What the Peer-Reviewed Research Actually Says About COVID-19 Vaccine Injuries and Safety Data

▶ Listen to this article

Two things can be true about COVID-19 vaccines at once: the injuries they cause are real, documented, and rarer than the illness they prevent, and the federal safety-net program set up to compensate those injuries approves a very small share of the claims filed against it. This week’s research adds a genuinely new data point to that picture — and it’s worth walking through what the actual peer-reviewed record says, not the shouting on either side of it.

This week’s new finding: does timing in your cycle matter?

A study published July 29 in npj Women’s Health, led by researchers at the London School of Hygiene & Tropical Medicine and the University of Montpellier, is the first to ask whether where a woman is in her menstrual cycle when she gets vaccinated changes her outcomes^[1].

Using prospectively tracked cycle data from the Clue period-tracking app matched to an in-app survey of 1,474 women, the researchers compared women vaccinated during the estrogen-dominant follicular phase against those vaccinated during the progesterone-dominant luteal phase^[1]:

Follicular-phase vaccination was linked to 35% higher odds of reporting any side effect than luteal-phase vaccination.
Follicular-phase vaccination was also linked to a 35-day longer median gap before a subsequent infection (200 days vs. 164 days) — though the researchers flag this as exploratory, since it’s based on only 82 recorded infections.
There was no evidence that follicular-phase vaccination produced more severe side effects, just more of the milder, expected kind.

The likely biology: estrogen tends to prime immune activity, while progesterone dampens it^[1]. It’s one study, not a reason to time your shot around your cycle — but it’s exactly the kind of sex-specific variable vaccine research has historically ignored, and the authors are explicit that it needs replication.

What the consensus science says about real injuries

The most rigorous single document on this question isn’t a press release from either side — it’s the National Academies of Sciences, Engineering, and Medicine’s 2024 evidence review, a congressionally mandated, independent evaluation of the full epidemiological and clinical literature^[2].

The review found the evidence sufficient to accept a causal relationship between the mRNA vaccines (Pfizer-BioNTech and Moderna) and myocarditis, concentrated overwhelmingly in young men shortly after a second dose^[2]. It also found causal evidence linking Johnson & Johnson’s Ad26.COV2.S vaccine to thrombosis with thrombocytopenia syndrome and to Guillain-Barré syndrome, and linking vaccination generally to anaphylaxis^[2].

Context the review is careful to include: in a study of nearly 4 million people in Hong Kong, vaccine-associated myocarditis had one death among 104 cases, versus 84 deaths among 762 cases of myocarditis caused by viral infection itself — the same condition, triggered a different way, with a much better prognosis^[2].

The federal injury-claims numbers

Separately from the science, there’s a federal compensation program for people who believe they were injured: the Countermeasures Injury Compensation Program (CICP), run by HRSA. Its own published data, current as of mid-2026, shows^[3]:

14,146 total COVID-19 countermeasure claims filed, of which 11,055 allege a vaccine-caused injury or death.
Of the claims decided so far, 135 (1.8%) have been found eligible for compensation across all COVID-19 countermeasures; 95 of those are vaccine claims specifically.
CICP has paid out $6 million-plus across 34 approved claims to date.
Against roughly 670 million U.S. vaccine doses administered, that works out to about 16 claims filed per million doses and 0.14 compensation-eligible claims per million doses.

Read that carefully: a low approval rate isn’t proof injuries are rare, and a low claims-per-dose rate isn’t proof the approval bar is fair — CICP has drawn real criticism, including from Congress, over its evidentiary standard and its lack of a formal appeals process. Both numbers are true at once, and neither one alone settles the argument.

Weighed against the other side of the ledger

The injury data doesn’t exist in a vacuum; the same current vaccines are still being measured for effectiveness. A CDC-network study published in JAMA Network Open this June found the 2025-2026 formulation cut medically attended COVID-19 emergency department and urgent care visits by 50% and hospitalizations by 55%, drawn from over 111,000 encounters across 253 ED/urgent-care sites and 179 hospitals in seven states^[4].

That study is also part of the story for a less scientific reason: it was pulled from a scheduled CDC publication in March by agency leadership over stated methodology concerns, before a co-author published it independently in JAMA Network Open^[4]. Whatever the reason, it’s a reminder that the fight over this data isn’t confined to journals — it’s also about who gets to publish it and when.

The Bottom Line

The peer-reviewed record supports real, rare, mostly mild-to-moderate vaccine injuries — myocarditis in young men chief among them — alongside a federal compensation program that has approved fewer than 100 vaccine-injury claims against roughly 670 million doses given. The same current vaccine formulation is independently measured to cut hospitalization risk by more than half. New research, like this week’s menstrual-cycle finding, keeps refining who’s more likely to feel side effects — it isn’t overturning the core safety-and-effectiveness picture.

Have you or someone you know filed a CICP claim, or noticed side effects that seemed to track with timing, cycle, or dose number? Share your experience in the comments — real data points like that are exactly what researchers like these are trying to capture.

References

Cooper, P., Alvergne, A., et al. (2026, July 29). Menstrual cycle phase and its association with COVID-19 vaccine outcomes among period tracking app users. npj Women’s Health. https://www.nature.com/articles/s44294-026-00150-x ↩a ↩b ↩c
National Academies of Sciences, Engineering, and Medicine. (2024, August). Evidence review of the adverse effects of COVID-19 vaccination and intramuscular vaccine administration. https://www.nationalacademies.org/publications/27746 ↩a ↩b ↩c ↩d
Health Resources and Services Administration. (2026, July). Countermeasures Injury Compensation Program (CICP) data. https://www.hrsa.gov/cicp/cicp-data ↩
Wiegand, R.E., Chickery, S., Yang, D., et al. (2026, June). Interim estimated effectiveness of 2025-2026 COVID-19 vaccines in adults using a test-negative design. JAMA Network Open, 9(6), e2625152. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2850668 ↩a ↩b