“The 2021 military-database spikes came from underreported 2016–2020 records, which created an artificially low baseline.”

• The claim restates the official Defense Health Agency (DHA) explanation for the 'DMED' controversy of January–February 2022. After attorney Thomas Renz and three military medical whistleblowers testified that the Defense Medical Epidemiology Database (DMED) showed enormous 2021 increases in myocarditis, cancers, miscarriages, infertility, neurological disease, etc., versus a 2016–2020 five-year average, DHA spokesman Peter Graves stated that a review found the 2016–2020 data in DMED 'represented only a fraction of actual ailment cases' and that only the 2021 data was accurate; once corrected, total 2016–2020 diagnoses were comparable to or higher than 2021.
• DoD attributed the discrepancy to a data-transfer/corruption problem between the source system (Defense Medical Surveillance System, DMSS) and the query tool (DMED), i.e., records existed but were not fully loaded for the earlier years — exactly the mechanism of an 'artificially low baseline.'
• DMED was taken offline in early 2022 specifically to correct and validate the data before being restored, an action consistent with the department conceding the earlier-year counts were defective rather than the 2021 counts being anomalous.
• Internal-consistency red flags support an artifact rather than a real biological signal: the reported 2021 'spikes' hit codes with no plausible vaccine mechanism (e.g., HIV diagnoses, chest pain, labor and delivery complications, 'male infertility'), and were of implausible magnitude (several hundred to >1,000 percent). Across-the-board, non-specific inflation is the classic fingerprint of missing records in the comparison years.
• The spikes were not corroborated by other DoD surveillance streams. The Armed Forces Health Surveillance Division's Medical Surveillance Monthly Report (MSMR) and DoD's dedicated post-vaccination myocarditis surveillance reported increases measured in dozens-to-low-hundreds of excess cases in young males, orders of magnitude below the DMED-derived figures Renz presented.
• Mainstream fact-checking organizations (Reuters, AP, PolitiFact, FactCheck.org) independently reviewed the episode in Feb 2022 and converged on the same conclusion: the apparent 2021 surge was a database artifact in the pre-2021 baseline, not a measured surge in illness.

• The correction rests substantially on DoD's own assertion. The department did not publish a full forensic audit, line-level before/after dataset, or independent third-party verification of what was missing and why, so the 'underreporting' explanation is authoritative but not externally replicated.
• Sen. Ron Johnson, Renz, and the whistleblower physicians continued to dispute the explanation, noting that the pre-correction queries had been run by credentialed users with database access and arguing DoD changed the data after the fact; DoD's replies to Johnson's oversight letters were criticized as incomplete.
• The claim's blanket framing ('the spikes came from underreported records') over-generalizes slightly: it is not established that every elevated 2021 figure was pure artifact. There is genuine, independently documented evidence of a real — but much smaller — excess of myocarditis/pericarditis in young males after mRNA vaccination, including in military populations (e.g., the 2021 JAMA Cardiology report by Montgomery et al. on 23 military myocarditis cases).
• 2021 also plausibly contains some real signal from non-vaccine causes (COVID-19 infection itself, deferred-care catch-up after 2020 pandemic disruptions), which the simple 'bad baseline' story does not address.
• The statement as written is vague: it does not name the database (DMED), the years' mechanism, or the source of the correction, so it relies on the reader already knowing the context; 'underreported' is a loose description of what DoD characterized as a data-transfer/corruption failure rather than deliberate under-reporting.

76/100 — the lab’s own confidence in the score above, not a second reading of the claim.

Reuters Fact Checkfactcheck · 88 · JAMA Cardiology (Montgomery et al., myocarditis following COVID-19 mRNA vaccination in members of the US military)journal · 93

• Military Times reported that the Pentagon attributed the dramatic 2021 DMED comparisons to a database problem in which 2016-2020 records were undercounted, making the pre-2021 baseline artificially low rather than showing a real 2021 surge.
• FactCheck.org reported the same core explanation: Defense officials said the DMED comparison used by vaccine-injury claimants was invalid because the 2016-2020 figures were incomplete/incorrect, while 2021 figures had not been similarly affected.
• Reuters reported there was no evidence that COVID-19 vaccines caused the alleged military-wide surges and that the Defense Department said the apparent spikes stemmed from a DMED data problem affecting earlier years.
• The underlying logic is consistent: if the 2016-2020 denominator/baseline is artificially depressed, ordinary or less-extraordinary 2021 counts can appear as huge percentage increases even without an actual 2021 explosion in diagnoses.

• The strongest evidence for the claim is mostly the Defense Department’s explanation as reported by journalists and fact-checkers; a fully public, diagnosis-by-diagnosis audit trail showing every corrected 2016-2020 count is not readily available.
• The statement is broad: it implies all reported 2021 military-database spikes were due to underreported 2016-2020 records. Public reporting supports that explanation for the viral DMED spike claims, but it does not independently prove that every possible 2021 increase in every condition was entirely artificial.
• Administrative medical databases such as DMED/DMSS are useful for surveillance but can be affected by coding, query, completeness, and data-processing issues; that supports skepticism toward both the original spike claims and any overly precise reconstruction without the corrected raw data.

80/100 — the lab’s own confidence in the score above, not a second reading of the claim.

Reutersfactcheck · 92

• The Department of Defense's Armed Forces Health Surveillance Division (AFHSD) confirmed that a data corruption glitch caused only a small fraction of actual medical records from 2016 to 2020 to correctly port over into the Defense Medical Epidemiology Database (DMED).
• Because the five-year baseline data (2016-2020) was artificially low due to this glitch, querying the correctly populated 2021 data gave the false mathematical appearance of a massive spike (often 300% to 1000% higher) in medical conditions such as myocarditis, cancer, and miscarriages.
• Once the military temporarily took DMED offline to restore the complete 2016-2020 medical figures from the master Defense Medical Surveillance System (DMSS), the 2021 incident numbers were shown to be consistent with historical norms.

• Vaccine skeptics who originally publicized the raw DMED data spikes rejected the Department of Defense's explanation of a 'database glitch' as an intentional cover-up, though they have not provided empirical evidence to demonstrate that the DoD retroactively fabricated the corrected baseline data.

95/100 — the lab’s own confidence in the score above, not a second reading of the claim.

• The claim closely matches the Defense Health Agency’s explanation of the 2021 Defense Medical Epidemiology Database (DMED) spike. DHA said a review found that DMED data for 2016–2020 were incorrect and represented only a small fraction of the actual diagnoses when compared with the Defense Medical Surveillance System (DMSS). Because 2021 data were up to date, comparing them with the underreported earlier years created the appearance of large increases.[https://www.reuters.com/article/fact-check/dod-says-data-error-caused-spike-in-numbers-of-medical-diagnoses-in-their-medica-idUSL1N2UY1S2/][https://www.politifact.com/factchecks/2022/jan/31/instagram-posts/numbers-were-based-faulty-data-military-spokespers/]
• PolitiFact independently reported the same explanation after obtaining it from a Defense Health Agency spokesperson: the 2016–2020 data were faulty, while the 2021 figures were up to date.[https://www.politifact.com/factchecks/2022/jan/31/instagram-posts/numbers-were-based-faulty-data-military-spokespers/]
• FactCheck.org and AFP likewise described the apparent increases as a database error involving underreported 2016–2020 records, rather than evidence of a genuine 2021 surge.[https://www.factcheck.org/2022/12/scicheck-died-suddenly-pushes-bogus-depopulation-theory/][https://factcheck.afp.com/doc.afp.com.333Y8JV]

• The available evidence does not independently establish every detail implied by the wording. The explanation is primarily a statement from the Defense Health Agency, reported by fact-checkers and Reuters; the search results do not provide a publicly released, detailed audit showing the underlying year-by-year corrections for every diagnosis category.
• The original apparent spikes were based on DMED data presented by critics of COVID-19 vaccination, and later reporting indicates that the database figures were revised or found unreliable. Therefore, the claim should not be interpreted as proof that all 2021 military-health statistics were accurate or that no real increases occurred in any category.[https://www.politifact.com/factchecks/2022/jan/31/instagram-posts/numbers-were-based-faulty-data-military-spokespers/][https://fullfact.org/health/us-military-data-covid-vaccine/]
• The wording says the spikes 'came from' underreported records, which is well supported for the widely discussed DMED comparison, but it does not specify which database, diagnoses, or release it means. That ambiguity limits how broadly the claim can be applied.

91/100 — the lab’s own confidence in the score above, not a second reading of the claim.

Reutersnews · 96 · PolitiFactfactcheck · 94 · FactCheck.orgfactcheck · 94 · AFP Fact Checkfactcheck · 92 · Full Factfactcheck · 91

• DoD/Defense Health Agency stated that a 2021 DMED system/data-load error left 2016–2020 records incomplete, so complete 2021 counts produced the appearance of huge year-over-year spikes against an artificially low baseline.
• After historical records were reloaded, most of the viral 300–1000%+ jumps (cancer, neurological, HIV, etc.) shrank or vanished, consistent with a baseline artifact rather than a true 2021 epidemic in the force.
• Independent civilian surveillance (VSD, national cancer registries, etc.) did not show matching 2021 explosions in those same conditions among young adults, supporting that the military charts were distorted by bad denominators/prior-year counts.

• Whistleblower physicians released contemporaneous DMED screenshots/printouts taken before the announced correction that already showed the low 2016–2020 counts; critics argue later ‘fixes’ altered history rather than merely completing it.
• DoD never released the full raw before/after extracts or a detailed technical post-mortem of the alleged load error for outside audit.
• Documented, biologically expected rises in myocarditis/pericarditis did occur in 2021 among service members (and civilians) after mRNA vaccination; those are not explained by a 2016–2020 undercount.
• Announcement of the ‘error’ followed public leak of the charts, which some interpret as reactive rather than a pre-existing technical finding.

72/100 — the lab’s own confidence in the score above, not a second reading of the claim.

JAMA Cardiologyjournal · 95 · CDCgov · 93