If the first report is late or incomplete, Day 0, case validity, and everything after them work from delayed or missing data. Six published figures below. Each one has a limit printed on the card.
These are literature and regulatory figures, not AE Connect measurements. Spontaneous-report totals are not incidence rates. A reported outcome is not proof that a medicine caused the event.
-
Underreporting
94%
Median ADR underreporting
A 2006 systematic review found a median underreporting rate of 94%, with an interquartile range of 82–98%.
Limit. Hazell and Shakir pooled 37 studies and 43 estimates from 12 countries. Methods and settings differ. This is a 2006 literature median. It is not an official FDA or EMA rate, and it is not a rate for this year.
-
When the clock starts
Day 0
Valid case, then 15 days
Day 0 is the date the organization first has the minimum criteria for a valid individual case safety report. An incomplete report is followed up. Expedited reports are due as soon as possible and no later than 15 calendar days after Day 0.
Limit. This is the ICH E2D(R1) and EU GVP Module VI rule. National rules add detail, so this card is not every country’s clock.
-
Manufacturer to FDA
9.94%
Expedited reports past day 15
About 9.94% of roughly 1.6 million expedited manufacturer reports were not received by FDA within 15 days. Reports associated with a death were delayed more often than reports that were not.
Limit. Ma and colleagues studied January 2004 through June 2014, and they measured the lag after the manufacturer had the information. That is not the lag from a reporter’s first email to case creation. The paper does not count deaths that followed from the delay.
-
Onset to FDA
99–126 days known ADRs173–328 days unknown ADRs
Unknown reactions were reported later
For the pairs in this study, the median time from onset to the FDA report was about 99–126 days when the ADR was already known, and about 173–328 days when it was not. In their counterfactual, earlier reporting shortened time-to-signal.
Limit. Six drug–ADR pairs only. The shorter time-to-signal is a model that assumed reports at day 15 from onset. It is not a trial of intake software.
-
Death outcome flag
1.88 million
Reports flagged with a death outcome
The openFDA public extract contained about 1.88 million drug-event records with a death outcome flag. That flag marks the reported outcome. It does not establish that a drug caused the death.
Limit. The API total was 1,883,487 records with seriousnessdeath:1. The extract’s meta.last_updated was 30 July 2026. openFDA is a public extract, not a live FAERS dashboard, and spontaneous reports are not a death toll.
-
Global database
>40 million
Reports in VigiBase
VigiBase, the WHO global database of reported potential side effects, held more than 40 million reports as of February 2025.
Limit. Uppsala Monitoring Centre figure. This is the size of a spontaneous-report database. It is not an incidence rate and it is not a count of deaths.
Sources
Citations assembled 28 October 2026. Figures are quoted from the sources named. AE Connect did not recompute them. The openFDA count is the public extract dated in that entry, not a later recount.
- Hazell L, Shakir SAW. Under-Reporting of Adverse Drug Reactions: A Systematic Review. Drug Safety. 2006. https://pubmed.ncbi.nlm.nih.gov/16689555/ · DOI https://doi.org/10.2165/00002018-200629050-00003
- ICH E2D(R1) Post-Approval Safety Data: Definitions and Standards for Expedited Reporting. Step 4, 2025.
https://database.ich.org/sites/default/files/ICH_E2D%28R1%29_Step4_FinalGuideline_2025_0819.pdf
· FDA copy https://www.fda.gov/media/176977/download
EMA. Guideline on good pharmacovigilance practices (GVP) Module VI, Rev 2 (in effect 22 November 2017). https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/guideline-good-pharmacovigilance-practices-gvp-module-vi-collection-management-submission-reports-suspected-adverse-reactions-medicinal-products-rev-2_en.pdf - Ma P, Marinovic I, Karaca-Mandic P. Drug Manufacturers’ Delayed Disclosure of Serious and Unexpected Adverse Events to the US Food and Drug Administration. JAMA Internal Medicine. 2015. https://doi.org/10.1001/jamainternmed.2015.3565 · https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2398401
- Hashiguchi M, et al. Factors Affecting the Timing of Signal Detection of Adverse Drug Reactions. PLOS ONE. 2015. https://doi.org/10.1371/journal.pone.0144263
- openFDA drug-event API, records with seriousnessdeath:1. Public extract meta.last_updated 30 July 2026. Total in that extract: 1,883,487. https://api.fda.gov/drug/event.json?search=seriousnessdeath:1 · openFDA FAERS notes https://open.fda.gov/data/faers/ · FDA, FAERS reporting by patient outcomes by year https://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/faers-reporting-patient-outcomes-year
- Uppsala Monitoring Centre. About VigiBase. Figure stated as of February 2025. https://who-umc.org/vigibase-data-access/about-vigibase/