My name is Jason Page. Along with my colleague Gary Hawkins, I lead the VAERS Data Group (vaersdata.org), an independent research project maintaining a longitudinal archive of all public VAERS data releases for the purpose of comparative historical analysis. I am writing to you because you are one of the longest-standing independent custodians of VAERS data in the country, and because the circumstances I am about to describe leave us in need of exactly the kind of partnership your work makes possible.

I write with both a request and a disclosure, in that order.

1. The Request

We are asking whether you would be willing to make your VAERS data drop archive — or a synchronized mirror of it — available to the VAERS Data Group via sFTP server or secure HTTPS download, either on a recurring basis as new drops are released or as a one-time transfer of your historical archive.

The reason this request has become urgent is explained below.

2. Background — Who We Are and What We Have Built

The VAERS Data Group was formed to do what no single government interface makes easy: maintain a continuous, timestamped, unmodified archive of every VAERS public data release so that researchers can compare what was reported across time, not just what the current cumulative file contains.

We have been building and publishing findings from this archive at vaersdata.org and through research articles at deepdots.substack.com (2023–2024), with analytical tools accessible at chart.vaersdata.org. Our work has been summarized in public communications addressed to federal officials and is documented at:

https://shavidica.cc/page/Projects/VAERS/letter-public-official

That summary includes six research findings derived from comparative analysis of successive VAERS data snapshots — the kind of analysis only a longitudinal archive makes possible:

Exhibit A — Narrative field removal from foreign VAERS reports.
Narrative text fields were removed from a subset of foreign-originated VAERS reports, reducing the contextual clinical detail available to researchers and the public. This change was detectable only by comparing successive releases.

Exhibit B — Deleted follow-up reports confirming deaths.
Follow-up reports that confirmed deaths were removed from the database between release cycles. These deletions were not announced and would not be visible to any researcher relying solely on the current cumulative file.

Exhibit C — Lot code absences in FOIA responses.
Approximately 1,290 Pfizer and 958 Moderna lot codes linked to significant numbers of adverse event reports were absent from FOIA responses, despite being present in the public VAERS data. The discrepancy between FOIA-produced records and public dataset contents is a significant data integrity concern.

Exhibit D — Untagged pulmonary embolism cases.
An 8% sample review identified approximately 408 pulmonary embolism cases lacking proper adverse event tagging, meaning they would not appear in standard symptom queries and are effectively invisible to researchers using filtered search interfaces.

Exhibit E — Lot code errors corrected through AI analysis.
Approximately 150,000 COVID-19 vaccine reports contained lot code entry errors. AI-assisted reanalysis was required to identify and correct these, suggesting a systemic data quality issue that affects the reliability of lot-specific adverse event counts as reported.

Exhibit F — Serious classification undercount.
Analysis found that approximately 85% of reports not tagged as "serious" in the VAERS database nonetheless met established criteria for the serious classification. This systematic undercount materially affects every analysis that relies on the CDC's own seriousness designations.

None of these findings would be detectable without a time-series archive of successive releases. The current cumulative VAERS file reflects only the present state of the database. Historical snapshots are what allow a researcher to ask: what changed between last month and this month, and why?

This is the analytical foundation on which our work — and, we believe, much of the meaningful independent VAERS research being done — depends.

3. The Disclosure — Why This Request Is Urgent

On May 8, 2026, we completed a forensic examination of our VAERS archive following reports of file corruption. The findings were serious.

65 of our 342 archived VAERS ZIP files have been completely and irreversibly destroyed. The destruction was not accidental hardware failure. Forensic analysis of filesystem modification timestamps revealed a minimum of four coordinated attack waves:

The intrusion vector was an unauthorized TeamViewer remote-access daemon found running on our server. Files were overwritten using a wiper technique — 0xFF byte fill followed by pseudo-random data — that leaves zero recoverable content. There are no ZIP signatures anywhere in any of the 65 destroyed files. The technique is consistent with documented destructive malware (Dustman, IsaacWiper) rather than ransomware, indicating the actor's goal was elimination rather than financial leverage.

The destroyed files cover February 2023 through November 2025 — almost precisely the period of peak COVID-19 vaccine adverse event reporting, and the period containing our most significant findings. This was not a random target.

We intend to file incident reports with CISA and the CDC VAERS Program Office. Full forensic documentation is available upon request.

The attack has made clear that a single-point archive, however carefully maintained, is vulnerable. We need a trusted second source, and we need it from someone who has been independently archiving VAERS data longer than almost anyone outside of the government itself.

That is you.

4. Summary and Closing

The VAERS Data Group has spent years building a research-grade longitudinal archive of public health surveillance data and has produced documented findings — on narrative deletions, death report removals, lot code discrepancies, tagging failures, and systematic misclassification — that depend on exactly the kind of comparative historical access that no government interface provides. That archive has now been deliberately targeted and substantially destroyed by a threat actor whose goal appears to be the elimination of independent VAERS research capacity.

We are rebuilding. But we are asking not to rebuild alone.

You have been doing this work since 2003. You understand better than most what it takes to maintain an independent VAERS archive and why it matters. We are asking whether you would be willing to extend that infrastructure to include us — as a verified, credentialed second recipient of the data you already process — so that a documented attack on one archive does not mean the permanent loss of a two-year historical record.

Gary Hawkins and I are both available to discuss any arrangement at your convenience. Please feel free to contact either of us:

Jason Page — pagetelegram@proton.me
Gary Hawkins — VAERS Data Group, vaersdata.org

We would welcome a call, an email exchange, or any form of communication that works for you. We are not asking for a fast answer.

Thank you for more than twenty years of independent work on VAERS data access. It has mattered, and it matters now more than ever.

Respectfully submitted,

Jason Page
pagetelegram@proton.me
Gary Hawkins
VAERS Data Group — vaersdata.org

Attachments available upon request:

• VAERS Archive Corruption — Forensic Report (full technical detail, 65 destroyed files, attack timeline)
• VAERS Data Group Research Summary — Exhibits A–F (shavidica.cc/page/Projects/VAERS/letter-public-official)
• CISA Incident Report (generated May 8, 2026)
• CDC VAERS Program Office Notification (generated May 8, 2026)