Summary of results
- The COVID lots flagged for elevated adverse-event reporting were administered mainly in the earliest rollout phases, when specific groups — not the general public — were eligible.
- Anaphylaxis-flagged lots were among the first doses (Dec 2020), a period when healthcare workers (median reporter age ~45) were being vaccinated and acute allergic reactions were under heightened surveillance.
- Death-flagged lots were administered weeks later (Jan–Feb 2021), during the phase covering the elderly and long-term-care population (median reporter age ~67), a group with high background mortality.
- Vaccination date, recipient age, and the rollout schedule point in the same direction: much of the between-lot difference is associated with when the lot was used and who received it. How much of any residual reflects the lots themselves cannot be determined from these data.
⚠️ Scope
- VAERS is passive, unverified reporting. This page examines factors that vary alongside the flagged lots — it is not a safety determination about any lot in either direction.
- “Flagged” lots are those that showed elevated reporting of specific serious events in a dose-denominated disproportionality screen (see the cluster analysis).
- Recipient age is taken from the reports themselves; the age distribution of everyone who received each lot is not available.
The question
A dose-denominated screen flagged a handful of lots for elevated serious-event reporting — lots sometimes called “hot lots.” Two explanations are commonly offered: a batch-specific defect, or differences in the recipient population. Because the U.S. rollout was sharply phased by group, the second can be examined directly: when were these lots administered, and to whom? This page reports that comparison; it does not resolve the first explanation.
The evidence: two populations, two rollout phases
Each point is a flagged lot — horizontal = median vaccination date, vertical = median recipient age. The dashed line is the median vaccination date across all COVID reports (2021-03-24). The flagged lots sit toward the left (earlier doses) and fall into a younger group and an older group.
| Lot | Signal | Median vax date | Median age | % Dec 2020 | % by Jan 31 2021 | Crude excess | Age-adjusted |
|---|
EH9899 | Anaphylaxis | 2020-12-21 | 44 | 77% | 98% | 2.1× | 1.7× |
EK5730 | Anaphylaxis | 2020-12-22 | 45 | 74% | 98% | 2.7× | 2.1× |
EK9231 | Anaphylaxis | 2021-01-08 | 45 | 22% | 97% | 2.4× | 2.0× |
EL9261 | Death | 2021-01-29 | 63 | 1% | 56% | 3.7× | 2.1× |
EM9810 | Death | 2021-02-11 | 69 | 0% | 7% | 3.5× | 1.9× |
EL9267 | Death | 2021-02-18 | 68 | 0% | 5% | 3.1× | 1.7× |
EN6201 | Death | 2021-02-19 | 68 | 0% | 4% | 3.2× | 1.7× |
Baseline: across all COVID reports, only 5.3% were vaccinated in Dec 2020 and 21.7% by Jan 31 2021. The anaphylaxis lots, at 74–77% vaccinated in Dec 2020, are about 14× that share — among the earliest doses administered. “Age-adjusted” = excess remaining after standardizing for the reporter age mix.
Mapping to the documented U.S. rollout
Dec 14, 2020 — Phase 1a begins
First doses nationwide. ACIP Phase 1a = healthcare personnel + long-term-care residents.
~Dec 21, 2020 — LTC program
Federal pharmacy long-term-care (nursing home) vaccination program begins. First U.S. anaphylaxis cases reported in healthcare workers this week → CDC issues intensive anaphylaxis monitoring guidance.
Jan 12, 2021 — Phase 1b
Recommended for adults 75+ and frontline essential workers.
Feb–Mar 2021 — Phase 1c
Adults 65–74 and high-risk 16–64. General adult eligibility (16+) not until Apr 19, 2021.
Anaphylaxis lots — Phase 1a period (healthcare workers)
Vaccinated Dec 21–22, 2020, median reporter age ~45. These were among the earliest doses, administered while healthcare workers were the eligible group and the first anaphylaxis cases were prompting heightened surveillance. The anaphylaxis signal is not explained by recipient age; it coincides with this early, intensively-monitored period, though the data cannot separate heightened reporting from a real difference.
Death lots — Phase 1b/1c period (elderly & LTC)
Vaccinated late Jan–Feb 2021, median reporter age ~67 — the phase covering the 65–74 / 75+ and long-term-care populations. Background mortality in this group is high: the death-report fraction rises steeply with age (below), so more death reports would be expected for a lot given to older recipients irrespective of the vaccine.
The effect of recipient age on death reporting
Death-report fraction by reporter age band, across all COVID reports:
<300.1%
30-490.3%
50-641.2%
65-743.0%
75+9.4%
The death-report fraction is roughly 90× higher in the 75+ band than under-30s. Standardizing for the reporter age mix removes 59–68% of the death lots’ apparent excess (see table); a residual remains.
Interpretation
Vaccination date, recipient age, and the ACIP rollout schedule point in the same direction: the flagged lots were used mainly in the earliest phases, for the specific groups eligible then, and separate by event type along a working-age-vs-elderly line. Much of the between-lot difference is associated with these timing and population factors.
A residual of roughly 2× remains after age adjustment. It is consistent with the more intensive reporting these first, closely-watched lots received, but the available data cannot confirm or rule out a contribution from the lots themselves. A reader weighing a batch-effect explanation should also account for the observation that the flagged lots coincide closely with the Phase-1a/1b lots and separate by event type along the two populations those phases covered.
Data: VAERS raw records (VAX_DATE, AGE_YRS) × lot disproportionality screen. Rollout milestones per CDC/ACIP. Passive surveillance; hypothesis-generating only. Reproducible: lot_rollout_corroboration.py.