mm7109e2 — Anomalies

Central finding #

This MMWR leans on VAERS for its serious-safety story (myocarditis work-up; the claim that serious events are rare; “No deaths were reported to VAERS”). From v-safe, readers mainly get mild reactogenicity plus one hospitalization anecdote—a teenager hospitalized for “new onset migraine.”

Compared to the v-safe dataset for the same booster week, that picture is incomplete. Families reported additional hospitalizations (including surgery), many emergency visits, and dozens of chest-pain reports—while coding often failed to mark hospitalizations correctly, and call-center follow-up for those hospital reports was recorded as unsuccessful. The published narrative makes booster injuries in this age group look rarer and milder than the underlying v-safe records support.

Study window: homologous Pfizer booster 9 Dec 2021–20 Feb 2022; day 0–7 check-ins; primary stratum ages 12–15 (boost on/after 3 Jan 2022). Full write-up: vsafe-analysis/mmwr/mm7109e2/anomalies_report.md.

A. Hospitalizations in v-safe — more than the paper’s single migraine case #

Paper (Results): “One (0.03%) adolescent received care at a hospital during the week after booster dose vaccination for treatment of a new onset migraine; whether hospitalization was the result of vaccination could not be determined.”

When a family checks Hospitalization on a v-safe survey, that is a direct report that the vaccinated person was admitted. Coders should also attach standard medical terms (MedDRA). The term Hospitalisation is the usual code for that event. Free text can record the reason (for example “appendectomy”).

In the 12–15 booster cohort, week-after-booster surveys include three hospitalization check-ins from two adolescents. None of those booster-week responses received the MedDRA term Hospitalisation. One response has no MedDRA codes at all.

So the paper tells the public there was one adolescent hospitalized after the booster, for migraine. In the v-safe dataset for the same kind of check-ins:

• LH6-45059-75313 (age 13): free text “Appendectomy” with ER + hospitalization checked — coded for appendicitis/surgery but not given Hospitalisation. This surgical admission is never described in the Results.
• 8W7-32430-24116 (age 13): “Was discharged today” — coded only as Unevaluable event (a placeholder). The same person had Migraine + Hospitalisation codes from May 2021 (dose-2 era), but the booster hospital row was not labeled as a hospitalization.
• M32-28354-68177 appears in the wider 16–20 age band (call-center age 19, outside 12–17) with Myocarditis + Hospitalisation coded correctly — useful as a contrast, not as an omitted adolescent case. Call-center outcome for all three hospital reporters: Unsuccessful.

Why this matters: A reader would think v-safe saw essentially no hospital-level harm after adolescent boosters except one migraine case. The dataset shows at least one additional serious hospital episode (appendectomy) in the paper’s younger age band, and shows hospitalization was systematically under-labeled in coding.

B. Emergency-room visits — reported in checkboxes, often invisible in coding #

Paper (Table 1): lists a small number of emergency visits after the booster among adolescents.

In the reconstructed 12–15 booster cohort alone, there are 14 day-0–7 check-ins where the family checked that the adolescent went to an emergency room or emergency department. Only one of those responses received an emergency-care-style MedDRA term (Emergency care on FY8-11778-30167). Three have no MedDRA coding at all.

Where free text or codes exist, they are not trivial—examples include chest pain / costochondritis, syncope (fainting), and facial swelling. Those are the kinds of medically attended events a safety paper would be expected to surface when arguing that serious outcomes are rare.

Why this matters: Checkbox ER visits are how v-safe captures “we needed urgent care.” If those visits are not coded—or lack any emergency-care term—the published story can treat ER utilization as a tiny Table 1 cell while the records contain a longer list of adolescents who sought emergency evaluation after the booster.

C–D. Chest pain (ages 12–15) unpublished next to the VAERS myocarditis story #

Paper (Results): “Sixty-four preliminary reports of myocarditis were received, among which 47 were considered serious; 32 (68.1%) of these reports were confirmed… All 32 reports were among adolescent boys…”

The MMWR’s serious cardiac discussion is almost entirely a VAERS myocarditis narrative. The v-safe Results do not tell readers how many adolescents in the booster cohort reported chest pain, even though chest pain after mRNA vaccination is a leading symptom used to evaluate possible myocarditis.

Compared to the v-safe dataset, 36 distinct 12–15 adolescents have a MedDRA Chest pain code in the coding window (9 Dec 2021–31 Mar 2022) after these boosters. Some free-text examples: “sharp chest pain when taking a deep breath,” “chest pains,” “Mild chest pain… when doing cardio.” Several sought clinic or ER care (EKG, bloodwork, etc.—see free-text / visit columns in the grid).

Why this matters: The paper reassures readers with a curated VAERS myocarditis count while leaving this parallel v-safe chest-pain signal off the page. A lay reader never learns that dozens of younger teens in the same booster program had chest pain coded in CDC’s active surveillance system.

Table: one row per registrant (preferring rows that include free-text mention of chest when available).

Cardiac MedDRA around the booster cohort #

Registrant CBD-11266-28604, call-center age 17 (inside the paper’s 12–17 range), wrote free text “pericarditis & long covid” and was coded Pericarditis. That in-scope cardiac diagnosis is not part of the paper’s v-safe serious-event story.

Myocarditis MedDRA hits F75-… / M32-… have call-center ages 18 and 19 (outside 12–17), so they are not counted as adolescent-inclusion cases—but they show cardiac coding does appear in v-safe around booster time, while the MMWR still did not discuss the in-range chest-pain and pericarditis signal above.

Deaths — none under inclusion #

Paper (Results): “No deaths were reported to VAERS.”

For this reconstructed booster cohort through 31 March 2022, the v-safe dataset shows no MedDRA Death or Cardiac arrest codes, and no death-like free text. Unlike several earlier MMWRs in this series, the “no deaths” claim is not contradicted here by death-coded v-safe rows. Noted for consistency; not an anomaly for this paper.

Call-center follow-up vs paper wording #

Paper (Methods): “CDC’s v-safe call center contacts registrants who indicate that medical care was sought after vaccination and encourages completion of a VAERS report, if indicated.”

That sentence implies people who report medical care are reached and guided toward VAERS when appropriate. All three reconstructed hospitalization reporters have call-center / GDIT final outcome Unsuccessful—including the age-13 appendectomy case the Results never mentioned (see hospital grid). The Methods language sounds like a functioning safety net; for these hospital-level reports, follow-up did not succeed.

Summary #

• Reporting gap: one migraine hospitalization in Results vs appendectomy + Unevaluable discharge in the v-safe dataset (ages 12–15)
• Coding failure: no Hospitalisation PT on those booster hospital rows; almost no ER-like coding on ER check-ins
• Reporting gap: 36 Chest pain (12–15) + age-17 Pericarditis unpublished beside VAERS myocarditis
• Anomaly: Unsuccessful call-center outcome for every reconstructed hospitalization reporter
• OK: no in-cohort v-safe Death codes under inclusion

Bottom line: Compared to the v-safe dataset, this paper understates medically serious care after adolescent boosters—omitting hospital-level reports, under-coding ER/hospital events, and leaving chest-pain / pericarditis signals off the page while leaning on VAERS myocarditis for reassurance.