Ana Szarfman, MD, PhD

Niu emails Anderson about a meeting with Szarfman to discuss new methods for data mining.

Szarfman organizes a virtual meeting on a new safety data mining methodology designed by DuMouchel.

Menschik and Baer exchange about classic data mining references written by Szarfman and DuMouchel.

Szarfman raises COVID-19 vaccine safety monitoring concerns with CBER Director Marks.

Szarfman presents RGPS to Marks and other CBER officials, calling MGPS not state of the art.

Szarfman confirms she has access to VAERS data through Oracle's Empirica.

Zinderman tells Baer he has no issue with Szarfman's VAERS access through Empirica.

Baer questions Szarfman's COVID vaccine data mining; Zinderman cautions against blocking access.

Baer acknowledges Szarfman knows more about data mining but characterizes her comments as atypical.

Niu writes Baer that she and Zinderman will meet with Szarfman about her VAERS objectives.

Szarfman shares DuMouchel's RGPS analysis showing 49 examples of extreme masking with senior FDA officials.

DuMouchel sends analysis adjusting for the fact that 99% of 2021 VAERS reports are for COVID-19 vaccines.

Niu asks whether DuMouchel's findings affect 2021 data mining results; Szarfman confirms they do.

Niu forwards DuMouchel's analysis to Zinderman without Szarfman; Zinderman flags it as worth considering.

Baer welcomes DuMouchel's input; Menschik pushes back on reaching out to Szarfman directly.

Szarfman shares DuMouchel's RGPS analysis revealing six new statistically significant safety signals.

Baer reports to colleagues that Szarfman offered to show Commonwealth her VAERS analysis; Niu distances herself.

Szarfman resends RGPS analysis flagging cardiac signals; Niu tells her CBER cannot collaborate.

Szarfman shares an EB data mining analysis; Niu confirms to Menschik she is not collaborating.

Szarfman circulates analysis showing MGPS undervalued acute myocardial infarction signals for Pfizer and Moderna.

Zinderman directs Szarfman to hold off on creating COVID-19 vaccine data mining reports and analyses.

Szarfman receives clearance from Stockbridge to publish her paper on masking and RGPS.

Szarfman resends RGPS analysis on acute myocardial infarction and invites Forshee to collaborate.

Szarfman calls Forshee to discuss increased risk of mortality following COVID-19 vaccination.

Szarfman writes Forshee that RGPS shows statistically significant mortality signals for Pfizer and Moderna vaccines.

Forshee forwards Szarfman's mortality findings to Marks, expressing concern about her interpretation.

Szarfman tells Commonwealth contractor that the 20-year-old MGPS model could potentially mask signals.

Szarfman clarifies masking concerns to Commonwealth and reiterates RGPS advantages over MGPS.

Menschik tells Szarfman that VAERS data mining is CBER's responsibility and discussions must stay in chain of command.

Anderson asks Marks that Szarfman refrain from using FDA email or communicating findings using CBER VAERS data.

Marks tells Cavazzoni that Szarfman has been asked to cease and desist on her vaccine analysis.

Menschik shares draft article language with CDC's Su that echoes Szarfman's masking concerns without crediting her.

Walinsky circulates Szarfman's preprint to FDA officials weighing Moderna's pediatric COVID-19 vaccine appeal.

Szarfman, DuMouchel, and co-authors publish 'Signaling COVID-19 Vaccine Adverse Events' in Drug Safety.

Szarfman emails her unmasking paper to Commissioner Califf, who replies, 'Thanks. These are good.'

Forshee notifies Marks and Anderson that FDA officials are addressing issues with Szarfman's published paper.

FDA officials discuss the masking effect three years after Szarfman raised it; Baer points to her published paper.

Baer tells Menschik RGPS yields significantly higher signals than MGPS; Menschik concurs but prefers to keep MGPS.