested an association, rated very low strength and high risk of bias [page read]. Adoption, not evidence.		reported officialsrc-who-gacvs-2025
3516 Mar 2026 (date only)16 Mar 2026: federal court stays the new schedule and the new ACIP's votes (AAP v. Kennedy)
On appeal to the First Circuit; the order was not read.		reported officialsrc-acp-injunction-2026
3618 May 2026 (date only)18 May 2026: expression of concern on Hooker & Miller 2020
The article is under investigation [read].		reported documentsrc-hooker-miller-2020-eoc
3722 Jul 2026 (date only)22 Jul 2026: CDC page revised again
Adds that scientists have not identified the root causes of autism [read via fetch tool].		reported officialsrc-cdc-autism-page
3810 Aug 2026 (date only)10 Aug 2026: Executive Order 14420 orders a childhood-schedule reassessment
Does not mention autism [read via fetch tool].		reported officialsrc-eo-14420
3929 Sep 2026 (date only)29 Sep 2026: survey: 68% of US adults say MMR does not cause autism, 13% say it does
Fielded 4-17 Aug 2026; statistically unchanged since 2022 [read via fetch tool].		reported datasrc-annenberg-2026

People and bodies involved

Sources

Generated from timeline.yaml by build/tools/render_timeline.py. This page records what is reported and where reports differ. It is not a finding about what happened.