paxchristi on Nostr: CRITICAL ASSESSMENT "Impact of Childhood Vaccination on Short and Long-Term Chronic ...

CRITICAL ASSESSMENT
"Impact of Childhood Vaccination on Short and Long-Term Chronic Health Outcomes in Children: A Birth Cohort Study"

Authors: Lamerato et al., Henry Ford Health System
Assessment Date: 2026-03-05

================================================================================

STUDY OVERVIEW

Design: Retrospective birth cohort, 18,468 children born 2000-2016
Setting: Henry Ford Health System / Health Alliance Plan (HAP), Michigan
Groups: 16,511 vaccinated vs. 1,957 completely unvaccinated
Headline Claim: Vaccination associated with 2.5x increased risk of developing
a chronic health condition (HR 2.53, CI 2.16-2.96)

================================================================================

VERDICT: LIKELY INVALID

The study's design cannot distinguish between "vaccines cause chronic
conditions" and "children who visit doctors more get diagnosed more." The
evidence strongly favors the latter explanation.

================================================================================

MAJOR METHODOLOGICAL FLAWS

1. DETECTION/ASCERTAINMENT BIAS (FATAL FLAW)

Vaccinated children averaged 7 healthcare visits per year. Unvaccinated
children averaged 2 per year. More visits means more opportunities to receive
a diagnosis. The study is effectively measuring healthcare utilization, not
vaccine effects.

The ear infection finding exposes this directly: vaccinated children showed a
7x higher rate of ear infections (HR 7.00, CI 6.05-8.10). Ear infections are
one of the most common childhood conditions and have no plausible biological
link to vaccination. This result alone demonstrates that the unvaccinated
group's conditions are being massively under-detected.

The study did not adjust for number of healthcare encounters in its regression
models -- arguably the single most important confounder.


2. ZERO CASES IN THE UNVACCINATED GROUP -- STATISTICALLY IMPLAUSIBLE

For several common conditions, the study found zero cases among ~2,000
unvaccinated children:

Condition Population Prevalence Cases in Unvaccinated
---------------------------------------------------------------------------
ADHD ~10% 0
Diabetes ~0.25% 0
Tics ~1% 0
Behavioral disability Common 0
Learning disability ~5-10% 0
Intellectual disability ~1-3% 0

Zero ADHD cases in 2,000 children is not biologically credible. This means
these children were not being evaluated and diagnosed -- not that they didn't
have these conditions.


3. MASSIVE DIFFERENTIAL FOLLOW-UP

Group Median Follow-Up
----------------------------------------
Vaccinated 970 days (2.7 years)
Unvaccinated 461 days (1.3 years)

Vaccinated children were observed for more than twice as long. More observation
time means more opportunity to develop and detect conditions.

The sensitivity analysis (restricting to longer enrollment periods) paradoxically
made the association stronger:

Minimum Enrollment Hazard Ratio
------------------------------------
Any 2.54
1 year 2.84
3 years 3.48
5 years 4.05

If the association were real, restricting to longer follow-up should stabilize
the HR, not increase it. This escalating pattern is a classic signature of
selection bias -- unvaccinated children who remain enrolled for 5+ years are an
increasingly select subgroup of healthy, low-utilization families.


4. UNCONTROLLED CONFOUNDING

The Cox models adjusted for gender, race, birthweight, prematurity, respiratory
distress, and birth trauma. They did NOT adjust for:

- Number of healthcare encounters (the most critical confounder)
- Socioeconomic status
- Parental education
- Diet and lifestyle
- Environmental exposures
- Breastfeeding status
- Household size / daycare attendance

Parents who refuse all vaccines tend to have fundamentally different healthcare-
seeking behaviors, diets, and lifestyles. Without controlling for these, the
study cannot isolate any effect of vaccines themselves.


5. HEALTHY USER BIAS / SELF-SELECTION

The unvaccinated group is self-selected -- these are families who deliberately
chose zero vaccines. This group differs from vaccinating families in ways that
go far beyond vaccine exposure:

- Lower healthcare utilization (confirmed: 2 vs. 7 visits/year)
- Likely different socioeconomic profile
- Different dietary and lifestyle patterns
- Different relationship with the medical system

The study cannot disentangle "effect of vaccines" from "effect of being the
type of family that vaccinates and engages with healthcare."


6. NOT PEER-REVIEWED

This paper was "entered into hearing record" -- submitted for a government
hearing. It was not published in a peer-reviewed journal, meaning it was not
subjected to the standard scrutiny where epidemiologists would have identified
these exact issues during review.


7. MULTIPLE COMPARISONS WITHOUT CORRECTION

The study tested 20+ individual outcomes without applying any correction for
multiple comparisons (e.g., Bonferroni, false discovery rate). At a significance
threshold of p<0.05 across 20+ tests, several statistically significant results
would be expected purely by chance.


8. NO DOSE-RESPONSE ANALYSIS

If vaccines truly caused these conditions, a dose-response relationship would be
expected -- more vaccines should correlate with more disease. The study
categorized exposure only as "any vs. none" and never tested whether the number
of vaccines (median 18 in the exposed group) correlated with risk. This omission
weakens any causal argument.


9. TIME-VARYING EXPOSURE CLASSIFICATION

Vaccination status was classified as "exposure prior to onset of each condition."
A child could accumulate years of "unexposed" person-time before receiving their
first vaccine and being reclassified. This creates immortal time bias -- the
unexposed group accumulates guaranteed disease-free time, artificially improving
their outcomes.

================================================================================

WHAT THE STUDY ACTUALLY SHOWS

The core finding -- that vaccinated children are diagnosed with more conditions
-- is entirely consistent with a simple, well-known phenomenon: children who
see doctors get diagnosed; children who don't see doctors don't.

This is supported by:

1. The 3.5x difference in annual healthcare visits between groups
2. Zero cases of common conditions (ADHD, learning disability) in unvaccinated
3. A 7x ear infection rate -- a condition with no vaccine mechanism
4. Escalating HRs with longer enrollment (selection bias signature)
5. No dose-response relationship tested

================================================================================

COMPARISON WITH SCIENTIFIC CONSENSUS

Finding Consistent with consensus?
-----------------------------------------------------------------------
No association with autism Yes
No association with cancer Yes
Association with anaphylaxis Yes (known rare adverse event)
2.5x overall chronic disease risk No -- not supported by large studies
5.5x neurodevelopmental risk No -- contradicts extensive research
4.3x asthma risk No -- mixed evidence, no effect this large

Large-scale studies in Denmark, Finland, and other countries with comprehensive
health registries and minimal ascertainment bias have not found associations of
this magnitude.

================================================================================

SUMMARY OF ISSUES

Issue Severity
-----------------------------------------------------------------------
Detection bias (3.5x visit difference) CRITICAL (likely explains all)
Zero cases of common conditions in unvaccinated CRITICAL (confirms under-detection)
Differential follow-up (2x longer in vaccinated) High
No adjustment for healthcare utilization High
Uncontrolled socioeconomic confounding High
Self-selection / healthy user bias High
Not peer-reviewed Moderate
Multiple comparisons without correction Moderate
No dose-response analysis Moderate
Immortal time bias potential Moderate

CONCLUSION: The study's methodology cannot support its claims. The overwhelming
likelihood is that the observed associations reflect differences in healthcare
utilization and detection, not causal effects of vaccination.