019736 S196 M5 c4591012 protocol

Pfizer Documents (PHMPT/FDA)

Pfizer Bla Submission

Pfizer 16 Plus Documents

145

Document text

BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 1of 144NON-INTERVENTIONAL ( NI) STUDY CONCEPT PR OTOCOL
Title Post-Emergency  Use Authorization Active 
Safety Surveillance Study  among Individuals 
in the Veteran’s Affairs Health Sy stem 
Receiving Pfizer -BioNTech Coronavirus 
Disease 2019 (COVID -19) Vaccine
Protocol number C4591012
Protocol version identifier Final Version 1.0
Date of last version of protocol 27January 2021
EUPost Authori zation Study (PAS) 
register numberTo be registered before the start of data 
collection 
Active substance COVID-19 mRNA Vaccine is single -stranded, 
5’-capped messenger RNA (mRNA) produced 
using a cell -free in vitro transcription from the 
corresponding DNA templates, encoding the 
viral spike (S) protein of SARS -CoV-2.
Medicinal product Pfizer-BioNTech COVID- 19 Vaccine
Research question and objectives Research question: what are the incidence rates 
of safety events of interest (based on adverse
events of special interest[AESI]) among 
individuals vaccinated with the Pfizer-
BioNTech COVID-19 vaccine within the US 
Veterans Health Administration (VHA) s ystem
overall and in sub- cohorts of interest, as 
compared to expected rates of those events?
Primary study objectives:
To assess whether in dividuals in the 
VHA system experience increased risk 
of safety events of interest following 
receipt of the Pfizer- BioNTech 
COVID-19 vaccine;
To assess whether sub- cohorts of 
interest (i.e., immunocompromised, 
elderly, individuals with specific 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073210
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 2of 144comorbiditi es, individuals receiving 
only one dose of the Pfizer -BioNTech 
COVID-19 vaccine, and individuals 
with prior SARS -CoV-2 infection) in 
the VHA s ystem experience increased 
risk of safet y events of interest 
following receipt of the Pfizer-
BioNTech COVID -19 vaccine.
Secondary study objective :
To characterize utilization patterns of 
the Pfizer -BioNTech COVID-19 
vaccine among individuals within the 
VHA, including estimating the 
proportion of individuals receiving 
vaccine, 2 -dose vaccine completion 
rate, and distri bution of time gaps 
between the first and second dose, 
demographics and health histories of 
recipients, overall and among the sub-
cohorts of interest. 
Authors Yinong Young- Xu, ScD, MA, MS
Director, Clinical Epidemiology  Program
Veterans Affairs Medical Center
White River Junction, VT 
Cynthia de Luise, PhD, MPH
Senior Epidemiologist/ Safety Surveillance 
Research Scientist ; Risk Management and 
Safety Surveillance Research
Pfizer, Inc.
New York, NY
Mei Sheng Duh, ScD, MPH 
Managing Principal and Chief Epidemiologist 
Analysis Group, Inc.
Boston, MA
This document contains confidential information belonging to Pfizer. Except as otherw ise agreed to in w riting, 
by accepting or reviewing this document, you agree to hold this information in confidence and not copy or 
disclose it to others (except where required by applicable law) or use it for unauthorized purposes. In the event 
of any actual or suspected breach of this obligation, Pfizer must be promptly notified.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073211
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 3of 1441.TABLE OF CONTENTS
1.TABLE OF CONTENTS.......................................................................................................3
2. LIST OF ABBREVIAT IONS................................ ................................ ................................ 5
3. RESPONS IBLE PARTI ES................................ ................................ ................................ ....7
4.ABSTRACT ................................ ................................ ................................ ........................... 8
5.AMENDMENTS AND UP DATES................................ ................................ ..................... 21
6.MILESTONES ................................ ................................ ................................ ..................... 22
7. RATIONALE AND BAC KGROUND ................................ ................................ ................ 23
8. RESEARCH QUESTION AND O BJECTIVES................................ ................................ .24
9. RESEARCH METHODS................................ ................................ ................................ ....25
9.1. Study  Design................................ ................................ ................................ ...........25
9.1.1. Self -Controlled Risk I nterval (SCRI) Design ................................ .............25
9.1.2. Active Comparator Design ................................ ................................ .........27
9.1.3. Study  Period................................ ................................ ................................ 28
9.2. Setting ................................ ................................ ................................ ...................... 28
9.2.1. Inclusion Criteria ................................ ................................ ........................ 28
9.2.2. Exclusion criteria ................................ ................................ ........................ 28
9.2.3. Subgroups ................................ ................................ ................................ ...28
9.3. Variables ................................ ................................ ................................ .................. 29
9.3.1. Exposure of I nterest................................ ................................ .................... 29
9.3.1.1. Pfizer -BioNTech COVID- 19 Vaccine Groups of Interest ........30
9.3.2. Baseline Characteristics ................................ ................................ ..............31
9.3.3. Outcomes ................................ ................................ ................................ ....32
9.4. Data Source................................ ................................ ................................ .............39
9.5. Study  Size................................ ................................ ................................ ................ 40
9.5.1. Power ................................ ................................ ................................ ..........40
9.6. Data Management ................................ ................................ ................................ ...42
9.6.1. Case report forms (CRFs)/Electronic data record ................................ ......42
9.6.2. Record retention ................................ ................................ .......................... 42
9.7. Data Anal ysis................................ ................................ ................................ ..........43
9.7.1. Baselin e Characteristics ................................ ................................ ..............43
9.7.2. Vaccine Utilization Patterns ................................ ................................ .......44
9.7.3. Safet y Signal Analyses ................................ ................................ ...............44
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073212
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 4of 1449.7.3.1. Signal Detection ................................ ................................ ........46
9.7.3.2.Signal Evaluation ................................ ................................ ......49
9.7.3.3. Signal Verification ................................ ................................ ....51
9.7.4. Seasonality -Adjusted Cases -Centered Method ................................ ...........51
9.7.5. End-of- Season and End -of-Surveillance Analy ses................................ .....52
9.7.6. Subgroup Analy sis................................ ................................ ...................... 53
9.7.7. Incidence Rates and Time to Safety  Event of Interest Anal ysis................. 53
9.8. Quality  Control................................ ................................ ................................ ........53
9.9. Strengths and Limitations of the Research Methods ................................ ...............54
9.10. Other Aspects ................................ ................................ ................................ ........55
10. PROTECTI ON OF HU MAN SUBJECTS ................................ ................................ ........55
10.1. Patient I nformation ................................ ................................ ................................ 55
10.2. Patient Consent ................................ ................................ ................................ ......56
10.3. Institutional Review board (I RB)/Independent Ethics Committee (I EC).............56
10.4. Ethical Conduct of the Study ................................ ................................ ................ 56
11. MANAGEMENT AND R EPORTING OF ADVERSE EVENTS/ADVERSE
REACTIONS................................ ................................ ................................ ...................... 56
12. PLANS FOR DI SSEMINATING AND COMMUNI CATING STUDY RESUL TS........58
13. REFERENCES ................................ ................................ ................................ .................. 59
14. LIST OF TABLES ................................ ................................ ................................ .............63
15. LIST OF FIGURES ................................ ................................ ................................ ...........63
16. ANNEX 1. LIST OF STAND ALONE DOCUMEN TS................................ ................... 64
17. ANNEX 2. ENCEPP CHECKLIST FOR STUDY PROTOCOL S................................ ...64
18. ANNEX 3. ADDITIO NAL INFORMATION ................................ ................................ ...64
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073213
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 5of 1442. LIST OF ABBREVIATIONS
Abbreviation Definition
ACIP Advisory Committee on Immunization Practices
ADEM Acute disseminated encephalomy elitis
AE Adverse event
AEM Adverse event monitoring
AESI Adverse event of special interest
AIDS Acquired immunodeficiency  syndrome
AMI Acute myocardial infarction 
BMI Body mass index
CAD Coronary  artery disease
CI Confidence Interval
CCI Charlson comorbidity  index
CDC Centers for Disease Control and Prevention
CDW Corporate Data Warehouse
CEP Clinical Epidemiology  Program
CIDP Chronic inflammatory  demyelinating pol yneuropathy
CMA Conditional Marketing Authorization
COPD Chronic obstructive pulmonary  disease
COVID-19 Coronavirus Disease 2019
CPT Current Procedural Terminology
CRADA Cooperative Research and Data Agreement
CRFs Case report forms
DIC Disseminated intravascular coagulation
DVT Deep vein thrombosis
TDap Diphtheria, tetanusand (acellular) pertussis
Td Diphtheria and tetanus
ED Emergency  department
EMA European Medicines Agency
EMR Electronic medical records
EU European Union
EUA Emergency  Use Authorization
EU PAS European Union Post -Authorization Safety
FDA Food and Drug Administration
GBS Guillain-Barré syndrome
GEP Good Epidemiological Practice
GPP Good Pharmacoepidemiology  Practices
H0 Null hypothesis
Ha Alternative h ypothesis
HBV Hepatitis B virus 
HCPCS Healthcare Common Procedure Coding S ystem
HCV Hepatitis C virus
HIV Human immunodeficiency virus
HPV Human papillomavirus
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073214
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 6of 144Abbreviation Definition
ICD-10-CM International Classification of Diseases, Tenth Revision, Clinical 
Modification
ICD-10-PCS International Classification of Diseases, Tenth Revision, Procedure 
Coding Sy stem
IEA International Epidemiological Association
IEC Independent Ethics Committee
IQR Interquartile range
IRB Institutional Review Board
KD Kawasaki disease
LLR Log-likelihood ratio
MaxSPRT Maximized sequential probability  ratio test
MenACWY Meningococcal conjugate
MenB Serogroup B meningococcal 
MIS-A Multisystem inflammatory  syndromein adults
mRNA Messenger RiboNucleic Acid
MS Multiple sclerosis
NIS Non-interventional study
ON Optic neuritis
PASS Post-Authorization Safety  Study
PRISM Post-Licensure Rapid Immunization Safety  Monitoring
RCA Rapid cycle analysis
RR Relative risk
SAP Statistical analy sis plan
SARS-CoV-2 Severe acute respiratory  syndrome coronavirus 2
SAS SAS Institute
SCRI Self-controlled risk interval
SD Standard deviation
SPEAC Safety Platform for Emergency  vACcines
TM Transverse m yelitis
UK United Kingdom
US United States
VA Department of Veterans Affairs
VAERS Vaccine Adverse Event Reporting S ystem
VHA Veterans Health Administration
VINCI VA Informatics and Computing Infrastructure
VISN Veterans Integrated Service Networks
VSD Vaccine Safet y Datalink
VTE Venous thromboembolism
WHO World Health Organization
WOC Without compensation
YRR Your Reporting Responsibilities
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073215
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 7of 1443.RESPONSIBLE PARTIES
Name, degree(s) Job Title Affiliation Address
Principal Investigator:
Yinong Young -Xu, 
ScD, MA, MSDirector, Clinical Epidemiology 
ProgramVeterans Affairs 
(VA) Medical 
Center163 Veterans Drive, 
White River Junction, 
VT 05009
Cynthia de Luise, 
PhD, MPHSenior Epidemiologist /Safety 
Surveillance Research Scientist ;
Risk Management and Safety 
Surveillance ResearchPfizer, Inc. 235 East 42nd Street, 
New York, NY 10017
Mei Sheng Duh, 
ScD, MPHManaging Principal and Chief 
Epidemiologist
Visiting Scientist, Department of 
BiostatisticsAnalysis Group, 
Inc.
Harvard T. H. 
Chan School of 
Public Health111 Huntington Ave
14thFloor
Boston, MA 02199
677 Huntington Ave
Boston, MA 02115
Maral DerSarkissian, 
PhD Vice President and Senior 
Epidemiologist
Adjunct Assistant ProfessorAnalysis Group, 
Inc.
Fielding School of 
Public Health, 
University of 
California, Los 
Angeles333 South Hope Street
27th Floor
Los Angeles, CA
90071
650 Charles E Young 
DriveSouth
Los Angeles, CA 
90095
Rachel Bhak, MS Manager and Senior Biostatistician Analysis Group, 
Inc. 111 Huntington Ave
14thFloor
Boston, MA 02199
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073216
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 8of 1444.ABSTRACT
Title:Post-Emergency  Use Authorization Active Safety  Surveillance Study  among 
Individuals in the Veteran’s Affairs Health S ystem Receiving Pfizer- BioNTech Coronavirus 
Disease 2019 (COVID -19) Vaccine
Protocol Version: 1.0; Dateof Protocol : 27January 2021
Authors: Yinong Young Xu, ScD, MA, MS , Veterans Affai rs Medical Center; Cy nthia de 
Luise, PhD, MPH, Pfizer, I nc.; Mei Sheng Duh, ScD, MPH, Anal ysis Group, I nc.
Rationale and b ackground :
In March 2020, the World Health Organization (WHO) declared a global pandemic for the 
coronavirus disease 2019 (COVID -19) due to the severe acute respiratory  syndrome 
coronavirus 2 (SARS -CoV-2), which was first identified by  public health officials in China 
in December 2019.1The COVID -19 pandemic presents an unprecedented public health 
crisis.As of January 7, 2021, over 21.4million COVID -19 cases and 364,000 deaths have 
been reported in the United States (US) alone.2
Pfizer and BioNTech have partnered to develop a novel messenger RiboNucleic Acid 
(mRNA)vaccine against SARS -CoV-2 for the prevention of COVID -19 (Candidate 
BNT162b2). Pfizer is conducting a Phase 1/2/3, randomized, placebo -controlled, observer -
blind, dose-finding, vaccine candidate- selection, and efficacy  study among healthy  
individuals (NCT04368728). The Food and Drug Administration (FDA) reviewed the 
available safet y data from 37,586 participants 16 years of age and older and did not identify  
anyspecific safet y concerns. In addition, the anal ysis of available efficacy  data from 36,523 
participants 12 years of age and older without evidence of prior SARS- CoV-2 infection at 
least 7 day s after receiving the second dose demonstrated 95% efficacy  of the vaccine in the 
prevention of COVID -19 (as confirmed by  8 vs. 162 COVID -19 cases in the vaccine and 
placebo groups, respectively ).3,4Based on these safety  and efficacy  data, as well as a revi ew 
of manufacturing information regarding product quality  and consistency , the FDA 
determined that the known and potential benefits of the vaccine outweighed the known and 
potential risks for the prevention of COVID -19 in individuals 16 y ears of age and ol der.4
Therefore on December 11, 2020, the Pfizer -BioNTech COVID-19 vaccine was granted an 
Emergency  Use Authorization (EUA) by  the FDA to prevent COVID -19 in individuals 16 
years of age and older.5
With respect to geographi c regions other than the US, on December 2, 2020, the United 
Kingdom (UK) was the first country  in the world to grant temporary  authorization for 
emergency  use of the Pfizer -BioNTech COVID -19 vaccine.6On December 21, 2020, the 
European Medicines Agency  (EMA) granted the Pfizer -BioNTech COVID- 19 vaccine a 
conditional marketing authorization (CMA) for use among individuals 16 years of age and 
older throughout all of the European Union’s (EU) 27 member states.7
As required b y the EUA, post-authorization observational studies using real -world data are 
needed in order to assess the association between Pfizer -BioNTech COVID- 19 vaccine and 
pre-determined safet y events of intere st (including deaths , hospitalizations, and severe 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073217
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 9of 144COVID-19) among individuals administered the vaccine in both the population at large and 
in populations of interest ( e.g., immunocompromised individuals, elderl y, and those with 
specific comorbidities) .4Pfizer in collaboration with the US Veterans Health Administration 
(VHA) and Anal ysis Groupherein propose post -EUA active safet y surveillance of safety 
events of interest based on the Priority List of Adverse Events of Special Interest from the 
Brighton Collaboration’s Safety  Platform for Emergency  vACcines (SPEAC) Project , the 
FDA and the Centers for Disease Control and Prevention’s (CDC) Advisory Committee on 
Immunization Practices (ACI P)enhanced safet y monitoring recommendation. This safety  
surveillance study willidentify and evaluate rapid,near real-time potential safet y signals 
associated with the Pfizer -BioNTech COVID-19 vaccine in the large -scale VHA electronic 
medical record (EMR) database. The observed safety  event of interest rates will be compa red 
to expected rates derived from self-controls and active comparators receiving seasonal 
influenza vaccination. Part of the methodologies used in this study  are constructed based on 
approaches previousl y used by the Post -Licensure Rapid I mmunization Safe ty Monitoring 
(PRISM) program for the H1N1 vaccine.8This non- interventional study  is designated as a 
Post-Authorization Safety Study  (PASS) commitment to the US FDA and is a Category  3 
commitment in the EU Risk Management Plan.
Research question and objectives : 
Research question: what arethe incidence rates of safet y events of interest (based on adverse 
events of special interest [AESI ])among individuals vaccinated with the Pfizer-BioNTech 
COVID-19 vaccine within the US VHA system overall and in sub -cohorts of interest, as 
compared to expected rates of those events?
Primary study objectives:
To assess whether individuals in the VHA s ystem experience increased risk of safet y 
events of interest following receipt of the Pfiz er-BioNTech COVID -19 vaccine;
To assess whether sub- cohorts of interest (i.e., immunocompromised, elderly , 
individuals with specific comorbidities, individuals receiving onl y one dose of the 
Pfizer-BioNTech COVID
-19 vaccine, and individuals with prior SARS -CoV-2 
infection) in the VHA system experience increased risk of safet y events of interest 
following receipt of the Pfizer-BioNTech COVID- 19 vaccine. 
Secondary study objective:
To characterize utilization patterns of thePfizer -BioNTech COVID -19 vaccine 
among individuals within the VHA , including estimating the proportion of 
individuals receiving vaccine, 2-dose vaccine completion rate,anddistribution of 
time gaps between the first and second dose , demographics and health histories of 
recipients, overall and among the sub- cohorts of interest .
Study design: This post-EUAactive safet y surveillance program will employ  a rapid-cycle, 
longitudinal, observational cohort study  designto provide early  real-world safet y 
information . 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073218
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 10of 144The self-controlled risk interval (SCRI) design will be used to sequentially  monitor 
occurrence of safety events of interest while controlling for time- invariant
confounders. The SCRI  design uses data from cases ( i.e., individuals who experience 
safety events of interest following vaccination) to compare the risk interval following 
vaccination to pre -or post-vaccination non -risk intervals (“pre -vaccination control 
interval” and “post -vaccination control interval”) in the same individual.
An active comparator design will be used t o sequentially  monitor occurrence of safety 
eventsof interest with Pfizer -BioNTech COVID -19 vaccinations as compared to 
recipients of influenza vaccinein the VHA during 2014/2015 through 2018/2019 flu 
seasons. Data in peri -COVID time periods from January 2020 to present are excluded 
because of pandemic- associated under -utilization of health resources and under -
reporting of medical events. 
Population : The exposed population will be kept as broad as possible in order to capture 
safety events of interest that occur among all individuals receiving the Pfizer -BioNTech 
COVID-19 vaccine in the period from December 11, 2020 to present . Individuals will be 
included if they  have a record of at least one dose of Pfizer -BioNTech COVID-19 vaccine. 
Individuals who rec eive at least one dose of COVID -19 vaccine from a manufacturer other 
than Pfizer -BioNTech will be identified and reported, but they  will be excluded from further 
analysis. All individuals will be required to be enrolled in and not disenrolled from VHA 
benefits during the 1 y ear prior to vaccination date ( i.e., baseline period). Depending on the 
attrition rate, the length of the baseline period may be modified to 6 months.
The influenza vaccine comparator cohort will be identified based on a record of at lea st one 
dose of seasonal influenza vaccine during prior flu seasons, from 2014/2015 through 
2018/2019.
Variables :
Exposure s: Administration of Pfizer-BioNTech COVID- 19 vaccine post -EUA 
approval will be identified based on the following: 
oCurrent Procedural Terminology  (CPT) code 91300 (Severe acute respiratory  
syndrome coronavirus 2 (SARS -CoV-2) (coronavirus disease [COVID- 19]) 
vaccine, mRNALNP, spike protein, preservative free, 30 mcg/0.3mL dosage, 
diluent reconstituted, for intramuscular use) and associate d vaccine 
administration HCPCS codes corresponding to the first dose: 0001A (ADM 
SARS-CoV-2 30 mcg/0.3mL  1st), and the second dose: 0002A (ADM SARS -
CoV-2 30 mcg/0.3mL 2nd);9,10OR
o10 and 11-dig it National Drug Codes (NDCs) 59267-1000 -1 (corresponds to 
first dose), 59267 -1000-01 (corresponds to second dose);9OR
oImmunization records that contain data on vaccine code descriptor, vaccine 
manufacturer ( i.e., Pfizer), lot number, injection site, and date(s) of 
immunization;9
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073219
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 11of 144Relevant codes will be continuously  reviewed and amended if new codes are added.
Administration of the seasonal influenza vaccine during 2014/2015 through
2018/2019 flu seasons will be identified based on the following:
oCPT codes
90654 (Influenza virus vaccine, trivalent (IIV3), split virus,
preservative -free, for intradermal use); OR
90656 (Influenza virus vaccine, trivalent (IIV3), split virus,
preservative free, 0.5 mL dosage, for intramuscular use); OR
90658 (Influenza virus vaccine, trivalent (IIV3), split virus, 0.5 mL
dosage, for intramuscular use); OR
o10 and 11- digit NDCs; OR
oImmunization records that contain data on vaccine code descriptor, vaccine
manufacturer, lot number, injection site, and date(s) of immunization.
Outcomes : Safety events of interest for active surveillance (see Table 1and Appendix
Table 2) are based on the Priority  List of Adverse Events of Special Interest from the
Brighton Collaboration’s Safety  Platform for Emergency  vACcines (SPEAC) Project ,
the FDA and the Centers for Dise ase Control and Prevention’s (CDC) Advisory
Committee on I mmunization Practices (ACIP) enhanced safet y monitoring
recommendations.
The list of safety events of interest may be revised over the course of the study, and if 
unanticipated po tential safety events of interest are identified during the course of 
surveillance, they will be added to the list and included in the anal yses. The risk and 
control intervals for  each safety event of interest are based on biological plausibility 
and precedents in the litera ture (see Table 1). Outpatient (including emergency 
department [ED]) and/or inpatient settings will be used to identify safety  events of 
interest depending on the ty pe of event. The specific encounter setting to be 
considered for each safety event of interest is summarized in Table 1 
and can be 
assigned to 1) the risk interval following vaccination Pfizer-BioNTech COVID-19 
vaccination, 2) the pre-v accination self-c ontrol interval, 3) the post-vaccination self-
control interval, or  4) risk interval for the active comparators of receiving seasonal 
influenza vaccine. Events outside the intervals will not b
e counted.
Only the individual’s first instance of asafety eventof interest following a specified
clean window ( i.e., the occurrence -free baseline period used to define incide nt
outcomes during which individuals enter the study  cohort only  if the safety  event of
interest did not occur during this period) will be captured; this means that if a safety
event of interest is identified but diagnosis codes corresponding to the safety event of
interest are also observed during the clean window, it will not be counted . The
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073220
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 12of 144duration of the pre-specified clean window will differ by  safety event of interest (see 
Appendix Table 2) in order to rule out pre -existing events.
Key Covariates: Baseline demographic ( i.e., age, sex, race/ethnicit y, state) and 
clinical characteristics ( i.e., smoking, body  mass index [ BMI],history of 
anaphylaxis/allergic reactions, previous anaphy laxis to vaccine component, history  of 
hospitalizations, Charlson Comorbidity  Index [CCI ], selected comorbidities, and 
concurrent immunizations)11will be assessed based on available data ( i.e., during 1-
year baseline ) prior to the date of vaccination with Pfizer -BioNTech COVID -19 
vaccine and date of seasonal influenza vaccination for active comparators .
Subgroups : Immunocompromised individuals, elderly , individuals with specific 
comorbidities, those receiving only  one dose of Pfizer -BioNTech COVID -19 vaccine, 
thosewith prior SARS -CoV-2 infection, those with regular use of VHA medical care, 
and VA priorit y group 1 veterans will be identified .
Data source :The VHA is the largest integrated health care s ystem in the US, providing both 
inpatient and outpatient clinical care to over 9 million Veterans enrolled at more than 170 
medical centers and 1,074 community -based outpatient clinics.12This studywill use data 
from VHA’s Corporate Data Warehouse (CDW), which is an integrated electronic medical 
record (EMR) s ystem with a centralized data warehouse that is updated on a daily  basis.The 
CDWdoes not include information on an y care received outside of a VHA facility .The VA 
offers eligible Veterans long -term care services ranging from nursing homes and assisted -
living centers to caregiver support in the Veterans’ own homes.13
Study size: The sample size achieved will depend on the number of recipients of Pfizer -
BioNTech COVID -19 vaccine within the VHA database, which will increase over time with 
subsequent anal yses. As of January  21, 2021, 112,201 doses of Pfi zer-BioNTech COVID -19 
vaccine have been administered within the VHA (based on CPT code 91300) to a total of 
107,458 patients.
Data analy sis: A stepwise approach, illustrated in the diagram , will be performed for signal 
detection, evaluation, and verification.   
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073221
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 13of 144
Notes: 
[1] List of safety events of interest and corresponding definitions may be refined as the study progresses based on additiona l 
available information.
[2] The risk and cont rol intervals selected for the SCRI analysis for each safety event of interest are based on biological 
plausibility and precedents in the literature. Only the individual’s first instance during the specified clean window ( i.e., the 
interval used to define incident outcomes) will be included.  Note that only the first inpatient or outpatient occurrence of a 
safety event of interest following the clean window will be used to identify incident events ( e.g., if an inpatient safety event 
of interest occurs in the clean window, a repeat occurrence will not be counted in the risk interval). However, event 
worsening will be counted as a safety event of interest. For example, if an outpatient safety event of inte rest occurs in the 
clean window and an inpatient occurrence for the same type of safety event of interest occurs in the risk interval, the 
inpatient occurrence will be counted as a safety event of interest. 
1) Signal detection : The goal is to provide rapid- cycle, near real -time safety  surveillance. In 
the signal detection phase, the SCRI  analysis will only  include pre -vaccination control 
intervals as the post -vaccination control intervals will require a longer time to accumulate 
andwill be used in the sig nal evaluation phase. To account for multiple testing and bi-weekl y 
review of the data, the maximized sequential probability  ratio test (MaxSPRT )using a 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073222
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 14of 144binomial probability  model will be applied. For comparison with individuals who received 
seasonal infl uenza vaccination, the Poisson -based MaxSPRT will be applied. 
Sequential anal yses for each safet y event of interest will commence once at least 3 events 
occur. This approach is consistent with the FDA’s COVID -19 Vaccine Safety  Surveillance 
Projectto avoid spurious signals from a few early  events.14Signals will be detected if the 
critical values are reached via the SCRI or active comparator anal ysis.Critical values will be 
determined for each safety event of interest base d on historical incidence rate, expected upper 
limit of the number of events under the null hy pothesis, and pre -specified significance level 
and power. Incidence rates will also be calculated and Kaplan -Meier methods wi llbe used to 
analyze time to safety event of interest .
2) Signal evaluation: If signals are detected for safety  events of interest based on the anal ysis 
described above, further evaluation will be conducted to refine and confirm such detections.  
This will include comprehensive quality  assurance (for example, check for possible 
duplications of claims or medical records, checking for unusual clustering in claim or 
medical record accrual by service date for potential coding issues, check for geographical 
distribution of cases that may  be related to lot numbers or diagnostic practice ) and 
multivariate adjustment using Poisson regression to account for baseline differences between 
Pfizer-BioNTech COVID- 19 vaccinated and active comparator cohorts. SCRI  analyses using 
the post-vaccination control intervals will be conducted as an additional inferential analy sis 
once enough post -vaccination time has accumulated. Lastly, the assessment of temporal 
clustering will also be conducted. 
3) Signal verification: diagnostic validation of the detected safety eventsof interest via 
adjudication of medical records by VHA clinicians for outcome verification will be 
conducted in a representative sample of cases. For rare events, potentiall y all cases may  be 
adjudicated.
End-of-season anal yses (over the course of the 30 -month period) and an end-of- surveillance 
analysis (i.e., at 30 months) will be conducted. Various subgroup anal yses will also be 
conducted, examining different age groups, immunocompromised individuals, individuals 
with specific comorbidities, those who only  received one dose of the Pfizer -BioNTech 
COVID-19 vaccine, those with prior SARS -CoV-2 infection based on medical history  or pre-
vaccination serology , those receiving care regularly  at VA facilities, and lastly  those with VA 
Priority group 1 statu s, which determines these individuals are of highest priority  for VHA 
care and likel y receive all of their care within the VHA s ystem.
Milestones:
Registration in the EU PAS register: To be registered before the start of data 
collection;
VHA Cooperative R esearch and Data Agreement ( CRADA) and Institutional Review 
Board (IRB) approvals (estimated) : March-April 2021 ;
Start of data collection( estimated planned date for starting data extraction for 
analysis): May2021;
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073223
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 15of 144Interim reports: 30 June 2021; 31 December 2021; 30June 2022, 31 December 2022;
End of data collection ( estimated planned date for final data cut): 10 June2023;
Final study  report: 31 December 2023
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073224
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 16of 144SUMMARY
Objective Primary 1 Primary 2 Secondary
Aim To assess whether individuals in the 
Veterans Health Administration 
(VHA) system experience increased 
risk of safety events of interest 
following receipt of the Pfizer -
BioNTech COVID -19 vaccine. To assess whether sub -cohorts of 
interest (i.e.immunocompromised, 
elderly, with specific comorbidities, 
individuals receiving only one dose 
of the Pfizer -BioNTech COVID -19 
vaccine, and individuals with prior 
SARS-CoV-2 infection ) in the VHA 
system experience increased risk of 
safety events of interest following 
receipt of the Pfizer -BioNTech 
COVID-19 vaccine. To characterize utilization patte rns of
thePfizer-BioNTech COVID -19 
vaccine among individuals within the 
VHA including estimating the 
proportion of individuals receiving 
vaccine, 2 -dose vaccine completion 
rate, and distribution of time gaps 
between the first and second dose, 
demographics and health histories of 
recipients, overall and among the 
sub-cohorts of interest.
Study design This post-EUA active safety surveillance program will employ a rapid -cycle, longitudinal, observational cohort study 
design to provide early real -world safety information.
The self-controlled risk interval (SCRI) design to sequentially monitor occurrence of safety events of interest 
while controlling for time -invariant confounders. This design allows inclusion of either a pre -vaccination 
control interval or a p ost-vaccination control interval, depending on the safety event of interest (e.g., post-
vaccination control intervals are used for outcomes where there is concern for bias due to indication or 
contraindication); 
An active comparator design will be used to sequentially monitor occurrence of safety events of interest with 
Pfizer-BioNTech COVID -19 vaccinations as compared to recipients of influenza vaccine in the VHA during 
2014/2015 through 2018/2019 flu seasons. Data in peri- COVID time periods from January 2020 to present are 
excluded because of pandemic -associated under -utilization of health resources and under -reporting of medical 
events.
Study population The study will be kept as broad as possible in order to capture safety events of interest that occur among vaccinated 
individuals.
Inclusion criteria:
Record of at least one dose of Pfizer -BioNTech COVID -19 vaccine in the period of December 11, 2020 to 
present, or
Record of at least one dose of seasonal influenza vaccine during prior flu seasons, from 2 014/2015 through 
2018/2019 ( applies to active comparator s only); and
At least 1 year of enrollment in and no disenrollment from VHA benefits ( i.e., the baseline period) prior to 
Pfizer-BioNTech COVID -19 or seasonal influenza vaccination date. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073225
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 17of 144Exclusion criteria:
Individuals who receive at least one dose of Pfizer -BioNTech COVID -19 vaccine in addition to a COVID -19 
vaccine from a manufacturer other than Pfizer- BioNTech will be identified and reported, but they will be 
excluded from further analysis. 
Study Period The study will be conducted for a period of 30 months, starting on December 11, 2020 onward, with data collection 
concluding on June 10, 2023.
Exposure Administration of Pfizer -BioNTech COVID -19 vaccine post -EUAapproval will be identified based on records of the 
following:
Current Procedural Terminology (CPT) code 91300 (Severe acute respiratory syndrome coronavirus 2 
(SARS-CoV-2) (coronavirus disease [COVID- 19]) vaccine, mRNALNP, spike protein, preservative fre e, 30 
mcg/0.3mL dosage, diluent reconstituted, for intramuscular use) and associated vaccine administration 
HCPCS codes corresponding to the first dose: 0001A (ADM SARS -CoV-2 30 mcg/0.3mL 1st), and the second 
dose: 0002A (ADM SARS -CoV-2 30 mcg/0.3mL 2nd); OR
10 and 11 -digit National Drug Codes (NDCs) 59267 -1000-1 (corresponds to first dose), 59267 -1000-01 
(corresponds to second dose); OR
Immunization records that contain data on vaccine code descriptor, vaccine manufacturer ( i.e., Pfizer), lot 
number, injec tion site, and date(s) of immunization; 
Administration of the seasonal influenza vaccine during 2014/2015 through 2018/2019 flu seasons will be identified 
based on records of the following:
CPT codes 
o90654 (Influenza virus vaccine, trivalent (IIV3), spli t virus, preservative -free, for intradermal use); 
OR
o90656 (Influenza virus vaccine, trivalent (IIV3), split virus, preservative free, 0.5 mL dosage, for 
intramuscular use); OR
o90658 (Influenza virus vaccine, trivalent (IIV3), split virus, 0.5 mL dosage, f or intramuscular use); 
OR
10 and 11 -digit NDCs; OR
Immunization records that contain data on vaccine code descriptor, vaccine manufacturer, lot number, 
injection site, and date(s) of immunization. 
Safety Events of Interest Safety events of interest for active surveillance were identified based on the Priority List of Adverse Events of Special 
Interest from the Brighton Collaboration’s Safety Platform for Emergency vACcines (SPEAC) Project, the FDA and 
the Centers for Disease Control and Prevention’s (CDC) Advisory Committee on Immunization Practices (ACIP) 
enhanced safety monitoring recommendations. The list of safety events may be revised over the course of the study, 
and if unanticipated potential safety events of interest are identified during the course of surveillance, they will be 
added to the list and included in the analyses. The risk and control intervals for each safety event of interest are based 
on biological plausibility and precedents in the literature. Outpatient (including emergency dep artment) and/or 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073226
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 18of 144inpatient settings will be used to identify safety events of interest depending on the type of event. Safety events of 
interest can be assigned to 1) the risk interval following vaccination Pfizer -BioNTech COVID -19 vaccination, 2) the 
pre-vaccination self -control interval, 3) the post -vaccination self -control interval, or 4) risk interval for the active 
comparators of receiving seasonal influenza vaccine. Events outside the intervals will not be counted. Only the 
individual’s first instance of a safety event of interest following a specified clean window (i.e. , the occurrence -free 
baseline period used to define incident outcomes during which individuals enter the study cohort only if the safety 
event of interest did not occur during this peri od) will be included; this means that if a safety event is identified but 
diagnosis codes corresponding to the safety event are also observed during the clean window, it will not be counted . 
The duration of the pre -specified clean w indow will differ by typ e of safety event of interest in order to rule out pre -
existing events. 
Neurologic
Generalized convulsions/seizures;
Guillain-Barré syndrome (GBS);
Aseptic meningitis;
Encephalitis/encephalomyelitis;
Other acute demyelinating diseases;
Transverse myelitis (TM);
Multiple sclerosis (MS);
Optic neuritis (ON);
Bell’s palsy
Immunologic
Anaphylaxis;
Vasculitides;
Arthritis and arthralgia/joint pain;
Multisystem inflammatory syndrome in adults (MIS -A);
Kawasaki disease (KD);
Fibromyalgia;
Autoimmune thyroiditis
Cardiac
Myocarditis;
Pericarditis;
Acute myocardial infarction (AMI)
Hematologic
Thrombocytopenia;
Disseminated intravascular coagulation (DIC)
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073227
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 19of 144COVID-19 (for all COVID- 19-related safety events of interest listed below, an inpatient diagnosis of COVID -
19 will be required in combination with the codes or laboratory values specified in Appendix Table 2; in 
addition, COVID- 19 related safety events of interest will only be evaluated using data from 2020 onw ard 
using the SCRI design ):
Severe COVID- 19 disease;
Microangiopathy;
Heart failure and cardiogenic shock;
Stress cardiomyopathy;
Coronary artery disease (CAD);
Arrhythmia;
Deep vein thrombosis (DVT);
Pulmonary embolus;
Cerebrovascular hemorrhagic stroke;
Cerebrovascular non -hemorrhagic stroke;
Limb ischemia;
Hemorrhagic disease;
Acute kidney injury;
Liver injury;
Chilblain-like lesions;
Single organ cutaneous vasculitis;
Erythema multiforme
Other
Death;
Narcolepsy/cataplexy;
Non-anaphylactic allergic reactions;
Appendicitis
Data source The VHA Corporate Data Warehouse (CDW) database w ill be used.
Data analysis A stepwise approach w ill be performed for signal detection, evaluation, and verification. 
1) Signal detection: The goal is to provide rapid -cycle, near real -time safety surveillance. In the signal detection phase, 
the SCRI analysis will only include pre -vaccination control intervals as the post -vaccination control intervals will 
require a longer time to accumulate and will be used in the signal evaluation phase. To account for multiple testing and 
bi-weekly review of the data, the maximized sequential probability ratio test (MaxSPRT) using a binomial probability 
model will be applied. For compar ison with individuals who received seasonal influenza vaccination, the Poisson -
based MaxSPRT w ill be applied. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073228
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 20of 144Sequential analyses for each safety event of interest will commence once at least 3 events occur. This approach is 
consistent with the FDA’s COVI D-19 Vaccine Safety Surveillance Project to avoid spurious signals from a few early 
events.  Signals will be detected if the critical values are reached via the SCRI or active comparator analysis. Critical 
values will be determined for each safety event of interest based on historical incidence rate, expected upper limit of 
the number of events under the null hypothesis, and pre -specified significance level and pow er. Incidence rates will 
also be calculated and Kaplan -Meier methods will be used to analyze t ime to safety event of interest.
2) Signal evaluation: If signals are detected for safety events of interest based on the analysis described above, further 
evaluation will be conducted to refine and confirm such detections. This will include comprehensive quality assurance 
(for example, check for possible duplications of claims or medical records, checking for unusual clustering in claim or 
medical record accrual by service date for potential coding issues, check for geographical distribution of cases that may 
be related to lot numbers or diagnostic practice) and multivariate adjustment using Poisson regression to account for 
baseline differences between Pfizer -BioNTech COVID -19 vaccinated and active comparator cohorts. SCRI analyses 
using the post -vaccination control intervals will be conducted as an additional inferential analysis once enough post -
vaccination time has accumulated. Lastly, the assessment of temporal clustering will also be conducted. 
3) Signal verification: diagnostic validation of the detected safety events of interest via adjudication of medical records 
by VHA clinicians for outcome validation will be conducted in a representative sample of cases. For rare events, 
potentially all cas es may be adjudicated.
End-of-season analyses (over the course of the 30 -month period) and an end -of-surveillance analysis ( i.e., at 30 
months) will be conducted. Various subgroup analyses will also be conducted, examining different age groups, 
immunocompr omised individuals, individuals with specific comorbidities, those who only received one dose of the 
Pfizer-BioNTech COVID -19 vaccine, those with prior SARS -CoV-2 infection based on medical history or pre -
vaccination serology, those receiving care regularl y at VA facilities, and lastly those with VA Priority group 1 status, 
which determines these individuals are of highest priority for VHA care and likely receive all of their care within the 
VHA system.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073229
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 21of 1445.AMENDMENTS AND UPDAT ES
None.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073230
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 22of 1446.MILESTONES
Milestone Planned date
Registration in the EU PAS register To be registered before the start of 
data collection 
VHA CRADA and IRB approvals (estimated) March -April 2021
Start of data collection (estimated) May2021[1]
Interim reports 30June 2021
31December 2021 
30June 2022 
31 December 2022 
End of data collection (estimated) 10June 2023[2]
Final study  report 31December 2023 
Abbreviations : CRADA , Cooperative Research and Data Agreement; IRB, Institutional Review Board; 
VHA, Veterans Health Administration. 
Notes:
[1] Start of data collection is the planned date for starting data extraction for the purposes of the study 
analysis. The initial data analysis will include the Pfizer -BioNTech COVID -19 vaccine exposure since 
December 11, 2020, the EUA approval date by the US FDA.
[2] End of data collection is the planned date on which the Pfizer -BioNTech COVID -19 vaccine exposure 
reached 30 months post -EUA approval.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073231
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 23of 1447.RATIONALE AND BACKGR OUND
In March 2020, the World Health Organization (WHO) declared a global pandemic for the 
coronavirus disease 2019 (COVID -19) due to the severe acute respiratory  syndrome 
coronavirus 2 (SARS -CoV-2), which was first ide ntified by  public health officials in China 
in December 2019.1The COVID -19 pandemic presents an unprecedented public health 
crisis. As of January 7, 2021, over 21.4million COVID -19 cases and 364,000 deaths have 
been reported in the United States (US) alone.2To date, the incidence of COVID -19 has 
continued to rise, largel y affecting the elderl y and middle -aged individuals, with worsening 
clinical sequelae linked to increasing age and comorbid conditions ( e.g., cardiovascular 
disease, active cancer, obesity , diabetes and chronic lung disease).15,16SARS-CoV-2is a 
well-adapted highl y infectious human pathogen with a case fatality rate that ranges between 
0.5% and 20%, based on the individual’s age, gender, race, and comorbidites.17
Pfizer and BioNTech have partnered to developa novel messenger RiboNucleic Acid 
(mRNA)vaccine against SARS -CoV-2for the prevention of COVID -19 (Candidate 
BNT162b2). To this end, Pfizer is conducting a Phase 1/2/3, randomized, placebo -controlled, 
observer-blind, dose -finding, vaccine candidate- selection, and efficacy  study among healthy  
individuals (NCT04368728). In their Phase 1 trial evaluating safety  and immunogenicit y of 
two mRNA vaccine candidates ( i.e., BNT162b1, BNT162b2) at various dose levels, 
candidate BNT162b2 was selected for advancement to a pivotal Phase 2/3 safet y and efficacy 
evaluation due to its milder systemic reactogenicity  profile, especially  in older adults.18The 
study was initiated in July  2020 with a target enrollment of 43,998 individuals.19
The US Food and Drug Administration (FDA) announced that regulatory  emergency  use 
authorization (EUA) as well as full approval of any  COVID-19 vaccine will require 
demonstrating prevention of the disease or decrease in its severity  in at least 50% of the 
individuals who receive it. I n addition, data from Phase 3 studies are required to include a 
median follow -up duration of at least 2 months after completion of the full vaccination 
regimen to assess the vaccine’s benefit -risk profile, especiall y adverse events and cases of 
severe COVID -19 in vaccinated study  subjects.20,21The FDA reviewed the available safet y 
data of the Phase 1/2/3 trial from 37,586 partici pants 16 y ears of age and older and did not 
identify any specific safety  concerns. In addition, the anal ysis of available efficacy  data from 
36,523 participants 12 years of age and older without evidence of prior SARS- CoV-2 
infection at least 7 day s after receiving the second dose demonstrated  95% efficacy  of the 
vaccine in the prevention of COVID -19 (as confirmed by  8 vs. 162 COVID -19 cases in the 
vaccine and placebo groups, respectivel y).3,4Based on these safety  and efficacy  data, as well 
as a review of manufacturing information regarding product quality  and consistency , the 
FDA determined that the known and potential benefits of the vaccine outweighed the known 
and potential risks for the pre vention of COVID -19 in individuals 16 y ears of age and older.4
Therefore on December 11, 2020, the Pfizer -BioNTech COVID-19 vaccine was granted an 
Emergency  Use Authorization (EUA) by  the FDA to prevent COVID -19 in individuals 16 
years of age and older.5
With respect to geographic regions other than the US, on December 2, 2020, the United 
Kingdom (UK) was the first country  in the world to grant temporary  authorization for 
emergency  use of the Pfizer -BioNTech COVID -19 vaccine.6On December 21, 2020, the 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073232
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 24of 144European Medicines Agency  (EMA) granted the Pfizer -BioNTech COVID- 19 vaccine a 
conditional marketing authorization (CMA) for use among individuals 16 years of age and 
older throughout all of the European Union’s (EU) 27 member states.7
As required b y the EUA, post-authorization observational studies using real -world data are 
needed in order to assess the association between Pfizer -BioNTech COVID- 19 vaccine a nd 
pre-determined safet y events of interest (including deaths , hospitalizations, and severe 
COVID-19) among individuals administered the vaccine in both the population at large and 
in populations of interest ( e.g., immunocompromised individuals, elderl y, and those with 
specific comorbidities) .4Post-authorization safety evaluations are important for identify ing 
rare, serious safet y events of interest in larger populations that may  not have been detected 
during clinical trials (either due to sample size or s elected stud y populations), and ensure a 
favorable benefit -risk ratio post -trial.Pfizer in collaboration with the US Veterans Health 
Administration ( VHA)and Analy sis Group herein propose post -EUA active safet y 
surveillance of safet y events of interest ba sed on the Priority List of Adverse Events of 
Special Interest from the Brighton Collaboration’s Safety  Platform for Emergency  vACcines 
(SPEAC) Project , the FDA andthe Centers for Disease Control and Prevention’s (CDC) 
Advisory Committee on Immunization P ractices (ACI P)enhanced safet y monitoring 
recommendation . This safety surveillance study  willidentify and evaluate rapid, near real-
time potential safety  signals associated with the Pfizer -BioNTech COVID -19 vaccine in the 
large-scale VHA electronic medical record ( EMR)database. The observed rates of safet y 
event of interest  will be compared to expected rates derived from self -controls and active 
comparators. Part of the methodologies used in this study  are constructed based on 
approachespreviouslyused by the Post-Licensure Rapid I mmunization Safety  Monitoring 
(PRISM) program for the H1N1 vaccine.8This non- interventional study  is designated as a 
Post-Authorization Safety Study  (PASS) commitment to the US FDA andisa Category  3 
commitment in the EU Risk Management Plan .
8.RESEARCH QUESTION AND OBJECTIVES
Research question: what are the incidence rates of safety events of interest (based on adverse 
events of special interest [AESI ]) among individuals vaccinated with the Pfizer-BioNTech 
COVID-19 vaccine within the US VHA system overall and in sub -cohorts of interest a s 
compared to expected rates of those events?
Primary study objectives:
To assess whether individuals in the VHA system experience increased risk of safet y 
events of interest following receipt of the Pfizer- BioNTech COVID -19 vaccine;
To assess whether sub-cohorts of interest ( i.e., immunocompromised, elderly , with 
specific comorbidities, individuals receiving onl y one dose of the Pfizer -BioNTech 
COVID-19 vaccine, and individuals with prior SARS -CoV-2 infection) in the VHA 
system experience increased risk of safety
 events of interest following receipt of the 
Pfizer-BioNTech COVID- 19 vaccine. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073233
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 25of 144Secondary study objectiv es:
To characterize utilization patterns of the Pfizer -BioNTech COVID -19 vaccine 
among individuals within the VHA including estimating the proportion of individuals
receiving vaccine, 2-dose vaccine completion rate,and distribution of time gaps 
between the first and second dose , demographics and health histories of recipients, 
overall and among the sub-cohorts of interest. 
9.RESEARCH METHODS
9.1.Study Design 
This post-EUAactive safety  surveillance program will employ  a rapid-cycle, longitudinal, 
observational cohort study  designto provide earl y real-world safet y information . The self-
controlled risk interval (SCRI) design will be used to sequentially  monitor occurrence of 
safety events of interest while controlling for time- invariant confounders (such as sex, race, 
chronic illness, and state). I n addition, safety events of interest associated with Pfizer-
BioNTech COVID -19 vaccinations will be sequentially  monitored and compared to 
recipients of influenza vaccine in the VHA between 2014/2015 to 2018/2019 .8,22
9.1.1.Self-Controlled Risk Interval (SCRI) Design 
The SCRI  design uses data from cases ( i.e., individuals who experience safety events of 
interest following vaccination) to compare the risk interval following vaccination to pre -or
post-vaccination non -risk intervals (“pre -vaccination control interval” and “post- vaccination 
control interval”) in the same individual.23Whether a pre -or post-vaccination control 
interval is used will depend on the clinical nature, seasonality , and frequency  of the safety 
event of interest , as described in greater detail below. A length of 42 day s has been used to 
define the risk interval in SCRI design studies for signal detection to ascertain the safet y 
profile of the H1N1 vaccine.8,22The same length of risk interval is proposed here, subject to 
further modification based on clinical input, clinical trial data, biologic plausibility , and 
published literature. The day  of vaccination will only  be included in the risk period for those 
safety eventsof interest for which a same -day occurrence is biologically  plausible 
(e.g.,anaphylaxis).
As some individuals may choose to decline or delay Pfizer- BioNTech COVID -19 
vaccination soon after an illness (known as the “healthy vaccinee effect”),24the 
pre-vaccination control interval will exclude the 14- day period before vaccination.25While 
using a pre -vaccination control period allows for timely  analysis, especially  pertinent for 
rarer safety events of interest , a post-vaccination control interval would be more appropri ate 
and will be used for certain safet y events of interest  for the following reasons: (1) a recent 
prior safet y event of interest might preclude vaccination ( i.e.,anaphylaxis), (2) individuals 
might have an underl ying condition that is also a contraindication for vaccination ( i.e., 
seizure disorder), or (3) safet y event of interest and vaccination may  be seasonal in nature.26
The time between the risk and control intervals will be determined based on the biological 
mechanism of action for each safet y eventsof interest assessed , and may  be subject to change 
based on further clinical input. Examples of the SCRI  design with a pre -vaccination control 
interval and a post -vaccination control interval (in an individual who only receives the first 
dose of vaccine) is presented in Figure 1below.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073234
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 26of 144Figure 1. Example of SCRI Design for Assessment of a Safety Event of Interest with a 
42-day Risk Interval in an Individual who Receives Only One Vaccine Dose, Showing 
Both Pre -and Post-vaccination Control Intervals
*The risk interval may include day 0, date of Pfizer -BioNTech COVID -19 vaccination, for some of the safety
eventsof interest assessed ( e.g., anaphylaxis). The length of the risk interval will vary across each safety event 
of interest and may be subject to change based on clinical input. Note that some individuals may not receive the 
complete course of vaccination, and thus may only receive the first dose of vaccine. This is represented in 
Figure 1while Figure 2represents an example where the complete course with 2 doses are received.
Two doses of the Pfizer -BioNTech COVID -19 vaccine are recommended 3 weeks apart.  
This studyprogram will monitor safety  eventsof interest that occur after dose 1 before dose 
2 (i.e., during risk interval 1), after dose 2 ( i.e., during risk interval 2), and aggregate for 
doses 1 and 2 ( i.e., risk interval 1 + risk interval 2), respectively , for individuals receiving 
both doses. 
For individuals who receive two doses of the vaccine, two separate control intervals will be 
defined to correspond to the risk interval associated with each dose (regardless of whether 
pre
- or post-vaccination control intervals are used). See Figure 2 below fo r an example in an 
individual who receives tw o doses of Pfizer-B ioNTech COVID-19 vaccine, with the second 
dose received 21 days after the
 first. Safety events of in terest that occur during the 
overlapping period of risk interval 1 and risk interval 2 (shown in gray shading in Figure 
2) 
may be flagged for separate analy ses to discern th
e additive effect of Pfizer-B ioNTech 
COVID-19 vaccine dose 1 and dose 2.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073235
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 27of 144Figure 2. Example of SCRI Design with Overlapping Risk Intervals when Two Doses of 
Pfizer-BioNTech COVID -19 Vaccine are Administered, Showing a Pre- and 
Post-vaccination Control Interval
9.1.2.Active Comparator Design
In the active comparato rdesign, the frequency  of safety events of interest among individuals 
who received Pfizer -BioNTech COVID -19 vaccine from December 11, 2020 onwardwill be 
compared with the event frequency  among recipients of the seasonal influenza vaccination in 
five prior seasons, between 2014/2015 t hrough 2018/2019. Data in peri -COVID time periods 
from January  2020 to present are excluded because of pandemic- associated under -utilization 
of health resources and under- reporting of medical events. The same risk interval len gth 
(e.g.,42 days)will be used to evaluate safety eventsof interest following vaccination with 
Pfizer-BioNTech COVID- 19 vaccine and to assess safet y eventsof interest occurring after 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073236
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 28of 144vaccination for seasonal influenza in prior seasons. The observed num ber of safety eventsof 
interest for Pfizer -BioNTech COVID -19 vaccine will be compared to the expected number 
calculated for the influenza vaccine in past seasons.8
9.1.3.Study Period 
The studywill be conducted for a period of 30 months , starting on December 11, 2020 
onward, with data collection concluding on June 10, 2023. 
9.2. Setting 
The study population will be kept as broad as possible in order to capture safety eventsof 
interest that occur among all vaccinated individuals.
9.2.1. Inclusion Criteria
Record of at l east one dose of Pfizer -BioNTech COVID- 19 vaccine in the period of 
December 11, 2020 to present, or
Record of at least one dose of seasonal influenza vaccine during prior flu seasons, 
from 2014/2015 to 2018/2019 (applies to active comparator s only); and
Atleast 1 year of enrollment in and no disenrollment from VHA benefits ( i.e., the 
baseline period) prior to Pfizer -BioNTech COVID- 19 or seasonal influenza 
vaccination date. 
9.2.2.Exclusion criteria
Individuals who receive at least one dose of Pfizer- BioNTech COV ID-19 vaccine in 
addition to a COVID -19 vaccine from a manufacturer other than Pfizer -BioNTech 
will be identified and reported, but they  will be excluded from further analy sis.
9.2.3.Subgroups
Safety surveillance may  be conducted for subgroups of interest, incl uding, but not limited to:
Immunocompr omised individuals; 
Different age groups, with a focus on the elderl y(e.g., <35, 35 - <45, 45 - <55, 55 -
<65, 65 -<75, >75);
Individuals with specific comorbidities;
Individuals receiving only  one dose of Pfizer-BioNTech COVID-19 vaccine;
Individuals with prior SARS -CoV-2 infection based on medical history  or 
pre-vaccination serology ;
Individuals with regular use of VHA medical care, defined as at least two outpatient 
(excluding ED, as ED visits may  not be considered regular) or inpatient encounters in 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073237
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 29of 144the one year prior to vaccination .The encounters must be separated b y >30days (for 
inpatient, by  admission date), and at least one must be within six months prior to the 
date of va ccination.This will ensure that individuals have ongoing health care 
encounters, particularl y near the vaccination date, and regularly  receive their 
healthcare from VHA facilities, rather than outside facilities that would not be 
captured in the CDW;
Individualswho are in the VApriority group 1Veteran. These individuals have either 
the highest levels of service connected disability (>50% disabling), are considered 
unemploy able, or have received the medal of honor.27Individuals categorized as 
priority group 1 are the highest priorit y for VHA care. This will ensure that the 
individual is more likely  to receive all of their care from a VA facility .
Additional subgroups of interest will be assessed as additional information becomes 
available from ongoing clinical trials, Vaccine Adverse Event Reporting Sy stem (VAERS), 
and other sources that will inform the Pfizer -BioNTech COVID -19 vaccine safet y profile.
Giventhat VA population has a median age of over 46 y ears for females and is comprised of 
approximatel y 90% males , the evaluation of the Pfizer-BioNTech COVID -19 vaccine safet y 
during pregnancy , including fetal death and infant outcomes, may  have poor feasibility and 
will therefore not be conducted. 
9.3.Variables
9.3.1.Exposure of Interest
Administration of Pfizer -BioNTech COVID -19 vaccine post-EUA approval will be identified 
based on the following:
Current Procedural Terminology  (CPT) code 91300 ( Severe acute respiratory  
syndrome coronavirus 2 (SARS -CoV-2) (coronavirus disease [COVID-19]) vaccine, 
mRNALNP, spike protein, preservative free, 30 mcg/0.3mL dosage, diluent 
reconstituted, for intramuscular use ) and associated vaccine administration HCPCS 
codes corresponding to the first dose: 0001A (ADM SARS -CoV-2 30 mcg/0.3mL  
1st), and the second dose: 0002A (ADM SARS- CoV-2 30 mcg/0.3mL  2nd);9,10OR
10 and 11-digit National Drug Codes (NDCs) 59267-1000- 1 (corresponds to first 
dose), 59267-1000- 01 (corresponds to second dose); OR
Immunization records that contain data on vaccine code descriptor, vaccine 
manufacturer ( i.e., Pfizer), lot number, injection site, and date(s) of immunization.9
Relevant codes will be continuously  reviewed and amended if new codes are added.
Person-time at-risk exposure to the first dose onl y, overlapping first and second doses, and 
second dose onl y will be anal yzed separatel y.
Administration of the seasonal influenza vaccine during 2014/2015 through 2018/2019 flu 
seasons will be identified based on the following:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073238
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 30of 144CPT codes 
o90654 (Influenza virus vaccine, trivalent (IIV3), split virus, preservative -free, 
for intradermal use) ; OR
o90656 (Influenza virus vaccine, trivalent (IIV3), split virus, preservative free, 
0.5 mL dosage, for intramuscular use) ;OR
o90658 (Influenza virus vaccine, trivalent (IIV3), split virus, 0.5 mL  dosage, 
for intramuscular use );OR
10 and 11- digit NDCs; OR
Immunization records that contain data on vaccine code descriptor, vaccine 
manufacturer, lot number, injection site, and date(s) of immunization. 
9.3.1.1.Pfizer-BioNTech COVID -19 Vaccine Groups of Interest
While the primary  vaccination group of interest is all in dividuals receiving Pfizer- BioNTech 
COVID-19 vaccine (irrespective of receipt of seasonal influenza vaccination), additional 
subsets of the study population will be studied, similar to the PRI SM safety  surveillance 
program of H1N1 vaccine safet y:8
Cohort A: Individuals vaccinated with Pfizer -BioNTech COVID- 19 vaccine who did not 
receive the influenza vaccine during the flu season in which COVID -19 vaccination 
occurred;
Cohort B: Individuals vaccinated with Pfizer -BioNTech CO VID-19 vaccine who received 
the seasonal influenza vaccine at least 42 day s prior to COVID -19 vaccination during the 
same flu season in which COVID -19 vaccination occurred;
Cohort C: Individuals vaccinated with Pfizer -BioNTech COVID- 19 vaccine who received
the seasonal influenza vaccine within 42 day s before or any  time after COVID -19 
vaccination during the same flu season in which COVID- 19 vaccination occurred;
Cohort D: Individuals vaccinated with both Pfizer -BioNTech COVID -19 vaccine and the 
seasonal inf luenza vaccine on the same day .
The following sub -cohorts will be assessed for each of the Cohorts A -D:
Individuals vaccinated with only  1 dose (i.e., incomplete course) of Pfizer -BioNTech 
COVID-19 vaccine;
Individuals vaccinated with 2 doses ( i.e., comple te course) of Pfizer -BioNTech 
COVID-19 vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073239
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 31of 1449.3.2.Baseline Characteristics
The following data elements regarding baseline demographic and clinical characteristics will 
be assessed based on a 1-year baseline period prior to the date of vaccination with Pfizer -
BioNTech COVID -19 vaccine and date of seasonal influenza vaccination for active 
comparator s.Depending on the attrition rate, the length of the baseline period may  be 
modified to 6 months. All diagnoses, procedures, and medications will be identified by the 
International Classification of Diseases, Tenth Revision, Clinical Modification ( ICD-10-CM)
diagnosis codes, ICD -10-PCS (procedure coding system) codes, ICD-10-CM Current 
Procedural Terminology  (CPT), orHealthcare Common Procedure Coding Sy stem (HCPCS ) 
procedure codes, and generic drug names, as appropriate( Appendix Table 1). The following 
demographic and clinical characteristics will be assessed:
Demographic s:
Age 
Sex
Race/ethnicit y
State
Clinical characteristics:
Smoking status
Body mass index (BMI)
History of anaphylaxis/
allergic reactions
Previous anaph ylaxisof vaccine component 
History of hospitalizations
Charlson c omorbidity  index (CCI)
Selected c omorbidities
oAutoimmune disease
oAsthma
oBleeding diathesis or condition associated with prolonged bleeding
oCancer
oCardiovascular conditions
oChronic kidney  disease/dialy sis
oChronic obstructive pulmonary  disease (COPD)/interstitial lung disease
oDiabetes mellitus
oDown syndrome
oSickle cell disease
oHepatitis B virus ( HBV)
oHepatitis C virus ( HCV)
oHuman immunodeficiency virus (HIV)
oHyperlipidemia
oHypertension
oLiver disease
oNeurological disease
oOther immune deficiencies
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073240
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 32of 144oSolid organ transplant
oVenous thromboembolism (VTE)
Concurrent immunizations
oSeasonal influenza vaccine
oTetanus diphtheria and pertussis (Tdap or Td)
oChickenpox (varicella)
oShingles (herpes zoster recombinant and/or live)
oHuman papillomavirus (HPV)
oPneumococcal conjugate
oPneumococcal polysaccharide
oHepatitis A
oHepatitis B
oMeningococcal conjugate (MenACWY) and serogroup B meningococcal 
(MenB)
oHaemophilus influenza type b
9.3.3.Outcomes
The safety events of interest for active surveillance wereidentified based on the Priority List 
of Adverse Events of Special Interest from the Brighton Collaboration’s Safety  Platform for 
Emergency  vACcines (SPEAC) Project, the FDA and Centers for Disease Control and 
Prevention (CDC) enhanced safet y monitoring recommendations.28,29Endpoints of special 
interest in signal detection, as noted by the FDA and CDC’s Advisory  Committee on 
Immunization Practices (ACIP) are denoted in italics .29If unanticipated pote ntial safety 
eventsof interest are identified during the course of surveillance, they  will be added to the 
list and included in the analy ses.See Appendix Table 2for the operational definitions of the 
outcome variables based on I CD-10-CM diagnosis codes, which may be refined as the study  
progresses based on additional available information and the publ ished literature (e.g., 
frequency  of ICD-10-CMcodes).Outpatient (including ED) and/orinpatient setting s will be 
used to identify  safety events of interest, depending on the t ype of event. The specific 
encounter setting considered for each safet y event of interest is summarized in Table 1.Any 
record of death will be captured, regardless of whether the individual died in a healthcare or 
non-healthcare setting. The following safet y events of interest will be assessed:
Neurologic :
Generalized convulsions/seizures
Guillain-Barré syndrome (GBS)
Aseptic meningitis
Encephalitis/encephalomyelitis
Other acute dem yelinating diseases
Transverse m yelitis (TM)
Multiple sclerosis (MS)
Optic neuritis (ON)
Bell’s pals y
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073241
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 33of 144Immunologic :
Anaphylaxis
Vasculitides
Arthritis and arthralgia/joint pain
Multisystem inflammatory syndrome in adults (MIS-A)
Kawasaki disease (KD)
Fibromyalgia
Autoimmune thy roiditis
Cardiac:
Myocarditis
Pericarditis
Acute myocardial infarction (AMI)
Hematologic :
Thrombocytopenia
Disseminated intravascular coagulation (DIC)
COVID-19(for all COVID -19-related safet y eventsof interest listed below, an inpatient 
diagnosis of COVID -19 will be required in combination with the codes or laboratory  values 
specified in Appendix Table 2; in addition, COVID- 19 related safety  events of interest will 
only be evaluated using data from 2020 onward using the SCRI  design):
Severe COVID -19 disease
Microangiopath y
Heart failure and cardiogenic shock
Stress cardiom yopathy
Coronary  artery disease (CAD)
Arrhythmia
Deep vein thrombosis (DVT)
Pulmonary  embolus
Cerebrovascular hemorrhagic stroke
Cerebrovascular non -hemorrhagic stroke
Limb ischemia
Hemorrhagic disease
Acute kidney  injury
Liver injury
Chilblain-like lesions
Single organ cutaneous vasculitis
Erythema multiforme
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073242
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 34of 144Other:
Death
Narcolepsy/cataplexy
Non-anaphylactic allergic reactions
Appendicitis
The risk and control intervals selected for the SCRI anal ysis for each safety event of interest 
arebased on biological plausibility  and precedents in the publ ished literature ( Table 1). A 
safety event of interest will only  be counted if it can be assigned to 1) the risk interval 
(following Pfizer -BioNTech COV ID-19 vaccination, 2) the pre -vaccination control interval,  
3) the post -vaccination control interval, or 4) the risk interval for the active comparators 
receiving seasonal influenza vaccine. Events outside the intervals will not be counted. Onl y 
the individual’s first instance of a safet y event of interest following a specified clean window 
(i.e., the occurrence -free baseline period used to define incident outcomes during which 
individuals enter the study cohort only  if the safety  event of interest did not occur during this 
period) will be included ; this means that if a safet y event of interest is identified but 
diagnosis codes (or laboratory  values in the case of se lect safety events of interest) 
corresponding to the safety  event of interest are also observed during the clean window, it 
will not be counted. The duration of the pre -specified window will differ by safet y events of 
interest in order to rule out pre -existing events. This approach is consistent with the FDA’s 
COVID-19 Vaccine Safety  Surveillance Project.14By way of example, a safety  events of 
interest for the SCRI  design can be considered in the following way s:
If a safetyevent of interest occurs in the individual’s pre- vaccination control 
interval and there are no other diagnosis codes for the same safety  event of 
interest in the clean window ( e.g., 1-year prior to that date), the safet y event of 
interest should be assign ed to the pre -vaccination control interval.
oIf a safety event of interest occurs in the pre -vaccination control interval 
but another diagnosis code for the same safety  event of interest is 
identified during the risk interval, then the safet y event of inter est will not 
be assigned to the risk interval and will onl y be assigned to the pre -
vaccination control interval as it will have occurred in the required clean 
window preceding the risk interval. However, if an outpatient safety event
of interest occurs in the clean window and an inpatient occurrence for the 
same type of safety eventof interest occurs in the risk interval, the 
inpatient occurrence will be counted in order to capture event 
exacerbation.
If a safety event of interest occurs in the risk inter val and there are no other 
diagnoses for the same safety event of interest in the clean window ( e.g., one-year 
prior to this date), which also includes the pre -vaccination control interval, then 
the safety event of interest will be assigned to the risk int erval. 
The same approach will be applied for the post- vaccination control intervals.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073243
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 35of 144The risk intervals for outcome evaluation for the active comparators who received 
seasonal influenza vaccination will be the same as for the individuals who received Pfizer-
BioNTech COVID -19 vaccine. 
However, it is possible that some safety eventsof interest do not have a precise time 
interval from which to evaluate risk, for example if biological plausibility  is unknown or the 
diagnostic time window is more delayed than anticipated. In these cases, misspecification of 
the risk (and control) intervals could result in misclassification and introduce bias, often 
toward the null. For instance, the assumption of a longer risk interval than is true may  result 
in “washing out” the signal, and an erroneousl y short risk interval may  similarly  result in 
underestimation of effect when using post -vaccination time intervals for self -control. To 
address this, sensitivity  analyses may be conducted with vary ing risk intervals (longer as well 
as shorter) in order to increase the likelihood that the safet y risk is detected accuratel y. 
Additionally , if further refinement and evaluation is necessary , temporal scan statistics may  
be used to empiricall y identify the at-risk time int erval by evaluating clusters of safety events 
of interest. This will be further described in the SAP. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073244
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 36of 144Table 1.Outcome algorithms for SCRI analysis , with risk and control intervals
Safety Event of Interest * Setting
(Inpatient [IP], 
Outpatient [OP])Clean 
windowPre-vaccination 
control interval 
(days)Risk 
interval 
(days)Post-
vaccination 
control interval 
(days)
Neurologic
Generalized convulsion/seizures8IP or OP86 months N/A 0-14 15-29
GBS8,22IP, primary  position 
only141 year N/A 1-42 43 -84
Aseptic meningitis30IP only141 year N/A 1-42 43-84
Encephalitis/ encephalomyelitis8IP only141 year -56 through -15 1-42 N/A
Other acute dem yelinating diseases8IP or OP81 year -98 through -15 1-42 N/A
TMaIP only141 year -98 through -15 1-42 N/A
MS8,22IP or OP81 year -98 through -15 1-42 N/A
ON8,22IP or OP81 year -98 through -15 1-42 N/A
Bell’s pals y8,22IP or OP141 year -56 through -15 1-42 N/A
Immunologic
Anaphylaxis8,22IP or OP146 months N/A 0-2 7-9
VasculitideseIP only 1 year N/A 1-28 29-56
Arthritisand arthralgia /joint paincIP or OP 1 year N/A 1-42 43-84
MIS-AbIP only141 year N/A 1-42 43-84
KD31IP only311 year N/A 1-28 29-56
FibromyalgiacIP or OP 1 year N/A 1-42 43-84
Autoimmune thy roiditiscIP or OP 1 year N/A 1-42 43-84
Cardiac
Myocarditis8,22IP or OP141 year -56 through -15 1-42 N/A
Pericarditis8,22IP or OP141 year -56 through -15 1-42 N/A
AMIdIPonly141 year -56 through -15 1-42 N/A
Hematologic
Thrombocy topenia30IP or OP141 year N/A 1-42 43-84
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073245
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 37of 144Table 1.Outcome algorithms for SCRI analysis , with risk and control intervals
Safety Event of Interest * Setting
(Inpatient [IP], 
Outpatient [OP])Clean 
windowPre-vaccination 
control interval 
(days)Risk 
interval 
(days)Post-
vaccination 
control interval 
(days)
DICeIPonly141 year N/A 1-42 43-84
COVID-19(for all COVID -19-related safet y events of interest listed below, an inpatient diagnosis of COVID -19 will be required in 
combination with the codes or laboratory  values specified in Appendix Table 2; in addition, COVI D-19 related safet y events of 
interest will only  be evaluated using data from 2020 onward using the SCRI  design):
Severe COVID -19 diseasebIP only 1 year N/A 1-42 43-84
Microangiopath yeIP only 1 year N/A 1-42 43-84
Heart failure and cardiogenic shockdIP only 1 year -56 through -15 1-42 N/A
Stress cardiom yopathydIP only 1 year -56 through -15 1-42 N/A
CADdIP only 1 year -56 through -15 1-42 N/A
ArrhythmiadIP only 1 year -56 through -15 1-42 N/A
DVTeIP or OP141 year N/A 1-42 43-84
Pulmonary  emboluseIP or OP141 year N/A 1-42 43-84
Cerebrovascular hemorrhagic stroke8IP only141 year N/A 1-42 43-84
Cerebrovascular non -hemorrhagic 
stroke8IP only141 year N/A 1-42 43-84
Limb ischemiaeIP only 1 year N/A 1-42 43-84
Hemorrhagic diseaseeIP only 1 year N/A 1-42 43-84
Acute kidney  injurygIP only 6 months N/A 1-42 43-84
Liver injurygIP or OP 1 year N/A 1-42 43-84
Chillblain -like lesionseIP or OP 1 year N/A 1-42 43-84
Single organ cutaneous vasculitiseIP only 1 year N/A 1-42 43-84
Erythema multiformefIP only 6 months N/A 1-2 8-9
Other
Narcoleps y and cataplexyaIP or OP141 year -98 through -15 1-42 N/A
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073246
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 38of 144Table 1.Outcome algorithms for SCRI analysis , with risk and control intervals
Safety Event of Interest * Setting
(Inpatient [IP], 
Outpatient [OP])Clean 
windowPre-vaccination 
control interval 
(days)Risk 
interval 
(days)Post-
vaccination 
control interval 
(days)
Non-anaphylactic allergic 
reactions8,22IP or OP86 months N/A 1-2 8-9
Appendicitis32IPonly146 months N/A 0-42 43-84
*Safety events of interestare based on the Priority List of Adverse Events of Special Interest from the Brighton Collaboration’s Safety Platform for 
Emergency vACcines (SPEAC) Project, the FDA and the Centers for Disease Control and Prevention’s (CDC) Advisory Committee on Immunization 
Practices (ACIP) enhanced safety monitoring recommendations.
Notes: 
aPublished risk and control intervals for demyelinating diseases and cranial disorders w ere applied to TM and narcolepsy/catap lexy.
bAs severe COVID- 19 ranges from severe pneumoni a, acute respiratory distress syndrome, and multisystem organ failure/MIS -A, a 1-42 day risk interval 
was applied in order to capture the 14 -day incubation period of the disease and 4 -5 day period from exposure to symptom onset.
c Published risk and control intervals for autoimmune disorders w ere applied to similar autoimmune rheumatic conditions (i.e., fibromyalgia and au toimmune 
thyroiditis).
dPublished risk and control intervals for myocarditis and pericarditis w ere applied to other cardiovascular conditions (i.e., heart failure and cardiogenic shock, 
stress cardiomyopathy, CAD, arrhythmia, AMI). 
eSimilar risk and control intervals were applied to all cardiovascular and hematological disorders characterized by damage to the blood vessels and/or arteries 
and clotting (i.e., microangiopathy, DVT, pulmonary embolus, limb ischemia, hemorrhagic disease, DIC, chilblain -like lesions). The published risk and 
control intervals for KD were applied to vasculitides given that KD is a type of medium and small -vessel vasculitis. 
fPublished risk and control intervals for non -anaphylactic allergic reactions were applied to hypersensitivity disorders (i.e., erythema multiform e).
gRisk intervals of 42 days were applied for acute kidney injury and liver injury to be consistent with other COVID -19 related safety events of interest .
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073247
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 39of 1449.4.Data Source 
The VHA is the largest integrated health care s ystem in the US, providing both inpatient and 
outpatient clinical care to over 9 million Veterans enrolled at more than 170 medical centers 
and 1,074 community -based outpatient clinics.12VHA’s health care delivery  system is 
organized regionall y around 18 Veterans Integrated Service Networks (VISNs) across the 
US. Each VISN is responsible for health care planning and resource allocation in a particular 
geographical region. For example, the VA New England Healthcare S ystem (VISN 1) covers 
VHA facilities in Massachusetts, Connecticut, New Hampshire, Maine, and Rhode I sland, 
while the VA Heart of Texas Health Care Network (VI SN 17) oversees the facilities in 
Texas. 
The VHA also maintains its own mortality  data where 99% of enrollees’ deaths are reported 
within one month of occurrence .As of January  7, 2021, the VHA has ha d over 174,000
confirmed COVID -19 cases.33Among active and convalescent cases , approximately  145,000
are Veterans and approximately  15,000are employees (with an estimated 630 as Veteran 
employees).33While African America n Veterans make up approximately  12% of the VHA,34
the burden of COVID -19 cases are skewed, with African American Veterans comprising 
approximately 20% of all COVID -19 cases.33Approximately 7,099COVID-19-infected VA 
patientshave died, an estimated 2,738in VHA hospitals.33
The objectives of this study  will be addressed using data from VHA’s Corporate Data 
Warehouse (CDW), which is an integrated EMR system with a centralized data warehouse 
that is updated on a dail y basis. The CDW stores data in separate databases, one for each t ype 
of clinical information ( e.g., inpatient medication, inpatient admission, outpatient medication, 
outpatient visit). Individualdemographic information such as date of birth and gender are 
also available. Immunization records include information on manufacturer, lot number, 
injection site, and concurrent immunizations. The CDW does not include information on an y 
care receiv ed outside of a VHA facility . 
Each individual is assigned a unique identification number to allow for longitudinal 
follow-up as well as to cross -reference to the various separate databases. For example, in 
each inpatient admission record, there is informa tion on the primary  discharge diagnosis (and 
as many as 15secondary  diagnoses), date of admission, date of discharge, and length of stay . 
This record can then be linked to other information of that inpatient stay  located in other 
files, including procedur es that the patient underwent during the hospitalization, medical 
specialty of the provider, and prescriptions dispensed. Other files are similarly  structured, 
and therefore may  be linked together to provide comprehensive information about the patient 
and his/her medical encounters. 
The VHA database is an appropriate data source to evaluate the safet y of the Pfizer -
BioNTech COVID-19 vaccine for the following reasons. First, as the vaccine will be
distributed through government facilities (including VHA) as part of initial distribution, 
analysis of VHA data will provide earl y data on the safet y of the vaccine. Veterans living in 
long-term care facilities and Veteran s who are healthcare workers will be prioritized in the 
first wave of Pfizer -BioNTech COVID -19vaccinations.35The VA offers eligible Veterans 
long-term care services ranging from nursing homes and assisted- living centers to caregiver 
support in the Veterans’ own homes.13Secondly, and relatedl y, VHA data are refreshed dail y 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073248
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 40of 144and would thus enable early  and rapid data anal ysis. Third, the VHA population is on 
average older than the general US population.36Of these, about 30% (roughly 
1,000,000 individuals) use VHA health services almost exclusively  (i.e., those with a priority  
group of 1 or 4; Veterans assigned to Priorit y group 4 are either accepting VA assistance or 
housebound benefits, or have been determined to be “catastrophicall y disabled” b y the 
VA.27), which lends itself to having complete, longitudinal healthcare data for such 
individuals who may be at higher risk of COVID -19 due to older age.37,38These priorit y 
groups include Veterans with the highest levels of service- connected disability  and are 
therefore, the highest priority  for VHA care.27Finally, the VHA population ha s, on average, 
more comorbid conditions t han the general population, which also indicates that these 
individuals may be at higher risk of COVID -19.39While the VHA pop ulation is 
predominantly  male (approximately  90%), and thus lacks generalizability  to females, it will 
still provide a useful setting to examine real -world vaccine safet y.
9.5.Study Size 
The sample size achieved will depend on the number of recipients of Pfizer -BioNTech 
COVID-19 vaccine within the VHA database during the study  period, which will increase 
over time with subsequent analy ses. The population size will increase with each bi-weekl y 
analysis as the Pfizer -BioNTech COVID-19 vaccine becomes more readily available and a 
greater number of individuals are vaccinated. Specificall y, the data wi ll be refreshed on a 
biweekly basis and a continuous sequential test procedure will be used to reevaluate data 
according to this schedule. As of January  21, 2021, 112,201 doses of Pfizer -BioNTech 
COVID-19 vaccine have been administered within the VHA (base d on CPT code 91300) to a 
total of 107,458 patients.
As a result of the ability  to perform near -real-time analysis, the risk interval (and post -
vaccination control interval, for applicable safety events of interest ) may have only partially 
elapsed in some cases. To account for this, we will use methods adopted in previous 
studies,8,25,40whereby risk intervals will be scaled (or truncated) in order to ensure an
equivalent length (or a fixed ratio) of time is assessed between the control and risk intervals.
9.5.1.Power
Power calculations for the rapid cy cle analysis (RCA) approaches proposed for safety event 
of interest signal detection will be conducted according to t he methods of Kulldorff et al.41,42 
Table 2 illustrates the estimated po wer for the RCA approach using the Poisson -based 
maximized sequential probability ratio test (MaxSPRT), and provides an overview of the 
power required to detect vary ing relative risk (RR ) estimates with an alpha level of 0.01. T 
denotes the expected number of safety events of interest to occur du ring the risk interval of 
interest (Table 2 and Table 3). Power of ≥ 80% is typically desirable in drug safety research. 
Usually
 the FDA views a RR of > 3 as meaningful, so this has been used for power 
calculations here.43 As an example, as shown in Table 2, the surveillance system would  have 
sufficient power (80.0%) to detect an increased risk of 
safety events of interest associated 
with the Pfizer -BioNTech COVID- 19 vaccine b y 3 fold when the expected number of safety 
events of interest
 reaches 6 events. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073249
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 41of 144Table2. Estimated Statistical Power for the Poisson -based MaxSPRT41
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073250
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 42of 1449.6.Data Management
Data for this stud y will be stored and extracted from the VHA database (p reviously described 
in Section 9.4) that contain information about patient demographics, vaccinations, 
procedures, diagnoses, and death.  
9.6.1.Case report forms (CRFs)/Electronic data record
As used in this protocol, the term CRF should be understood to refer to either a paper form or 
an electronic data record or both, depending on the data collection method used in this study .
A CRF is required and should be completed for each included patient in the signal 
verification phase that requires EMR and chart review ( see Section 9.7.3.3). The completed 
original CRFs should not be made available in any form to third parties, except for 
authorized representatives of Pfizer or appropriate regulatory  authorities, without written 
permission from Pfizer. The CRF will consist of two parts: (1) a database CRF that will be 
populated based on a direct extraction of data from the VA CDW for review by  the 
adjudicators; (2) an adjudication page that will be completed by  an adjudicator after 
reviewing data in the completed CRFs. Analysis Group shall ensure that the CRFs are 
securely stored on VHA servers in an encry pted electronic and/or paper] form and will be 
password protected or secured in a locked room to prevent access b y unauthorized third 
parties.
Analysis Group has ultimate responsibility  for the collection and reporting of all clinical, 
safety, and laboratory data entered on the database CRFs and an y other data collection forms 
(source documents) and ensuring that they  are accurate, authentic/original, attributable, 
complete, consistent, legible, timely  (contemporaneous), enduring, and available when 
required.  The adjudication page must be signed by  the adjudication committee members to 
attest that the data contained on the formsare true and accurate based on their review of the 
EMR data . Any corrections to entries made in the CRFs or source documents must be dated, 
initialed, and explained (if necessary) and should not obscure the original entry .
The source documents are the hospital or the ph ysician's chart. In these cases, data collected 
on the CRFs must match those char ts. 
9.6.2.Record retention
To enable evaluations and/or inspections/audits from regulatory  authorities or Pfizer, 
Analysis Group agrees to keep all study -related records, which includes study  documents and 
deliverables such as the protocol, SAP, aggregated results tables, SAS programming files, 
and study  report. The records should be retained by Analysis Group according to local 
regulations or as specified in the vendor contract, whichever is longer. Analy sis Group must 
ensure that the records continue to be sto red securely  for so long as they  are retained.
If Analysis Group becomes unable for any  reason to continue to retain study  records for the 
required period, Pfizer should be prospectivel y notified. The study records must be 
transferred to a designee accepta ble to Pfizer.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073251
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 43of 144Study records must be kept for a minimum of 15 years after completion or discontinuation of 
the study, unless Analy sis Group and Pfizer have expressly  agreed to a different period of 
retention via a separate written agreement. Record must be retained for longer than 15 years 
if required b y applicable local regulations.  
Analysis Group must obtain Pfizer's written permission before disposing of any  records, even 
if retention requirements have been met.
9.7.Data Analysis
Detailed methodology for summary  and statistical anal yses of data analyzedin this study  will 
be documented in a statistical anal ysis plan (SAP), which will be dated, filed ,and maintained 
by the sponsor. The SAP may  modify the plans outlined in the protocol; any major 
modification s of primary endpoint definitions or their anal yses would be ref lected in a 
protocol amendment. The SAP will also provide additional detail regarding the evaluation of 
a threshold of excess risk for each of the safet y events of interest. Consistent with th e 
approach of Kulldorff et al. , this will be determined based on background incidences for each 
event (e.g., based on historical influenza vaccinated active comparator cohort data to be 
evaluated during the study), in addition to pre -specified significance level (e.g., alpha=0.01 
or 0.05) and power.41This information, in conjunction with a clinically  meaningful RR ( e.g., 
2 or 3) and the expected upper limit of events under the null hy pothesis will allow for the 
calculation of critical values of each safet y event of interest using the MaxSPRT method. 
Greater power ( e.g., 80%) is also a natural criterion to use when selecting the upper limit on 
the length of surveillance, and in turn, the expected number of events to occur, although there 
is ultimately  a tradeoff between that power and the time allowed to identify  the expected 
number of events to occur.
Data analy ses will be conducted using SAS Enterprise Guide version 7.1 (SAS I nstitute Inc., 
Cary, NC) or R Version 3.5.3 or its latest version (R Core Team, Vienna, Austria). In 
addition, SaTScan will also be used to conduct specific temporal anal yses.
9.7.1.Baseline Characteristi cs
Baseline demographics and clinical characteristics for individuals receiving Pfizer -BioNTech 
COVID-19 vaccine and individuals who received seasonal influenza vaccination will be 
summarized using descriptive statistics, consisting of the mean and standard deviation (SD) 
and median (interquartile range [IQR]) values for continuous variables and frequency  
distributions for categorical variables. Incidence rates ( i.e., per-patient per -month) for prior 
hospitalizations may  be calculated as the number o f events divided b y person-time of 
observation since the length of the baseline period may vary  between individuals. 
Standardized differences will be calculated between Pfizer BioNTech COVID -19 vaccine 
recipients and active comparators who received seasona l influenza vaccination to evaluate 
whether there are an y major differences in individuals’ baseline characteristics. Standardized 
differences <10% will indicate that matching has appropriatel y balanced the charac teristics 
between recipients of the Pfizer-BioNTech COVID- 19 vaccine and seasonal influenza 
vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073252
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 44of 1449.7.2.Vaccine Utilization Patterns
Descriptive statistics will also be used to summarize vaccine utilization patterns, including 
proportion of individuals receiving vaccine, 2 -dose completion rate ,distribution of time gaps 
between the first and second dose , and care setting where immunization was received ( e.g., 
outpatient clinic, pharmacy , inpatient ward). Counts of individuals who received a COVID -
19 vaccine from a different manufacturer in addition to the Pfizer -BioNTech COVID-19 
vaccine will be reported.
9.7.3.Safety Signal Analyses
Several analy ses corresponding to the designs discussed previousl y will be conducted to 
detect safet y signals associated with Pfizer -BioNTech COVID -19 vaccine. Analy ses will be 
conducted among all individuals receiving the vaccine, individuals who received 
Pfizer-BioNTech COVID-19 vaccine without seasonal flu vaccine (Cohort A will be used for 
SCRI; Cohort B +Cwill be used for active comparator analyses), and individuals receivin g 
Pfizer-BioNTech COVID- 19 vaccine and seasonal flu vaccine on the same day  (Cohort D), 
along with sub- cohorts receiving onl y one dose vs. two doses. 
A stepwise process, illustrated below, will be performed for signal detection, evaluation, and 
verification (Figure 3).This approach has been adapted from the Active Monitoring Protocol 
of the FDA’s COVID -19 Vaccine Safet y Surveillance Project.14The statistical approach 
described below may  be modified further based on data availability , additional clinical input, 
and for consistency  or to complement similar studies of Pfizer -BioNTech COVID-19 
vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073253
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 45of 144Figure 3. Steps in Signal Detection, Evaluation , and Verification
Notes: 
[1] List of safety events of interest and corresponding definitions may be refined as the study progresses based 
on additional available information.
[2] The risk and control intervals selected for the SCRI analysis for each safet y event of interest are based on 
biological plausibility and precedents in the literature. Only the individual’s first instance during the specified 
clean window  (i.e., the interval used to define incident outcomes) will be included.  Note that only the first 
inpatient or outpatient occurrence of a safety event of interest following the clean window will be used to 
identify incident events ( e.g., if an inpatient safety event of interest occurs in the clean window , a repeat 
occurrence will not be counted i n the risk interval). However, event worsening will be counted as a safety event 
of interest. For example, if an outpatient safety event of interest occurs in the clean window and an inpatient
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073254
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 46of 144occurrence for the same type of safety event of interest occurs in the risk interval, the inpatient occurrence will be counted as 
a safety event of interest. 
9.7.3.1.Signal Detection
9.7.3.1.1. Sequential Testing -SCRI Design using the Binomial -based MaxSPRT for 
Comparison to Pre -vaccination Control Intervals
The goal is to provide ra pid-cycle, near real -time safet y surveillance. In the signal detection 
phase, the SCRI  analysis will only  include pre -vaccination control intervals as the post -
vaccination control intervals will require a longer time to accumulate and thus will not allow 
for timely analysis. The post -vaccination control period will be assessed during the signal 
evaluation phase (see Section 9.7.3.2), to allow for additional observation time to accrue as 
well as to more deeply  investigate potential signals. This will allow for timely  RCA without 
the need to wait for data to accumulate for safety events of interest with post-vaccination 
control intervals. 
To account for multiple testing and bi -weekly review of the data, the MaxSPRT using a 
binomial probability  model will be applied. The null hy pothesis (H 0) assumes that the risk of 
a safety event of interest during the risk interval is equivalent to the risk of the same safety 
event of interest developing during the control interval, accounting for differences in interval 
duration as needed ( e.g., for safety events of interest such as demyelinating disease), meaning 
a RR of 1 is specified under H 0.22The one-sided composite alternative h ypothesis (H a) 
assumes that the risk of a safety event of interest during the risk interval is greater than the 
risk of the same safety eventof interest developing during the control interval, accounting for 
differences in interval duration ( i.e., RR>1, H ais applicable across a range of RRs).41
Specifically, for the Pfizer -BioNTech COVID- 19 vaccine, let xrepresent the total count of 
safety events of interest in the control interval ( Figure 4), let y represent the total count of 
safety events of interest in the risk i nterval, and let rrepresent the ratio of yto x under the 
null hypothesis. Thus, when the total control interval duration and total risk interval duration 
are equal, r will be 1. The RR is estimated by  ..
.25The RR and corresponding 9 9% 
confidence intervals (CIs) will be calculated. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073255
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 47of 144Figure 4. Example of SCRI Design for a Safety Event of Interest with a 42- day Risk 
Interval and a Pre- vaccination Control Interval
For the binomial mode l, the log-likelihood ratio (LLR) is calculated as the log probability  of 
observing this distribution of y under H a, divided by  the probability  of this occurring under 
H0.41This ratio is calculated whenever new data arereceived to account for the continuous 
data stream until the full 42 -day risk period is complete.
=ln(|)
(|0)
Once the LLR test statistic reaches a pre -specified critical value , a signal is detected.  
Specificall y, the null hypothesis will be rejected if the LLR exceeds the critical value. The 
null hypothesis will not be rejected if the LLR does not reach or exceed the critical value, if 
the total number of safety events of inte restreaches a pre -specified upper limit, or if 
surveillance ends without reaching this upper limit.25
For each safety event of interest (and specific to each age group, if age- stratified anal yses are 
conducted), the criti cal value of the LLR will be determined based on the safet y event of 
interest specific upper limit of expected safety events of interest and alpha level.25Upper 
limits will be determined based on the expected number of safety events of interest under the 
null hypothesis, assuming the risk after Pfizer -BioNTech COVID- 19 vaccination is no 
greater than the risk of safety events of interest after seasonal influenza vaccination. 
Therefore, upper limits will be chosen such that t hey would not usually  be reached. 
9.7.3.1.2. S equential Testing -Poisson-based MaxSPRT for Comparison to Active 
Comparators who Received Seasonal Influenza Vaccination
For comparison with active comparators who received seasonal influenza vaccination, the 
Poisson-based MaxSPRT will be applied, following the same statistical approach as 
described above, but using a Poisson probabilit y distribution. In the Poisson MaxSPRT 
approach, the event frequency  of safety events of interest in the risk interval after Pfizer-
BioNTech COVID -19 vaccination will be compared to a background rate ofsafety events of 
interest in the risk interval after seasonal influenza vaccination in five prior seasons, ranging 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073256
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 48of 144from 2014/15 through2018/19. This approach is particularl y important for extremely  rare 
safety events of interest (i.e., less than 50 anticipated based on historical influenza vaccine 
ratesof safety events of interest ).22Poisson MaxSPRT is used to mo nitor very  rare safet y 
events of interest as binomial MaxSPRT may  not detect a signal, despite a clinically  
meaningful RR.25This will also allow for more timely anal ysis using historical data, as well 
as improved power and sample size.
GBS is of particular interest relative to the safety profile of Pfizer -BioNTech 
COVID-19vaccine. As GBS is an extremely  rare safety event of interest , the primary  RCA 
proposed will focus on Poisson MaxSPRT and apply  an alpha of 0.05. The Poi sson 
MaxSPRT has increased power to detect a signal with fewer occurrences of the safet y event 
of interest. However, this method cannot fully  control for confounding by  indication. 
9.7.3.1.3. Critical Values and Alpha Spending
Critical values for the LLR test statistic are shown below in Table 3based on calculations 
conducted b y Kulldorff et al 2011.41For example, assuming T =6 (number of expected 
events under the null) and RR =3, which corresponds to a power of 80.0% (See 
Section9.5.1), the critical value would be 5.14 using alpha of 0.01 for the Poisson -based 
MaxSPRT. As noted previously , each safety  event of interest will be evaluated separatel y to 
determine a critical va lue based on background incidence, alpha, power, and clinically  
meaningful RR. These details will be addressed in the SAP.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073257
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 49of 144Table 3. Critical Values for Poisson -based MaxSPRT
Multiple ty pes of alpha spending functions can be employ ed to calculate the cumulative rate 
at which Ty pe 1 error (alpha) probabilit y is spent during sequential testing.44To achieve 
optimal expected time -to-signal, especiall y when historical Poisson data are used with 
surveillance data, a power-typeconvex alpha spending shape will be used based on published 
literature.44Additionally , ρ =1.5 is referenced as a “rule of thumb” as it is suggested to be 
appropriate in most applications.
9.7.3.2.Signal Evaluation
Signals are detected when the event frequency  of asafety eventof interest during the risk 
interval following vaccination with Pfizer -BioNTech COVID -19 vaccine is significantl y 
increased compared to the event frequency  of the same safety events of interest in the control 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073258
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 50of 144comparator (i.e., the critical value is achieved and surpassed). If signals are indeed detected 
for safety  events of interest based on the anal ysis described above, further evaluation is 
warranted to refine and conf irm such detections. This will include the following additional 
analyses to assess the robustness of the findings. 
9.7.3.2.1. Post-Signal Quality Assurance
Quality assurance will first be conducted in order to assess the quality  of the data and 
analysis that produce d the signal. While quality  control measures will be conducted during 
the signal detection phase (seeSection 9.8), post-signal qualit y assurance will also be 
performed during the signal evaluation phase. This will include a comprehensive quality 
assurance (for example, check for possible duplications of claims or medical records, 
checking for unusual clustering in claim or medical record accrual b y service date for 
potential coding issues, check for geographical distribution of cases that may be related to lot 
numbers or diagnostic practice) . In addition, for signals detected via active comparison, 
additional anal yses comparing to pre- vaccination contro l intervals may  be formed to check 
for consistency . Signals will also be confirmed across all of the safety  studies planned to be 
performed (i.e., C4591008, C4591011, C4591012) to confirm that specific data sources are 
not biased. 
9.7.3.2.2. Multivariate Adjustment using Poisson Regression
If signals are detected and persist after conducting quality  assurance, further evaluation via 
statistical measures are warranted. Specifically , to investigate whether potential signals 
identified via Poisson MaxSPRT for the compar ison to active comparators with seasonal 
influenza vaccination are not confounded ( i.e., to take into account baseline differences 
between the Pfizer BioNTech COVID -19 vaccinated and active comparator populations), a 
multivariate Poisson regression anal ysis will be conducted to compare the incidence rates of 
the safety events of interest occurring within the risk intervals. T he predictor would be
whether the individual had received the Pfizer -BioNTech COVID -19 vaccine or had received 
the influenza vaccine d uring historical seasons. Analy ses will be adjusted for relevant 
baseline and/or clinical characteristics ( e.g., age, sex, race, CCI and/or specific comorbidities 
of interest, state, etc.).8
If the signal remains, based on an IRR > 3 with a p- value <0.01 from the adjusted Poisson 
regression, further evaluation may be considered via signal verification. 
9.7.3.2.3. Assessment of Temporal Clusters
Vaccine safety  surveillance must allow for sufficient ty pe I error probability for rapid 
detectionofsafety events of interest , and statistically  significant signals must be studied 
further to ensure that a true association is present.45Therefore, the presence of temporal 
clusters will be assessed using the so ftware SaTScan to calculate temporal scan statistic in 
order to further refine safety  signals detected from the signal detection analy ses.22A temporal 
scan statistic accounts for multiple testing present during overlapping risk intervals. The null 
hypothesis assumes that there is no association between the safet y eventsof interest and 
immunization, and safety events of interest are assumed to be distributed independentl y and 
uniformly  during a period of time subsequent to Pfizer -BioNTech COVID- 19 vaccination.22
A temporal scan statistic will be generated b y moving a time interval of fixed length across 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073259
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 51of 144the risk interval, comparing the number of observed versus expected safety events of inter est
within the time interval under the null hy pothesis.46
9.7.3.2.4. Sequential Testing -SCRI Design using the Binomial MaxSPRT for 
Comparison with Post -Vaccination Control Intervals
Similar to the SCRI  design using the binomial -based MaxSPRT method for pre -vaccination 
control intervals, sequential testing anal yses will be conducted using the post- vaccination 
control intervals as appropriate for specific safety events of interest . This will be conducted 
during the signal evaluation p hase in order to allow time to accumulate during the post -
vaccination control period. The same statistical methodology  as described for the pre -
vaccination control intervals will be applied.
9.7.3.3.Signal Verification
If a signal persists after conducting signal evaluation , signal verification through medical 
records review may  be conducted .
9.7.3.3.1. Medical Records Review
As part of the signal evaluation process, diagnostic validation of the detected safet y events of 
interest (i.e.,cases) via a djudication of patient medical records by VHA clinicians for 
outcome verification in a representative sample of cases will be conducted. The total number 
of charts to be reviewed will depend on the number of safety events of interest detected, such 
that all cases may be reviewed for safety  events of interest where a small number of events 
result in signal detection and a representative sub -sample may  be reviewed for safety events 
of interest where a larger number of events results in signal detection .47For rare events, 
potentially  all cases may  be adjudicated. An adjudication charter will be developed to govern 
signal evaluation and medical records review. Specificall y, validation of detected safet y 
events of interest will be performed through patient medical chart review in collaboration 
withan adjudication committee comprised ofthe treating or trained healthcare 
professionals.47
9.7.4.Seasonality- Adjusted Cases -Centered Method
A case-centered anal ysis for specific safety  events of interest for which signals were detected 
may also be conducted in order to account for bias caused b y seasonality of safety events of 
interest and vaccination.23This method will use data on all safety event of interest cases that 
occur after vaccination with Pfizer -BioNTech COVID -19 vaccine. Logistic regression will 
be used to compare the number of safety event of interest cases that were vaccinated inside 
versus outsid e a pre-specified risk interval , as of the date of the safet y events,where the total 
number of vaccinations given inside versus outside the risk interval (in the population of all 
vaccinees) is used as the offset term.25Specifically, the association of vaccination with risk 
of safety events of interest will be estimated from a logistic regression model that includes 
summarized data with one record per risk set. The key  independent variable will be the 
proportion of the risk set who were in the risk interval on the date of the safety  eventof 
interestoccurrence. In this way , risk sets are anchored to calendar dates, and confounding b y 
seasonality  of the safet y eventsof interest and vaccination is addressed .48Note that other 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073260
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 52of 144confounders may  also be adjusted for b y restricting risk sets to vaccinees similar with respect 
to select characteristics ( i.e., through stratification).  
9.7.5.End-of-Season and End -of-Surveillance Analyses
For any safety event of interest with signals detected, end-of- season anal yses (over the course 
of the 30- month period) and an end -of-surveillance anal ysis (i.e., at 30 months, after the end 
of surveillance) will be conducted. Similar methodology  will be applied for the end -of-
surveillance anal ysis and end-of- season anal ysis conducted for seasonal influenza vaccine in 
order to adjust for the seasonality  of both disease and vaccine administration.8This approach 
will be able to define the true risk intervals after each dose and estimate the risk for potential 
safety events of interest after both dose 1 and 2 of the Pfizer -BioNTech COVID- 19 vaccine, 
as well as the abilit y to discern whether or not on e or two doses of seasonal influenza vaccine 
were administered during the same period.  
The number of events in the sum of three distinct risk intervals will be compared to the 
control interval, adjusting for potential differences in interval length, to e stimate the RR of 
Pfizer-BioNTech COVID-19 vaccine compared to the influenza vaccine . In order to monitor 
the safety after the first and full course of the vaccine, the number of potential safety events 
of interest occurring in three separate risk intervals (P 1, P2, P3) will be estimated ( Figure 5). 
P1represents the risk interval after the first dose only , excluding any  overlap in risk intervals 
with the second dose. P 2represents the overlapping risk intervals for first and second dose of 
the vaccine. P 3represents the risk interval of the second dose of the vaccine, excluding the 
overlapping risk interval alread y captured in P 2. This design will allow fo r the assessment of 
risk during the appropriate periods, regardless of the time interval between vaccine doses. As 
multiple endpoints will be assessed, 99% CIswill be calculated around the RR in order to 
ascertain whether the Pfizer- BioNTech COVID -19 vaccine is associated with safety  events 
of interest. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073261
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 53of 144Figure 5. Example of Risk (P1, P2, P3) and Pre -vaccination Control Intervals for the 
SCRI End -of-surveillance Analyses of 1 or 2 Doses of Pfizer-BioNTech COVID -19 
Vaccine
In Figure 5A, P1+ P2+ P3represent the risk intervals where a safet y event of interest may 
occur. In Figure 5B,there is no overlapping risk interval so that P 1+ P3represent the risk 
intervals where a safety event of interest may occur.The timing of the risk and control 
intervals maybe adjusted for in order to control for the effect of seasonalit y across the 
intervals assessed.
9.7.6.Subgroup Analysis
Separate anal yses of baseline characteristics, vaccine utilization patterns, signal detection, 
signal evaluation, and signal verification insubgroups of interest may  be conducted based on 
feasibility , sample size, and data available. 
9.7.7.Incidence Rates and Time to Safety Event of Interest Analysis
Incidence rates (and corresponding CIs) will be calculated from safety event of interest signal 
detection anal yses. Kaplan -Meier methods will be used to anal yze time-to-event (i.e., time to 
safety event of interest ). If individuals do not experience thesafet y events of interest , they 
will be censored at the end of the risk interval. Median time to safety event of interest and 
corresponding CI swill be reported.
9.8.Quality Control
Data for the study  will be extracted from electronic databases in the CDW of the VHA. Each 
data content area in the CDW is subjected to similar checks, from high level variable 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073262
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 54of 144name/type checks, to detailed trending comparisons. As an example, the diagnostic data is 
subject to the following checks:
Referenced table exists
Diagnosis type is correctly assigned b y codes defining the diagnosis
Percentages, rates, are as expected (check ranges and for missing)
Both inpatient and outpatient diagnosis codes are captured. Referenced variables exist 
and are of appropriate length and t ype
Data retrieval will be coordinated b y an experienced programmer/anal yst. The anal yst will 
write programm ing for retrieval of each data element from the electronic databases. Double 
programming will be performed for the first iteration of the anal yses; results/datasets will be 
compared, and if an y discrepancies are identified, both programmers will determine a 
resolution, bringing in a third programmer if needed. Subsequent iterations of anal yses (i.e., 
re-runs of the anal yses) will be audited by  a senior programmer. All tables will be reviewed 
by the project manager and the principal investigator to evaluate for internal consistency  of 
counts and totals. All calculated variables will be checked against the component variables 
(cross tabs) to ensure accuracy . For example, categorical age would be compared with 
continuous age to confirm that each category  of agecontained onl y individuals of the 
expected age ranges within that category .
9.9.Strengths and Limitations of the Research Methods
To identify  individuals who experienced safety events of interest associated with Pfizer-
BioNTech COVID -19 vaccine, the SCRI  method of signal detection offers some key  
advantages. The SCRI  approach inherentl y adjusts for within-individual confounders, such as 
age, sex, and confounding b y indication. Additionally , the inclusion of a post -vaccination 
control period will account for in creased detection bias from stimulated reporting of safet y 
events of interest due to heightened vigilance on COVID -19 vaccines.49Specificall y, safety 
events of interest may  be more likely  to be reported or sought care for after vaccination with 
Pfizer-BioNTech COVID-19 vaccine than before ( i.e., during the pre -vaccination control 
interval)which may  result in bias against the Pfizer -BioNTech COVID -19 vaccine . Lastly, 
SCRI allows for near real- time monitoring of safety  risksassociated with the Pfizer -
BioNTech COVID -19 vaccine. 
The VHA CDW provides a range of benefits, including its comprehensive structure, large 
number of variables, and electronic accessibility. The VHA CDW also includes EMR data 
that include structured fie lds (which will be used for signal detection) and open fields (such 
as physician notes, which will be used for signal evaluation and case validation, as needed). 
Importantly , the VHA CDW retains electronic immunization records that include 
manufacturer nam e and lot numbers, facilitating the identification of brand- specific vaccines, 
such as the Pfizer -BioNTech COVID -19 vaccine. Moreover, the VHA CDW data are 
updated on a dail y basis, enabling near real -time rapid monitoring of potential safet y signals. 
However, there are several limitations when rely ing on VHA that should be noted. First, 
there could be gaps in the data since individuals may  receive healthcare services outside of 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073263
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 55of 144VHA facilities. As such, if individuals receive the Pfizer-BioNTech COVID-1 9 vaccine 
outside of a VHA facility, this information will not be captured in the VHA EMR sy stem. 
Similarly, individuals may  have also received past seasonal influenza vaccinations outside of 
the VHA s ystem, and thus would be misclassified as not having re ceived vaccine in the 
current anal ysis. For example, veterans with secondary  insurance or veterans who are 
65years of age or older who have Medicare may  receive health care services outside of VHA 
facilities. One stud y on VHA enrollees in seven different states found that of all individuals 
admitted to VHA hospitals in 2007, one fifth also had a non VHA hospitalization during that 
year.50Another study  reported that about 53% of Veterans 65 years of age and older who 
were dually  eligible for VHA and Medicare services in 2003 2004 used both.51Hence, it is 
important to note that data on vaccination status ma y be incomplete. However, this limitation 
will be addressed b y examining subgroups of individuals who receive care regularl y at VHA 
facilities, as well as those with Priority  group 1 status, to ensure that their healthcare data are 
complete to the extent p ossible in the CDW. Second, to the extent that the individuals in the 
VHA database are different from individuals outside of the VHA, the results may  not be 
generalizable to the broader US population. For example, since the VHA includes 
predominantly  male Veterans (approximately  90% male), findings from this study  may not 
be generalizable to women in the US. 
9.10.Other Aspects
Not applicable.
10. PROTECTION OF HUMAN SUBJECTS
10.1.Patient Information
All parties will comply  with all applicable laws, including laws regarding the implementation 
of organizational and technical measures to ensure protection of patient personal data. Such 
measures will include omitting patient names or other directl y identifiable data in an y 
reports, publications, or other disclosures, except where required b y applicable laws.
To protect the rights and freedoms of natural individuals with regard to the processing of 
personal data, when study  data are compiled for transfer to Pfizer and other authorized 
parties, an y patient names will be removed and will be replaced by  a single, specific, 
numerical code. All other identifiable data transferred to Pfizer or other authorized parties 
will be identified by  this single, patient -specific code. In case of data transfer, Pfizer will 
maintain high standar ds of confidentialit y and protection of individuals’ personal data 
consistent with the vendor contract, and applicable privacy  laws. 
No personal data is planned to be transferred off the VA servers. Specifically, the Clinical 
Epidemiology  Program (CEP) at White River Junction VA Medical Center will conduct this 
safety surveillance stud y with sponsorship from Pfizer and assistance from Analy sis Group, 
Inc. The project will be led by  the VA, with Dr. Yinong Young -Xu, Director of CEP, serving 
as the Principal Investigator. Data access will be granted through VA Informatics and 
Computing I nfrastructure (VINCI). VHA data will not be provided to Pfizer or Anal ysis 
Group.Rather, onl y VA employees, including those with research service without 
compensation (WOC) e mployee status, who have completed necessary  VA training and have 
proper clearance will access and anal yze data on secure VA servers and behind necessary  
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073264
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 56of 144firewalls, under the direction and supervision of Dr. Young -Xu. Given the sensitive nature of 
healthcare data, comprehensive securit y measures will be implemented to ensure the 
confidentiality , integrity, and protection of Veterans’ privacy and healthcare data.
10.2.Patient Consent
As this study  does not involve data subject to privacy  laws according to applica ble legal 
requirements, obtaining informed consent from individuals by  Pfizer is not required.  
10.3. Institutional Review board (IRB)/Independent Ethics Committee (IEC)
There must be prospective approval of the study  protocol, protocol amendments, and their 
relevant documents from the relevant IRBs/IECs. All correspondence with the I RB/IEC must 
be retained. Copies of IRB/IEC approvals must be forwarded to Pfizer. The study  protocol 
will be reviewed b y the IRB of the VA Medical Center, White River Junction, VT.
10.4.Ethical Conduct of the Study
The study  will be conducted in accordance with legal and regulatory  requirements, as well as 
with scientific purpose, value and rigor and follow generall y accepted research practices
described in Guidelines for Good Pharmacoepi demiology  Practices (GPP) issued by  the 
International Societ y for Pharmacoepidemiology ,53the FDA Guidance for Industry and FDA 
Staff: Best Practices for Conducting and Reporting, Pharmacoepidemiologic Safet y Studies 
Using Electronic Healthcare Data52and Good Epidemiological Practice (GEP) guidelines 
issued by the International Epidemiological Association (IEA).54
11.MANAGEMENT AND REPORTING OF ADVERSE EVENTS/A DVERSE 
REACTIONS 
Signal Detection and Signal Evaluation
This study  involves data that exist as structured data by  the time of study  start or a 
combination of existing structured data and unstructured data, which will be converted to 
structured form during the implementation of the protocol solely  by a computer using 
automated/algorithmic methods, such as natural language processing. In these data sources, 
individual patient data are not retrieved or validated, and it is not possible to link (i.e., 
identifya potential association between) a particular product and medical event for any  
individual.  Thus, the minimum criteria for reporting an adverse event (AE) (i.e., identifiable 
patient, identifiable reporter, a suspect product, and event) cannot be met. 
Signal Verification
This study  protocol requires human review of patient- level unstructured data; unstructured 
data refer to verbatim medical data, including text-based descriptions and visual depictions 
of medical information, such as medical records, imag es of physician notes, neurological 
scans, X-rays, or narrative fields in a database.  The reviewer is obligated to report adverse 
events (AEs) with explicit attribution to any  Pfizer drug that appear in the reviewed 
information (defined per the patient po pulation and study  period specified in the protocol).  
Explicit attribution is not inferred b y a temporal relationship between drug administration 
and an AE, but must be based on a definite statement of causality  by a healthcare provider 
linking drug admin istration to the AE.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073265
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 57of 144The requirements for reporting safet y events of interest on the non -interventional study  
(NIS) adverse event monitoring (AEM) Report Form to Pfizer Safety  are as follows:
All serious and non- serious AEs with explicit attribution to any Pfizer drug that 
appear in the reviewed information must be recorded on the data collection tool (e.g., 
chart abstraction form) and reported, within 24 hours of awareness, to Pfizer Safet y 
using the NIS AEM Report Form.
Scenarios involving drug exposure, including exposure during pregnancy , exposure 
during breast feeding, medication error, overdose, misuse, extravasation, lack of 
efficacy, and occupational exposure associated with the use of a Pfizer product must 
be reported, within 24 hours of awareness, to Pfizer Safet y using the NI S AEM 
Report Form.
For these AEs with an explicit attribution or scenarios involving exposure to a Pfizer 
product, the safety  information identified in the unstructured data reviewed is captured in the 
Event Narrative section of the report form, and constitutes all clinical information known 
regarding these AEs.  No follow-up on related AEs will be conducted.
All the demographic fields on the NI S AEM Report Form may  not necessarily  be completed, 
as the form designates, since not all elements will be available due to privacy  concerns with 
the use of secondary  data sources. While not all demographic fields will be completed, at the 
very least, at least one patient identifier (e.g., gender, age as captured in the narrative field of 
the form) will be reported on the NI S AEM Report Form, thus allowing the report to be 
considered a valid one in accordance with pharmacovigilance legislation. All identifiers will 
be limited to generalities, such as the statement “A 35- year-old female...” or “An elderl y 
male...”  Other identifiers will have been removed.      
Additionally , the onset/start dates and stop dates for “Illness”, “Study  Drug”, and “Drug 
Name” may  be documented  in month/y ear (mmm/yyyy ) format rather than identify ing the 
actual date of occurrence within the month /y ear of occurrence in the day /month/year 
(DD/MMM/YYYY ) format.
All research staff members must complete the following Pfizer training requirements:   
Your Reporting Responsibilities ( YRR)Training for Vendors Working on Pfizer 
Studies 
These trainings must be completed by  research staff members prior to the start of data 
collection.  All trainings include a “Confirmation of Training Certificate” (for signature b y 
the trainee) as a record o f completion of the training, which must be kept in a retrievable 
format.  Copies of all signed training certificates must be provided to Pfizer. 
Re-training must be completed on an annual basis using the most current Your Reporting 
Responsibilities training materials. 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073266
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 58of 14412.PLANS FOR DISSEMINAT ING AND COMMUNICATING STUDY RESULTS
This protocol will be posted on publicly available registers following its finalization. The 
final study  results will be made publicly  available via the European Union Post Authorisation 
Safety (EU PAS) Register and may  be submitted for publication in a peer reviewed medical 
journal.
In the event of an y prohibition or restriction imposed (e.g., clinical hold) by an applicable 
competent authorit y in any area of the world, or if the inves tigator is aware of an y new 
information which might influence the evaluation of the benefits and risks of a Pfizer 
product, Pfizer should be informed immediately.  
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073267
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 59of 14413.REFERENCES
1. World Health Organization (WHO). WHO Director-General’s opening remarks at th e 
media briefing on COVID -19 -11 March 2020.  
https://www.who.int/dg/speeches/detail/who- director-general-s-opening-remarks-at-
the-media-briefing-on-covid-19---11-march-2020. Accessed November 11, 2020.
2.Johns Hopkins University. Coronavirus Resource Cen ter.  
https://coronavirus.jhu.edu/. Accessed November 10, 2020.
3. U.S. Food and Drug Administration (FDA). Vaccines and Related Biological 
Products Advisory  Committee Meeting: FDA Briefing Document Pfizer -BioNTech 
COVID-19 Vaccine. Accessed December 10, 2020.
4. U.S. Food & Drug Administration (FDA). Pfizer COVID-19 Vaccine EUA Letter of 
Authorization reissued 12-23 -20. In:December 23, 2020.
5.U.S. Food and Drug Administration (FDA). FDA Takes Key  Action in Fight Against 
COVID-19 By Issuing Emergency  Use Authorization for First COVID- 19 Vaccine.  
https://www.fda.gov/news -events/press -announcements/fda -takes-key-action-fight-
against-covid-19- issuing-emergency -use-authorization -first-covid-19.
6.Pfizer Inc. Pfizer and BioNTech Achieve First Authorization in the World for a 
Vaccine to Combat COVID -19. https://www.pfizer.com/news/press -release/press-
release-detail/pfizer -and-biontech-achieve-first-authorization -world. Accessed 
January 4, 2021.
7.Pfizer Inc. Pfizer and BioNTech Receive Authorization in the European Union for 
COVID-19 Vaccine. https://www.pfizer.com/news/press -release/press- release-
detail/pfizer -and-biontech-receive-authorization- european- union. Accessed December 
28, 2020.
8.Yih WK, L ee GM, Lieu TA, et al. Surveillance for adverse events fol lowing receipt 
of pandemic 2009 H1N1 vaccine in the Post- Licensure Rapid Immunization Safety  
Monitoring (PRI SM) System, 2009 -2010. Am J Epidemiol. 2012;175(11):1120 -1128.
9.American Medical Association (AMA). Appendix Q: Severe Acute Respiratory 
Syndrome Coronavirus 2 (SARS -CoV-2) (coronavirus disease [COVID -19]) 
Vaccines. https://www.ama -assn.org/s ystem/files/2020-11/covid-19- immunizations -
appendix-q-table.pdf. Accessed January  12, 2021.
10.Centers for Medicare & Medicaid Services. COVID -19 Vaccines and Monoclonal 
Antibodies. 2020; https://www.cms.gov/medicare/medicare- part-b-drug-average-
sales-price/covid -19-vaccines- and-monoclonal -antibodies. Accessed January  14, 
2021.
11. Center for Disease Control (CDC). People with Certain Medical Conditions. 
Decembe r 29, 2020; https://www.cdc.gov/coronavirus/2019 -ncov/need -extra-
precautions/people -with-medical-conditions.html. Accessed January  4, 2021.
12.U.S. Department of Veterans Affairs. Veterans Health Administration. 2020; 
https://www.va.gov/health/. Accessed November 10, 2020.
13. U.S. Department of Veterans Affairs. VA nursing homes, assisted living, and home 
health care. 2020; https://www.va.gov/health -care/about -va-health-benefits/long -
term-care/. Accessed January  4, 2020.
14.U.S. Food and Drug Administrat ion (FDA). COVID -19 Vaccine Safet y Surveillance 
Project: Active Monitoring Protocol. December 16, 2020.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073268
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 60of 14415.Center for Disease Control (CDC) COVID -19 Response Team. Preliminary  Estimates 
of the Prevalence of Selected Underl ying Health Conditions Among Patients with 
Coronavirus Disease 2019 —United States, February  12–March 28, 2020. MMWR 
Morbidity and Mortality Weekly Report 2020;69(13):382 -386.
16.Gupta S, Hay ek SS, Wang W, et al. Factors Associated With Death in Critically  Ill 
Patients With Coronavirus Disease 2019 in the US. JAMA Intern Med. 
2020;180(11):1 -12.
17.Williamson EJ, Walker AJ, Bhaskaran K, et al. Factors associated with COVID -19-
related death using OpenSAFELY. Nature. 2020;584(7821):430-436.
18.Walsh EE, Frenck RW, Jr., Falsey  AR, et al. Safety  and Immunogenicity of Two 
RNA-Based Covid-19 Vaccine Candidates. N Engl J Med. 2020.
19.U.S. National Library of Medicine. Study  to Describe the Safet y, Tolerability, 
Immunogenicit y, and Efficacy of RNA Vaccine Candidates Against COVID -19 in 
Healthy Individuals. NCT04368728.
20. U.S. Food and Drug Administration (FDA) Center for Biologics Evaluation and 
Research (CBER). Emergency  Use Authorization for Vaccines to Prevent CO VID-
19: Guidance for Industry . https://www.fda.gov/media/142749/download Accessed 
November 10, 2020.
21. U.S. Food and Drug Administration (FDA) Center for Biologics Evaluation and 
Research (CBER). Development and L icensure of Vaccines to Prevent COVID -19; 
Guidance for Industry, June 2020. https://www.fda.gov/media/142749/download. 
Accessed November 10, 2020.
22.Lee GM, Greene SK, Weintraub ES, et al. H1N1 and seasonal influenza vaccine 
safety in the vaccine safety datalink project. Am J Prev Med. 2011;41(2):121 -128.
23.Fireman B, Lee J, L ewis N, Bembom O, van der Laan M, Baxter R. I nfluenza 
vaccination and mortality: differentiating vaccine effects from bias. Am J Epidemiol. 
2009;170(5) :650-656.
24.Virtanen M, Peltola H, Paunio M. Day -to-day reactogenicity  and the healthy  vaccinee 
effect of measles -mumps-rubella vaccination. Pediatrics. 2000;106(5):E62.
25.Greene SK, Kulldorff M, Yin R, et al. Near real -time vaccine safety  surveillance with 
partially accrued data. Pharmacoepidemiol Drug Saf. 2011;20(6):583 -590.
26.Keren R, Zaoutis TE, Bridges CB, et al. Neurological and Neuromuscular Disease as 
a Risk Factor for Respiratory  Failure in Children Hospitalized With I nfluenza 
Infection. JAM A. 2005;294(17):2188-2194.
27.U.S. Department of Veterans Affairs. VA priority  groups. 2020; 
https://www.va.gov/health- care/eligibility /priority-groups/.
28.Law B, Sturkenboom M. D2.3 Priority List of Adverse Events of Special Interest: 
COVID-19. 2020.
29.Shimabukuro T. CDC post -authorization/post -licensure safety  monitoring of COVID -
19 vaccines. National Center for Immunization &  Respiratory Diseases. 2020.
30.Liu CH, Yeh YC, Huang WT, Chie WC, Chan KA. Assessment of pre -specified 
adverse events follow ing varicella vaccine: A population- based self -controlled risk 
interval study . Vaccine. 2020;38(11):2495-2502.
31.Baker MA, Baer B, Kulldorff M, et al. Kawasaki disease and 13 -valent pneumococcal 
conjugate vaccination among young children: A self -controlled risk interval and 
cohort study  with null results. PLoS Med. 2019;16(7):e1002844.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073269
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 61of 14432.Gee J, Naleway A, Shui I, et al. Monitoring the safety  of quadrivalent human 
papillomavirus vaccine: findings from the Vaccine Safety  Datalink. Vaccine. 
2011;29(46):827 9-8284.
33.U.S. Department of Veterans Affairs. COVID -19 National Summary . 2020; 
https://www.accesstocare.va.gov/Healthcare/COVID19NationalSummary . Accessed 
November 10, 2020.
34.U.S. Department of Veterans Affairs. Racial and Ethnic Minority  Veterans. 2020; 
https://www.va.gov/HEALTHEQUITY/Race_Ethnicity .asp. Accessed January  18, 
2021.
35.U.S. Department of Veterans Affairs. COVID -19 vaccines at VA. 2020; 
https://www.va.gov/health- care/covid -19-vaccine/.
36.U.S. Department of Veterans Affairs. Profile o f Veterans: 2014. 2016; 
https://www.va.gov/vetdata/docs/SpecialReports/Profile_of_Veterans_2014.pdf. 
Accessed November 10, 2020.
37.U.S. Department of Veterans Affairs. Number of Veteran Patients b y Healthcare 
Priority Group: FY2000 to FY2017. 2018; 
https://www.va.gov/vetdata/docs/Utilization/Number_of_Veteran_Patients_by _HC_P
riority_Groups_2000_2017.pdf. Accessed November 10, 2020.
38.U.S. Department of Veterans Affairs. Average Expenditures Per Patient by 
Healthcare Priorit y Group: FY2000 to FY2013. 201 3; 
https://www.va.gov/vetdata/docs/Utilization/AvgCost_FINAL2.xlsx. Accessed 
November 10, 2020.
39.Agha Z, Lofgren RP, VanRuiswy k JV, Layde PM. Are patients at Veterans Affairs 
medical centers sicker? A comparative anal ysis of health status and medical re source 
use. Arch Intern Med. 2000;160(21):3252-3257.
40.Li R, Stewart B, McNeil MM, et al. Post licensure surveillance of influenza vaccines 
in the Vaccine Safet y Datalink in the 2013 -2014 and 2014 -2015 seasons. 
Pharmacoepidemiol Drug Saf. 2016;25(8):928 -934.
41.Kulldorff M, Davis RL, Kolczak† M, Lewis E, L ieu T, Platt R. A Maximized 
Sequential Probability  Ratio Test for Drug and Vaccine Safet y Surveillance. 
Sequential Analysis. 2011;30(1):58 -78.
42.Kulldorff M, Silva I R. Continuous Post -Market Sequentia l Safety Surveillance with 
Minimum Events to Signal. arXiv preprint arXiv. 2015(1503.01978).
43.When Are Risks Real? I n. Clinical Nutrition Insight. Vol 332007:6.
44.Silva IR, Lopes WM, Dias P, Yih WK. Alpha spending for historical versus 
surveillance Poisson data with CMaxSPRT. Stat Med. 2019;38(12):2126 -2138.
45.Silva IR. Type I error probability  spending for post -market drug and vaccine safet y 
surveillance with bin omial data. Stat Med. 2018;37(1):107-118.
46. Kulldorff M. A spatial scan statistic. Communications in Statistics - Theory and 
Methods. 2007;26(6):1481 -1496.
47. U.S. Food and Drug Administration (FDA) Center for Biologics Evaluation and 
Research (CBER). G uidance for Industry: Good Pharmacovigilance Practices and 
Pharmacoepidemiologic Assessment. 2005. Accessed November 12, 2020.
48.Baker MA, Lieu TA, L i L, et al. A vaccine study  design selection framework for the 
postlicensure rapid immunization safety  monitoring program. Am J Epidemiol. 
2015;181(8):608 -618.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073270
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 62of 14449. Hoffman KB, Demakas AR, Dimbil M, Tatonetti NP, Erdman CB. Stimulated 
reporting: the impact of US food and drug administration -issued alerts on the adverse 
event reporting s ystem (FAERS). Drug Saf. 2014;37(11):971 -980.
50.West AN, Charlton ME, Vaughan -Sarrazin M. Dual use of VA and non -VA hospitals 
by Veterans with multiple hospitalizations. BMC Health Serv Res. 2015;15:431.
51.Petersen LA, By rne MM, Daw CN, Hasche J, Reis B, Pietz K. Relationship between 
clinical conditions and use of Veterans Affairs health care among Medicare -enrolled 
veterans. Health Serv Res. 2010;45(3):762 -791.
52.U.S. Food and Drug Administration (FDA) Center for Drug Evaluation and Research 
(CDER). Best Practices for Condu cting and Reporting Pharmacoepidemiologic 
Safety Studies Using Electronic Healthcare Data. 2013; 
https://www.fda.gov/media/79922/download . Accessed January  19, 2021.
53.International Societ y for Pharmacoepidemiology  (ISPE). Guidelines for good 
pharmacoepi demiology  practices (GPP). Pharmacoepidemiol Drug Saf. 
2008;17(2):200 -208.
54. International Epidemiological Association (IEA). Good Epidemiological Practice 
(GEP), IEA Guidelines for Proper Conduct of Epidemiological Research. 2007; 
https://ieaweb.org/IEA Web/Content/I EA_Publications.aspx . Accessed January  25, 
2021.
55.Centers for Medicare & Medicaid Services. 2021 ICD -10-CM Official Guidelines for 
Coding and Reporting.  https://www.cms.gov/files/document/2021 -coding-
guidelines -updated-12162020.pdf . Accessed January  17, 2021.
56.Baxter R, Eaton A, Hansen J, Aukes L , Caspard H, Ambrose CS. Safet y of 
quadrivalent live attenuated influenza vaccine in subjects aged 2 -49years. Vaccine. 
2017;35(9):1254 -1258.
57.Johns Hopkins Vasculitis Center. Ty pes of Va sculitis.  
https://www.hopkinsvasculitis.org/ty pes-
vasculitis/#:~:text=%E2%80%9CAngiitis%E2%80%9D%20and%20%E2%80%9CAr
teritis%E2%80%9D,lit'%20i%20deez%E2%80%9D . Accessed January  19, 2021.
58.OptumInsight Inc. Guide to Clinical Validation, Documentation and Coding: Acute 
Kidney Injury.  
https://www.optum360coding.com/upload/pdf/ECDCG14/CDCG14_v2.pdf . 
Accessed January  17, 2021.
59. U.S. Department of Health and Human Services. Common Terminology Criteria for 
Adverse Events (CTCAE). 2017; 
https://ctep.cancer.g ov/protocoldevelopment/electronic_applications/docs/CTCAE_v
5_Quick_Reference_5x7.pdf . Accessed January  17, 2021.
60.Forns J, Cainzos -Achirica M, Hellfritzsch M, et al. Validity  of ICD-9 and ICD -10 
codes used to identify  acute liver injury : A study in three European data sources. 
Pharmacoepidemiol Drug Saf. 2019;28(7):965 -975.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073271
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 63of 14414.LIST OF TABLES
Table 1. Outcome algorithms for SCRI  analysis, with risk and control intervals ................... 36
Table2. Estimated Statistical Power for the Poisson -based MaxSPRT41............................... 41
Table 3. Critical Values for Poisson -based MaxSPRT................................ ............................ 49
Appendix Table 1. Demographic and Clinical Characteristics Definitions ..Error! Bookmark 
not defined.
Appendix Table 2. Operational Definitions of Safety  Events of Interest.....Error! Bookmark 
not defined.
15.LIST OF FIGURES
Figure 1. Example of SCRI  Design for Assessment of a Safet y Event of Interest with a 
42-day Risk Interval in an I ndividual who Receives Only  One 
Vaccine Dose, Showing Both Pre- and Post-vaccination Control 
Intervals................................ ................................ ................................ ....26
Figure 2. Example of SCRI  Design with Overlapping Risk Intervals when Two Doses 
of Pfizer-BioNTech COVID- 19 Vaccine are Administered, 
Showing a Pre-and Post-vaccination Control Interval ............................. 27
Figure 3. Steps in Signal Detection, Evaluation, and Verification ................................ ..........45
Figure 4. Example of SCRI  Design for a Safet y Event of Interest with a 42 -day Risk 
Interval and a Pre -vaccination Control I nterval................................ ........47
Figure 5. Example of Risk (P1, P2, P3) and Pre- vaccination Control I ntervals for the 
SCRI End-of-surveillance Anal yses of 1 or 2 D oses of Pfizer -
BioNTech COVID -19 Vaccine ................................ ................................ .53
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073272
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 64of 14416.ANNEX 1. LIST OF STAND ALONE DOCUMENTS
None. 
17.ANNEX 2. ENCEPP CHEC KLIST FOR STUDY PROT OCOLS
N/A
18.ANNEX 3. ADDITIONAL INFORMATI ON
Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Demographic Characteristics
Age Continuous variable;
Dichotomous variable: 
18-64
>65;
Categorical variable:
<35
35 -<45
45 -<55
55 -<65
65 -<75
≥75Age as of the date prior toPfizer-
BioNTech COVID -19 vaccination 
(and/or date prior to  seasonal 
influenza vaccination for active 
comparators)
Sex Categorical variable: 
Male
Female
Unknown
Race/ethnicit yCategorical variable:
White
Asian or Pacific 
Islander
Black
American Indian or 
Alaskan native
Other
Unknown
State Geographic regions in the US State of residence
Clinical Characteristics
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073273
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 65of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Smoking Dichotomous variable Defined b ythe “tobacco” variable. 
‘Y’ indicates the person is a tobacco 
user
ICD-9-CM codes:
305.1, Tobacco use disorder
V15.82, History  of tobacco 
use
ICD-10-CM codes:
F17.200, Nicotine 
dependence, unspecified, 
uncomplicated
Z7.20, Tobacco use
Z87.891, Personal history  of 
nicotine dependence
Body mass index 
(BMI)Continuous variable; 
Categorical variable:
Underweight (<18.5) 
Normal weight (18.5 -
24.9) 
Overweight (25 -29.9)
Obese (≥30 -<40)
Severe obesit y (>40)Calculated from height and weight 
data (kg/m2) 
ICD-9-CM codes:
V85.0, Bod y Mass Index less 
than 19, adult
V85.1, Bod y Mass Index 
between 19- 24, adult
V85.2, Bod y mass index 
between 25- 29, adult
V85.3, Bod y mass index 
between 30- 39, adult
V85.4, Bod y mass index 40 
and over, adult
ICD-10-CM codes:
Z68.1,Body Mass Index 
19.9 or less, adult
Z68.2, Body mass index 20-
29, adult
Z68.3, Bod y mass index 
between 30- 39, adult
Z68.4, Bod y mass index 40 
and over, adult
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073274
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 66of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
History of 
anaphylaxis/allergic 
reactionsDichotomous variable ICD-9-CM code:
V13.81, Personal history  of 
anaphylaxis
V14.0 -V14.6, V14.8, 
V14.9, Personal history  of 
allergy to drugs, medications 
and biological substances, 
excluding serum and vaccine
V15.0x, Other allergy
525.66, Allergy  to existing 
dental restorative material
995.0, Other anaph ylactic 
shock, not elsewhere 
classified
995.1, Angioneurotic edema, 
not elsewhere classified
995.21, Arthus phenomenon
999.27, Other drug allergy
995.3, Allergy , unspecified, 
not elsewhere classified
995.6x, Anaphy lactic shock 
due to food
999.41, Anaph ylactic
reaction due to 
administration of blood and 
blood products
999.49, Anaph ylactic 
reaction due to other serum 
ICD-10-CM code:
Z87.892  Personal history of 
anaphylaxis
Z88.0 -Z88.6, Z88.8, Z88.9, 
Allergy status to drugs, 
medications and biological 
substances, excluding serum 
and vaccine
T78.00xx -T78.09xx, 
Anaphylactic reaction due to 
food, initial encounter, 
subsequent encounter and 
sequela
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073275
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 67of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
T78.2xxx, Anaphy lactic 
shock, initial encounter, 
subsequent encounter and 
sequela
T78.3xxx, Angioneurotic 
edema, initial encounter, 
subsequent encounter and 
sequela
T78.41xx, Arthus 
phenomenon
T80.51xx, Anaphy lactic 
reaction due to 
administration of blood and 
blood products, initial 
encounter, subsequent 
encounter and sequela
T80.59xx, Anaphy lactic 
reaction due to other s erum,  
initial encounter, subsequent 
encounter and sequela
T88.6xxx, Anaphy lactic 
reaction due to adverse effect 
of correct drug or 
medicament properl y 
administered, initial 
encounter, subsequent 
encounter and sequela
Previous anaph ylaxis 
of vaccine 
componentDichotomous variable ICD-9-CM code:
999.42, Anaph ylactic 
reaction due to vaccination
V14.7, Personal history  of 
allergy to serum or vaccine
ICD-10-CM codes:
T80.52xx, Anaphy lactic 
reaction due to vaccination, 
initial encounter, subsequent 
encounter and sequela
Z28.04, Immunization not 
carried out because of patient 
allergy to vaccine or 
component
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073276
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 68of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Z88.7, Allergy  status to 
serum and vaccine
History of 
hospitalizations Dichotomous variable;
Continuous variableDefined b y having any  
hospitalizations (dichotomous) and 
number of hospitalizations 
(continuous)
Charlson 
Comorbidity  Index 
(CCI)Continuous variable ICD-9-CM codes:
410.x, 412.x, My ocardial 
infarction
398.91, 402.01, 402.11, 
402.91, 404.01, 404.03, 
404.11, 404.13, 404.91, 
404.93, 425.4 -425.9, 428.x, 
Congestive heart failure
093.0, 437.3, 440.x, 441.x, 
443.1 -443.9, 447.1, 557.1, 
557.9, V43.4, Peripheral 
vascular disease
362.34, 430.x -438.x, 
Cerebrovascular disease
290.x, 294.1, 331.2, 
Dementia
416.8, 416.9, 490.x -505.x, 
506.4, 508.1, 508.8, Chronic 
pulmonary  disease
446.5, 710.0 -710.4, 714.0 -
714.2, 714.8, 725.x, 
Rheumatic disease
531.x -534.x, Peptic ulcer 
disease
070.22, 070.23, 070.32, 
070.33, 070.44, 070.54, 
070.6, 070.9, 570.x, 571.x, 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073277
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 69of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
573.3, 573.4, 573.8, 573.9, 
V42.7, Mild liver disease
250.0 -250.3, 250.8, 250.9, 
Diabetes without chronic 
complication
250.4 -250.7, Diabetes with 
chronic complication
334.1, 342.x, 343.x, 344.0 -
344.6,344.9, Hemiplegia or 
paraplegia
403.01, 403.11, 403.91, 
404.02, 404.03, 404.12, 
404.13, 404.92, 404.93, 
582.x, 583.0 -583.7, 585.x, 
586.x, 588.0, V42.0, V45.1, 
V56.x, Renal disease
140.x -172.x, 174.x -195.8, 
200.x -208.x, 238.6, Any 
malignancy , including 
lymphoma and leukemia, 
except malignant neoplasm 
of skin
456.0 -456.2, 572.2 -572.8, 
Moderate or severe liver 
disease
196.x -199.x, Metastatic 
solid tumor
042.x -044.x, Acquired 
immunodeficiency  syndrome 
(AIDS)/Human 
immunodeficiency  virus 
(HIV)
ICD-10-CM codes:
I21.x, I21.xx, I 22.x, I25.2, 
Myocardial infarction
I09.9, I11.0, I13.0, I13.2, 
I25.5, I42.0, 
I42.5 -I42.9, I43, I43.x, 
I50.x, I50.xx, Congestive 
heart failure
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073278
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 70of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
I70.x, I71.x, I 73.1, I73.8, 
I73.9, I77.1, I79.0, I79.2, 
K55.1, K55.8, K55.9, Z95.8, 
Z95.9, Peripheral vascular 
disease
G45, G45.x, G46.x, H34.0, 
I60.x -I63.x, I60.xx -I63.xx, 
I60.xxx - I63.xxx, I65.x -
I69.x, I65.xx - I69.xx, 
I65.xxx - I69.xxx, 
Cerebrovascular disease
F00.x -F03.x, F00.xx -
F03.xx, F05, F05.1, G30.x, 
G31.1, Deme ntia
I27.8, I27.9, J40.x - J47.x, 
J40.xx -J47.xx, J40.xxx -
J47.xxx, J60.x -J67.x, J68.4, 
J70.1, J70.3, Chronic 
pulmonary  disease
M05, M05.x, M05.xx, 
M05.xxx, M06, M06.x, 
M06.xx, M06.xxx, M31.5, 
M32.x -M34.x, M32.xx -
M34.xx,  M35.1, M35.3, 
M36.0, Rheum atic disease
K25.x -K28.x, Peptic ulcer 
disease
B18.x, K70.0 - K70.3, 
K70.9, K71.3 - K71.5, 
K71.7, K73.x, K74.x, 
K74.xx, K76.0, K76.2 -
K76.4, K76.8, K76.9, Z94.4, 
Mild liver disease
E10.0, E10.1x, E10.6x, 
E10.6xx, E10.8, E10.9, 
E11.0x, E11.1x, E11.6x, 
E11.6xx, E11.8, E11.9, 
E12.0, E12.1, E12.6, E12.8, 
E12.9, E13.0x, E13.1x, 
E13.6x, E13.6xx, E13.8, 
E13.9, E14.0, E14.1, E14.6, 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073279
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 71of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
E14.8, E14.9, Diabetes 
without chronic complication
E10.2x -E10.5x, E10.2xx -
E10.5xx, E10.7, E11.2x -
E11.5x, E11.2xx - E11.5xx, 
E11.7, E12.2 - E12.5, E12.7, 
E13.2 -E13.5x, E13.7, E14.2 
-E14.5, E14.7, Diabetes 
with chronic complication
G04.1, G11.4, G80.1, G80.2, 
G81.x, G81.xx, G82.x, 
G82.xx, G83.0, G83.1 -
G83.3, G83.1x -G83.3x, 
G83.4, G83.9, Hemiplegia or 
paraplegia
I12.0, I13.1x, N03.2 - N03.7, 
N05.2 -N05.7, N18.x, N19, 
N25.0, Z49.0x -Z49.3x, 
Z94.0, Z99.2, Renal disease
C00-C75, C00.x -C75.x, 
C00.xx-C75.xx (excluding 
C44, C44.x and C44.xx), 
C7A., C7A.x, C7A.xx, C7B., 
C7B.x, C7B.xx, C76- C80, 
C76.x-C80.x, C76.xx -
C80.xx, C81- C96, C81.x-
C96.x, C81.xx -C96.xx, Any 
malignancy , including 
lymphoma and leukemia, 
except malignant neoplasm 
of skin
I85.0, I85.9, I86.4, I98.2, 
K70.4x, K71.1x, K72.1x, 
K72.9x, K76.5, K76.6, 
K76.7, Moderate or severe 
liver disease
C77.x -C80.x, C77.xx -
C80.xx, Metastatic solid 
tumor
B20, B97.35, AIDS/HIV
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073280
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 72of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Comorbidities Categorical variable:
Autoimmune 
disease 
Asthma
Bleeding diathesis 
or condition 
associated with 
prolonged bleeding
Cancer
Cardiovascular 
conditions ( e.g., 
heart failure, CAD, 
cardiomyopathies)
Chronic kidney  
disease/dial ysis
COPD/interstitial 
lung disease
Diabetes mellitus 
(ie, Type 2 diabetes)
Down syndrome
Sickle cell disease
HBV
HCV
HIV
Hyperlipidemia
Hypertension
Liver disease
Neurological 
disease
Other immune 
deficiencies
Solid organ 
transplant
VTEAutoimmune disease 
(immunocompromised state 
[weakened immune s ystem] from 
solid organ transplant):
ICD-9-CM  codes:
245.2, Chronic 
lymphocytic thyroiditis
340, Multiple sclerosis
357, Acute infective 
polyneuritis
357.4, Poly neuropathy  in 
other diseases classified 
elsewhere
696.1, Other psoriasis
694.3, Impetigo 
herpetiformis
696.1, Other psoriasis
696, Psoriatic 
arthropath y
695.4, Lupus 
erythematosus
714, 714.x, 714.xx, 
Rheumatoid arthritis and 
other inflammatory  
polyarthropathies
359.6, Symptomatic 
inflammatory  myopathy 
in diseases classified 
elsewhere
357.1, Poly neuropathy  in 
collagen vascular disease
714.89, Other specified 
inflammatory  
polyarthropathies
714.9, Unspecified 
inflammatory  
polyarthropathy
446.5, Giant cell arteritis
710.2, Sicca syndrome
ICD-10-CM codes:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073281
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 73of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
D69.3, Immune 
thrombocytopenic 
purpura
E06.3, Autoimmune 
thyroiditis
G35, MS
G61.0 and G65.0, GBS 
and sequelae of GBS
L40.x, L40.5x, Psoriasis
L93.x, Lupus 
erythematosus
M05.x, M05.xx, 
M05.xxx, Rheumatoid 
arthritis with rheu matoid 
factor
M06.x, M06.xx, 
M06.xxx, Other 
rheumatoid arthritis
M31.5, M31.6, Giant cell 
arteritis
M35.0x, Sicca 
(Sjogren’s) s yndrome 
E10, E10.x, E10.xx, 
Type 1 diabetes mellitus
N05.9, 
Glomerulonephritis
D84.9, 
Immunodeficiency , 
unspecified
Asthma: 
ICD-9-CM codes:
o493.xx, Asthma
ICD-10-CM codes:
oJ45.2x -J45.3x, 
Mild intermittent 
asthma
oJ45.4x, Moderate 
persistent asthma
oJ45.5x, Severe 
persistent asthma
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073282
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 74of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oJ45.9x, Other and 
unspecified 
asthma
Bleeding diathesis or condition 
associated with prolonged bleeding:
ICD-9-CM codes:
o286.x, 
Coagulation 
defects
o289.8x, Other 
specified diseases 
of blood and 
blood-forming 
organs
o287, 287.x, 
287.xx,  Purpura 
and other 
hemorrhagic 
conditions
ICD-10-CM codes:
oD65, 
Disseminated 
intravascular 
coagulation
oD66, Heredita ry 
factor VIII 
deficiency
oD67, Hereditary  
factor IX 
deficiency
oD68, D68.x, 
D68.xx, Other 
coagulation 
defects
oD69, D69.x, 
D69.xx, Purpura 
and other 
hemorrhagic 
conditions
Cancer:
ICD-9-CM codes:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073283
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 75of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o140.x -149.x, 
Malignant 
neoplasm of lip, 
oral cavit y, and 
pharynx
o150.x -159.x, 
Malignant 
neoplasm of 
digestive organs 
and peritoneum
o160.x -165.x, 
Malignant 
neoplasm of 
respiratory  and 
intrathoracic 
organs
o170.x -176.x, 
Malignant 
neoplasm of 
bone, connective 
tissue, skin, and 
breast
o179.x -189.x, 
Malignant 
neoplasm of 
genitourinary  
organs 
o190.x -199.x, 
Malignant 
neoplasm of other 
unspecified sites
o200.xx -208.xx, 
Malignant 
neoplasm of 
lymphatic and 
hematopoietic 
tissue
o209.0x -209.3x, 
Malignant 
neuroendocrine 
tumors
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073284
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 76of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o230.x -234.x, 
Carcinoma i n situ 
of digestive 
organs
ICD-10-CM codes:
oC00-C75, C00.x -
C75.x, C00.xx -
C75.xx, C7A., 
C7A.x, C7A.xx, 
C7B., C7B.x, 
C7B.xx, 
Malignant 
neoplasms, stated 
or presumed to be 
primary (of 
specified sites), 
and certain 
specified 
histologies, 
except 
neuroendocrine,
and of lymphoid, 
hematopoietic 
and related tissue 
oC76-C80, C76.x -
C80.x, C76.xx -
C80.xx, 
Malignant 
neoplasms of ill-
defined, other 
secondary  and 
unspecified sites
oC81-C96, C81.x -
C96.x, C81.xx -
C96.xx, 
Malignant 
neoplasms of 
lymphoid, 
hematopoietic 
and related tissue 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073285
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 77of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Cardiovascular conditions ( e.g., 
heart failure, coronary  artery disease 
[CAD], cardiomy opathies):
ICD-9-CM codes:
o428.xx, Heart 
failure
o414.01, 429.2, 
411.1, 413.9, 
414.11, 414.12, 
414.05, 414.02, 
414.04, 414.03, 
414.06, 414.07, 
414.2, 411.81, 
411.89, CAD
o425.xx, 
Cardiomy opathy
ICD-10-CM codes:
o150.x, 150.xx,  
Heart failure
oI24.0, I24.8, 
I24.9, I25.10, 
I25.110, I25.111, 
I25.118, I25.119, 
I25.41, I25.42, 
I25.700, I25.701, 
I25.708, I25.709, 
I25.710, I25.711, 
I25.718, I25.719, 
I25.720, I25.721, 
I25.728, I25.729, 
I25.730, I25.731, 
I25.738, I25.739, 
I25.750, I25.751, 
I25.758, I25.759, 
I25.760, I25.761, 
I25.768, I25.769, 
I25.790, I25.791, 
I25.798, I25.799, 
I25.810, I25.811, 
I25.812, CAD
oI42.x, 
Cardiomy opathy
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073286
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 78of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Chronic kidney  disease/dialy sis:
ICD-9-CMcodes:
o283.11, 
Hemolytic-
uremic syndrome
o403, 403.x, 
403.xx, 
Hypertensive 
chronic kidney  
disease
o404, 404.x, 
404.xx, 
Hypertensive 
heart and chronic 
kidney disease
o440.1, 
Atherosclerosis of 
renal artery
o442.1, Aneury sm 
of renal artery
o572.4, 
Hepatorenal 
syndrome
o274.1, Gouty  
nephropath y, 
unspecified
o710, Systemic 
lupus 
erythematosus
o710.2, Sicca 
syndrome
o580, 580.x, 
580.xx, Acute 
glomerulonephriti
s
o581.x, 581.xx, 
Nephrotic 
syndrome
o582, 582.x, 
582.xx, Chronic 
glomerulonephriti
s
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073287
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 79of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o583, 583.x, 
583,xx, Nephritis
and nephropath y, 
not specified as 
acute or chronic
o591, 
Hydronephrosis
o593.3, Stricture or 
kinking of ureter
o592, Calculus of 
kidney
o592.1, Calculus 
of ureter
o590.9, Infection 
of kidney , 
unspecified
o584.x, Acute 
kidney failure
o585.x, Chronic 
kidney disease
o588.x, 588.xx, 
Disorders 
resulting from 
impaired renal 
function
o587, Renal 
sclerosis, 
unspecified
o753.1x, Cy stic 
kidney disease
o753.2, 753.2x, 
Obstructive 
defects of renal 
pelvis and ureter
ICD-10-CM codes:
oD59.3, 
Hemolytic-
uremic syndrome
oI12.x, 
Hypertensive 
chronic kidney  
disease
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073288
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 80of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oI13.x, I13.xx, 
Hypertensive 
heart and chronic 
kidney disease
oI70.1, 
Atherosclerosis of 
renal artery
oI72.2 Aneury sm 
of renal artery
oK76.7, 
Hepatorenal 
syndrome
oM10.30-M10.39, 
M10.30x-
M10.37x, Gout 
due to renal 
impairment
oM32.14,
Glomerular 
disease in 
systemic lupus 
erythematosus
oM32.15, Tubulo -
interstitial 
nephropath y in 
systemic lupus 
erythematosus
oM3504, Sicca 
syndrome with 
tubulo-interstitial 
nephropath y
oN00.x-N07.x, 
N08, Glomerular 
diseases
oN13.1, N13.2, 
N13.3x, 
Obstructive a nd 
reflux uropathy
oN14.x, 
Nephropath y
oN15.x, Other 
renal tubulo -
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073289
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 81of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
interstitial 
diseases
oN16, Renal 
tubulo-interstitial 
disorders in 
diseases classified 
elsewhere
oN17.x, N18.x, 
N19, Acute 
kidney failure and 
chronic kidney  
disease
oN25.x, N26.x, 
N25.xx, Other 
disorders of 
kidney and ureter
oQ61.02, Q61.11x, 
Q61.2-Q61.9, 
Cystic kidney  
disease
oQ62.x, Q62.xx, 
Congenital 
obstructive 
defects of renal 
pelvis and 
congenital 
malformation of 
ureter
COPD/interstitial lung disease:
ICD-9-CM codes:
o491.9, 
Unspecified 
chronicbronchitis
o492.8, Other 
emphysema
o491.x, 491.xx, 
Chronic 
bronchitis
o493.2, Chronic 
obstructive 
asthma, 
unspecified
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073290
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 82of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o496, Chronic 
airway 
obstruction, not 
elsewhere 
classified
o516, 516.x, 
516.xx, Other 
alveolar and 
parietoalveolar 
pneumonopathy
o515, 
Postinflammatory  
pulmonary  
fibrosis
o518.x, 518.xx, 
Other diseases of 
lung
o714.81, 
Rheumatoid lung
ICD-10-CM codes:
oJ41.x Simple and 
mucopurulent 
chronic bronchitis
oJ42, Unspecified 
chronic bronchitis
oJ43.x, 
Emphysema
oJ44.x, Other 
COPD
oJ80, J81.x, 
J82.xx, J84.xx, 
J84.xxx, Other 
respiratory  
diseases 
principally  
affecting the 
interstitium 
oM05.10, 
Rheumatoid lung 
disease with 
rheumatoid 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073291
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 83of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
arthritis of 
unspecified site
Diabetes mellitus (ie, Type 2 
diabetes):
ICD-9-CM codes:
o250.xx, Diabetes 
mellitus
ICD-10-CM codes:
oE11.x, E11.xx, 
E11.xxx, Ty pe 2 
diabetes mellitus
Down syndrome:
ICD-9-CM codes:
o758.x, Down 
syndrome
ICD-10-CM codes:
oQ90.x, Down 
syndrome
Sickle cell disease:
ICD-9-CM codes:
o282.xx, Sickle -
cell disease
ICD-10-CM codes:
oD57, D57.x, 
D57.xx, D57.xxx,
Sickle-cell 
disorders
HBV:
ICD-9-CM codes:
o70.33, Chronic 
viral hepatitis B 
without mention 
of hepatic coma 
with hepatitis 
delta
o70.32, Chronic 
viral hepatitis B 
without mention 
of hepatic coma 
without mention 
of hepatitis delta
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073292
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 84of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o70.3, Viral 
hepatitis B 
without mention 
of hepatic coma, 
acute or 
unspecified, 
without mention 
of hepatitis delta
o70.2, Viral 
hepatitis B with 
hepatic coma, 
acute or 
unspecified, 
without mention 
of hepatitis delta
ICD-10-CM codes:
oB18.0, B18.1, 
Chronic viral 
hepatitis B
oB19.1, B1 9.1x, 
Unspecified viral 
hepatitis B
HCV:
ICD-9-CM codes:
o70.7, Unspecified 
viral hepatitis C 
without hepatic 
coma
o70.71, 
Unspecified viral 
hepatitis C with 
hepatic coma
o70.54, Chronic 
hepatitis C 
without mention 
of hepatic coma
ICD-10-CM codes:
oB18.2, Chronic 
viral hepatitis C
oB19.2x, 
Unspecified viral 
hepatitis C
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073293
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 85of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
HIV:
ICD-9-CM codes:
o42, HIV disease
o79.53, HIV t ype 2 
ICD-10-CM codes:
oB20, HIV disease
oB97.35, HIV t ype 
2 as the cause of 
diseases classified 
elsewhere
Hyperlipidemia
ICD-9-CM codes:
o272.0x, Pure 
hypercholesterole
mia
o272.1x, Pure 
hyperglyceridemi
a
o272.2x, Mixed 
hyperlipidemia
o272.4x, 
Hyperlipidemia, 
NOS
ICD-10-CM codes:
oE78.0-E78.5, 
E78.0x, E78.4x, 
Hyperlipidemia
Hypertension:
ICD-9-CM codes:
o401.1, Benign 
essential 
hypertension
o401.9, Essential 
hypertension, 
NOS
o405.1, Benign 
secondary  
hypertension
o405.9, Secondary  
hypertension, 
NOS
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073294
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 86of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o997.91, 
Hypertension, 
NOS
ICD-10-CM codes:
oH35.03x, 
Hypertensive 
retinopathy
oI10, I11.x -I16.x, 
I13.xx, 
Hypertensive 
diseases
oI67.4, 
Hypertensive 
encephalopa thy 
diseases
Liver disease:
ICD-9-CM codes:
o571, 571.x, 
Alcoholic fatt y 
liver
o572, 572.x, 
Hepatic 
encephalopath y
o573.x, Other 
disorder of liver
o570, Acute and 
subacute necrosis 
of liver
ICD-10-CM codes:
oK70.x, K70.xx, 
Alcoholic fatt y 
liver
oK71.x, K71.xx, 
Toxic liver 
disease
oK72.xx, Hepatic 
failure, not 
elsewhere 
classified
oK73.x, Chronic 
hepatitis, not 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073295
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 87of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
elsewhere 
specified
oK74.x, K74.xx, 
Fibrosis and 
cirrhosis of liver
oK75.x, K75.xx, 
Other 
inflammatory  
liver diseases
oK76.x, K76.xx, 
Other diseases of 
liver
oK77, Liver 
disorders in 
diseases classified 
elsewhere
Neurological disease:
ICD-9-CM codes:
o780.97, Altered 
mental status
o780.93, Memory  
loss
o781.8, Neurologic 
neglect syndrome
o797, Senility  
without mention 
of psychosis
oV62.89, Other 
psychological or 
physical stress, 
not elsewhere 
classified
o799.5x, Signs and 
symptoms 
involving 
cognition
o780.99, Other 
general symptoms
o780.4, Dizziness 
and giddiness
o781.1, 
Disturbances of 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073296
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 88of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
sensation of smell 
and taste
oV41.5, Problems 
with smell and 
taste
o368.16, 
Psychophysical
visual 
disturbances
o307.9, Other and 
unspecified 
special symptoms 
or syndromes, not 
elsewhere 
classified
o300.9, 
Unspecified 
nonpsychotic 
mental disorder
o300.9, 
Unspecified 
nonpsychotic 
mental disorder
o308.9, 
Unspecified acute 
reaction to stress
o307.9, Oth er and 
unspecified 
special symptoms 
or syndromes, not 
elsewhere 
classified
oV62.85, 
Homicidal 
ideation
oV62.84, Suicidal 
ideation
o799.24, 
Emotional lability
o799.23, 
Impulsiveness
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073297
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 89of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o799.29, Other 
signs and 
symptoms 
involving 
emotional state
oV40.39, Other 
specified 
behavioral 
problem
ICD-10-CM codes:
oR41, R41.x, 
R41.xx, Other 
symptoms and 
signs involving 
cognitive 
functions and 
awareness
oR42, Dizziness 
and giddiness
oR43, R43.x, 
Disturbances of 
smell and taste
oR44, R44.x, 
Other symptoms 
and signs 
involving general
sensations and 
perceptions
oR45, R45.x, R45. 
xx, Symptoms 
and signs 
involving 
emotional state
oR46, R46.x, R46. 
xx, Symptoms 
and signs 
involving 
appearance and 
behavior
Other immune deficiencies:
ICD-9-CM codes:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073298
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 90of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o279.x, 279.xx, 
Deficiency  of 
humoral 
immunity
o135, Sarcoidosis
o273.x, Disorders 
of plasma protein 
metabolism
ICD-10-CM codes:
oD80, D80.x, 
Immunodeficienc
y with 
predominantly  
antibody defects
oD81, D81.x, 
D81.xx, 
Combined 
immunodeficienci
es
oD82, D82.x, 
Immunodeficienc
y associated with 
other major 
defects
oD83, D83.x, 
Common variable 
immunodeficienc
y
oD84, D84.x, 
D84.xx, Other 
immunodeficienci
es
oD86, D86.x, 
D86.xx, 
Sarcoidosis
oD89, D89.x, 
D89.xx, Other 
disorders 
involving the 
immune 
mechanism, not 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073299
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 91of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
elsewhere 
classified
Solid organ transplant:
CPT codes: 
o32850-32856, 
Transplantation 
of lung
o33930-33945, 
Transplantation 
of heart
o44132, 44133, 
47133, 47135, 
47140-47147, 
Transplantation 
of liver
o44135-44137, 
44715, 44720, 
44721, 
Transplantation 
of intestine
o48160, 48550 -
48552, 48554, 
48556, 
Transplantation 
ofpancreas
o50300, 50320, 
50323, 50325, 
50327, 50328, 
50329, 50340, 
50340, 50360, 
50365, 50370, 
50380, Renal 
transplantation
ICD-9-PCS codes:
o00.91 -00.93, 
Transplant from 
donor or cadaver
o37.51, Heart 
transplantation
o33.51, Unilateral 
lung 
transplantation
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073300
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 92of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o33.52, Bilateral 
lung 
transplantation
o46.97, Transplant 
of intestine
o50.59, Other 
transplant of 
intestine
o52.82, 
Homotransplant 
of pancreas
o55.69, Other 
kidney transplant
ICD-10-PCS codes:
o02YA0Z0, 
02YA0Z1, 
Transplantation 
of heart
o0BYC0Z0, 
0BYC0Z1, 
0BYD0Z0, 
0BYD0Z1, 
0BYF0Z0, 
0BYF0Z1, 
0BYG0Z0, 
0BYG0Z1, 
0BYH0Z0, 
0BYH0Z1, 
0BYJ0Z0,  
0BYJ0Z1,  
0BYK0Z0,  
0BYK0Z1,  
0BYL0Z0, 
0BYL0Z1, 
0BYM0Z0, 
0BYM0Z1, 
Transplantation 
of lung
o0DY60Z0, 
0DY60Z1, 
Transplantation 
of stomach
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073301
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 93of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o0DY80Z0, 
0DY80Z1, 
Transplantation 
of small i ntestine
o0DYE0Z0, 
0DYE0Z1, 
Transplantation 
of large intestine
o0FY00Z0, 
0FY00Z1, 
Transplantation 
of liver
o0FYG0Z0, 
0FYG0Z1, 
Transplantation 
of pancreas
o0TY00Z0, 
0TY00Z1, 
0TY10Z0, 
0TY10Z1, 
Transplantation 
of kidney
VTE:
ICD-9-CM codes:
o415.1x, 
Pulmonary  
embolism and 
infarction
o451.x, 451.xx, 
Phlebitis and 
thrombophlebitis
o452, Portal vein 
thrombosis
o453.x, 453.xx, 
Other venous 
embolism and 
thrombosis
ICD-10-CM codes:
oI26, I26.x, I26.xx, 
Pulmonary  
embolism
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073302
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 94of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oI80, I80.x, I80.xx, 
I80.xxx, Phlebitis 
and 
thrombophlebitis
oI81, Portal vein 
thrombosis
oI82, I82.x, I82.xx, 
I82.xxx Other 
venous embolism 
and thrombosis
Concurrent 
immunizationsCategorical variable:
Seasonal influenza 
Tetanus diphtheria and 
pertussis (Tdap or Td)
Chickenpox (Varicella)
Shingles (Herpes Zoster 
recombinant and/or 
live)
Human papillomavirus 
(HPV)  
Pneumococcal 
conjugate
Pneumococcal 
polysaccharide
Hepatitis A
Hepatitis B
Meningococcal 
conjugate (MenACWY) 
and serogroup B 
meningococcal (MenB)
Haemophilus influenza 
type bDescription of immunization, 
immunization I D, lot number, and 
manufacturer code will be available.
Seasonal influenza:
CPT codes:
o90653, Influenza 
vaccine, inactivated 
(IIV), subunit, 
adjuvanted, for 
intramuscular use
o90724, Influenza virus 
vaccine
o90662, Influenza virus 
vaccine (IIV), split virus, 
preservative free, 
enhanced 
immunogenicit y via 
increased antigen 
content, for 
intramuscular use
o90662, Influenza virus 
vaccine (IIV), split virus, 
preservative free, 
enhanced 
immunogenicit y via 
increased antigen 
content, f or 
intramuscular use
o90694, Influenza virus 
vaccine, quadrivalent 
(aIIV4), inactivated, 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073303
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 95of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
adjuvanted, preservative 
free, 0.5 mL dosage, for 
intramuscular use
o90756, Influenza virus 
vaccine, quadrivalent 
(ccIIV4), derived from 
cell cultures, subunit, 
antibiotic free, 0.5 mL 
dosage, for intramuscular 
use
o90674, Influenza virus 
vaccine, quadrivalent 
(ccIIV4), derived from 
cell cultures, subunit, 
preservative and 
antibiotic free, 0.5 mL 
dosage, for intramuscular 
use
o90688, Influenza virus 
vaccine, quadrivalent 
(IIV4), split virus, 0.5 
mL dosage, for 
intramuscular use
o90686, Influenza virus 
vaccine, quadrivalent 
(IIV4), split virus, 
preservative free, 0.5 mL 
dosage, for intramuscular 
use
o90630, Influenza virus 
vaccine, quadrivalent 
(IIV4), split virus, 
preservative free, for 
intradermal use
o90682, Influenza virus 
vaccine, quadrivalent 
(RIV4), derived from 
recombinant DNA, 
hemagglutinin (HA) 
protein only , preservative 
and antibiotic free, for 
intramuscular use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073304
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 96of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o90672, Influenza virus 
vaccine, quadrivalent, 
live (LAIV4), for 
intranasal use
o90661, Influenza virus 
vaccine, trivalent 
(ccIIV3), derived from 
cell cultures, subunit, 
preservative and 
antibiotic free, 0.5 mL 
dosage, for intramuscular 
use
o90658, Influenza virus 
vaccine, trivalent (IIV3), 
split virus, 0.5 mL  
dosage, for intramuscular 
use
o90656, Influenza virus 
vaccine, trivalent (IIV3), 
split virus, preservative 
free, 0.5 mL dosage, for 
intramuscular use
o90654, Influenza virus 
vaccine, trivalent (IIV3), 
split virus, preservative -
free, for intradermal use
o90673, Influenza virus 
vaccine, trivalent (RIV3), 
derived from 
recombinant DNA, 
hemagglutinin (HA) 
protein only , preservative 
and antibiotic free, for 
intramuscular use
o90660, Influenza virus 
vaccine, trivalent, live 
(LAIV3), for intranasal 
use
o90659, Influenza vir us 
vaccine, whole virus, for 
intramuscular or jet 
injection use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073305
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 97of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
HCPCs codes:
oG0008, 
Administration of 
influenza virus 
vaccine
oG8482, Influenza 
immunization 
administered or 
previously  received 
oQ2034, Influenza 
virus vaccine, split 
virus, for 
intramuscular use 
(Agriflu) 
oQ2035, Influenza 
virus vaccine, split 
virus, when 
administered to 
individuals 3 y ears of 
age and older, for 
intramuscular use 
(Afluria) 
oQ2036, Influenza 
virus vaccine, split 
virus, when 
administered to 
individuals 3 y ears of 
age and older, for 
intramuscular use 
(Flulaval) 
oQ2037, Influenza 
virus vaccine, split 
virus, when 
administered to 
individuals 3 y ears of 
age and older, for 
intramuscular use 
(Fluvirin)
oQ2038, Influenza 
virus vaccine, split 
virus, when 
administered to 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073306
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 98of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
individuals 3 y earsof 
age and older, for 
intramuscular use 
(Fluzone)
oQ2039, Influenza 
virus vaccine, not 
otherwise specified
Tetanus diphtheria and pertussis 
(Tdap or Td):
CPT codes:
o90714, Tetanus and 
diphtheria toxoids 
adsorbed (Td), preservative 
free, when administered to 
individuals 7 y ears or 
older, for intramuscular use
o90715, Tdap administered 
to individuals 7 y ears or 
older, for intramuscular use
o90718, Tetanus and 
diphtheria toxoids (Td) 
adsorbed when 
administered to 
individuals 7 y ears or 
older, for intramuscular 
use
Chickenpox (Varicella)
CPT codes:
o90396, Varicella -zoster 
immune globulin, human, 
for intramuscular use
o90716, Varicella virus 
vaccine, live, for 
subcutaneous use
Shingles (Herpes Zoster 
recombinant and/or live)
CPT codes:
o90396, Varicella -zoster 
immune globulin, human, 
for intramuscular use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073307
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 99of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o90736, Zoster (shingles) 
vaccine (HZV), live, for 
subcutaneous injection
o90750, Zoster (shingles) 
vaccine (HZV), 
recombinant, subunit, 
adjuvanted, for 
intramuscular use
Human papillomavirus (HPV)
CPT codes: 
o90649,Human 
Papillomavirus vaccine, 
types 6, 11, 16, 18, 
quadrivalent (4vHPV), 3 
dose schedule, for 
intramuscular use
o90650, Human 
Papillomavirus vaccine, 
types 16, 18, bivalent 
(2vHPV), 3 dose 
schedule, for 
intramuscular use
o90651, Human 
Papillomavirus vaccine
types 6, 11, 16, 18, 31, 33, 
45, 52, 58, nonavalent 
(9vHPV), 2 or 3 dose 
schedule, for 
intramuscular use
Pneumococcal conjugate
CPT codes:
o90669, Pneumococcal 
conjugate vaccine, 7 
valent, for intramuscular 
use
o90670, Pneumococcal 
conjugate vaccine, 13 
valent (PCV13), for 
intramuscular use
HCPCS codes (used 
pneumococcal conjugate and 
polysaccharide):
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073308
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 100of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oG0009, Administration of 
pneumococcal vaccine 
oG8864, Code for 
Pneumococcal vaccine 
administered or 
previously  received
Pneumococcal pol ysaccharide:
CPT code:
o90732, Pneumococcal 
polysaccharide vaccine, 
23-valent (PPSV23), adult 
or immunosuppressed 
patient dosage, when 
administered to 
individuals 2 y ears or 
older, for subcutaneous or 
intramuscular use
Hepatitis A
CPT codes
o90632, Hepatitis A 
vaccine, adult do sage, for 
intramuscular use
o90633, Hepatitis A vaccine 
(HepA), 
pediatric/adolescent 
dosage-2 dose schedule, 
for intramuscular use
o90634, Hepatitis A vaccine 
(HepA), 
pediatric/adolescent 
dosage-3 dose schedule, 
for intramuscular use
o90730, Hepatitis A vacci ne
o90636, Hepatitis A and 
hepatitis B vaccine (HepA -
HepB), adult dosage, for 
intramuscular use
Hepatitis B
CPT codes: 
o907311, Hepatitis B 
vaccine
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073309
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 101of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o90739, Hepatitis B vaccine 
(HepB), adult dosage, 2 
dose schedule, for 
intramuscular use
o90740, Hepatitis B vaccine 
(HepB), dial ysis or 
immunosuppressed patient 
dosage, 3 dose schedule, 
for intramuscular use
o90743, Hepatitis B vaccine 
(HepB), adolescent, 2 dose 
schedule, for intramuscular 
use
o90744, Hepatitis B vaccine 
(HepB), 
pediatric/adolescent 
dosage, 3 dose schedule, 
for intramuscular use
o90745, Hepatitis B 
vaccine, adolescent/high 
risk infant dosage, for 
intramuscular use
o90746, Hepatitis B 
vaccine (HepB), adult 
dosage, 3 dose schedule, 
for intramuscular use
o90747, Hepatitis B 
vaccine (HepB), dial ysis 
or immunosuppressed 
patient dosage, 4 dose 
schedule, for 
intramuscular use
HCPCS codes:
oG0010, Administration of 
Hepatitis B vaccine
Meningococcal conjugate 
(MenACWY) and serogroup B 
meningococcal (MenB)
CPT codes:
o90619, Meningococcal 
conjugate vaccine, 
serogroups A, C, W, Y, 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073310
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 102of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
quadrivalent, tetanus 
toxoid carrier 
(MenACWY -TT), for 
intramuscular use
o90620, Meningococcal 
recombinant protein and 
outer membrane vesicle 
vaccine, serogroup B 
(MenB-4C), 2 dose 
schedule, for 
intramuscular use
o90621, Meningococcal 
recombinant lipoprotein 
vaccine, serogroup B 
(MenB-FHbp), 2 or 3 
dose schedule, for 
intramuscular use
o90733, Meningococcal 
polysaccharide vaccine, 
serogroups A, C, Y, W -
135, quadrivalent 
(MPSV4), for 
subcutaneous use 90734, 
Meningococcal conjugate 
vaccine, serog roups A, C, 
W, Y, quadrivalent, 
diphtheria toxoid carrier 
(MenACWY -D) or 
CRM197 carrier 
(MenACWY -CRM), for 
intramuscular use
Haemophilus influenza type b
CPT codes:
o90645, Hemophilus 
influenza b vaccine (Hib), 
HbOC conjugate (4 dose 
schedule), for 
intramuscular use
o90646, Hemophilus 
influenza b vaccine (Hib), 
PRP-D conjugate, for 
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073311
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 103of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
booster use onl y, 
intramuscular use
o90647, Haemophilus 
influenzae ty pe b vaccine 
(Hib), PRP -OMP 
conjugate, 3 dose 
schedule, for 
intramuscular use
o90648, Haemophilus
influenzae ty pe b vaccine 
(Hib), PRP -T conjugate, 4 
dose schedule, for 
intramuscular use
o90737, Hemophilus 
influenza B
o90748, Hepatitis B and 
Haemophilus influenzae 
type b vaccine (Hib-
HepB), for intramuscular 
use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073312
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 104of 144Appendix Table 2. Operational Definitions of Safety Events of Interest
Variable Operational Definition
Defined b y the presence of an y of the following 
ICD-9-CMcodes (inclusive)1:Defined b y the presence of an y of the 
following ICD -10-CMcodes 
(inclusive)1:
Neurologic
Generalized convulsions/seizures8,22345, Epilepsy  and recurrent seizures
780.3, Convulsions
780.31, Febrile convulsions (simple), 
unspecified
780.39, Other convulsionsG40.A01, Absence epileptic 
syndrome, not intractable, with 
status epilepticus
G40.A09, Absence epileptic 
syndrome, not intractable, without 
status epilepticus
G40.A11, Absence epileptic 
syndrome, intractable, with status 
epilepticus
G40.A19, Absence epileptic 
syndrome, intractable, without 
statusepilepticus
G40.101, L ocalization -related 
(focal) (partial) sy mptomatic 
epilepsy and epileptic s yndromes 
with simple partial seizures, not 
intractable, with status epilepticus
G40.109, L ocalization -related 
(focal) (partial) sy mptomatic 
epilepsy and epilep tic syndromes 
0
…[truncated]