Serious adverse events of special interest following

ed
mRNA vaccination in randomized trials

Joseph Fraiman, MD1

Juan Erviti, PharmD, PhD2

iew
Mark Jones, PhD3
Sander Greenland, MA, MS, DrPH, C Stat4
Patrick Whelan, MD PhD5
Robert M. Kaplan, PhD6
Peter Doshi, PhD7

ev
Affiliations
1 Louisiana State University, Lallie Kemp Regional Medical Center, Independence, LA
2 Unit of Innovation and Organization. Navarre Health Service, Spain

3 Institute of Evidence-Based Healthcare, Bond University, Gold Coast, QLD, Australia

r
4 Fielding School of Public Health, University of California, Los Angeles

5 University of California, Los Angeles

er
6 School of Medicine, Stanford University

7 University of Maryland School of Pharmacy, Baltimore, MD

Correspondence to: Peter Doshi, 220 N Arch Street, Baltimore, MD, 21201
pe
[email protected]
ot
tn
rin
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 1 Serious adverse events of special interest following

ed
2 mRNA vaccination in randomized trials
3
4 Joseph Fraiman, MD1
5 Juan Erviti, PharmD, PhD2

iew
6 Mark Jones, PhD3
7 Sander Greenland, MA, MS, DrPH, C Stat4
8 Patrick Whelan, MD PhD5
9 Robert M. Kaplan, PhD6
10 Peter Doshi, PhD7
11

ev
12 Affiliations
13 1 Louisiana State University, Lallie Kemp Regional Medical Center, Independence, LA
14 2 Unit of Innovation and Organization. Navarre Health Service, Spain

15 3 Institute of Evidence-Based Healthcare, Bond University, Gold Coast, QLD, Australia

r
16 4 Fielding School of Public Health, University of California, Los Angeles

17 5 University of California, Los Angeles

18
19
20
21
er
6 School of Medicine, Stanford University

7 University of Maryland School of Pharmacy, Baltimore, MD

Correspondence to: Peter Doshi, 220 N Arch Street, Baltimore, MD, 21201
pe
22 [email protected]
23
24
25
ot
tn
rin
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 26 ABSTRACT

ed
27
28 Introduction. In 2020, prior to COVID-19 vaccine rollout, the Coalition for Epidemic
29 Preparedness Innovations and Brighton Collaboration created a priority list, endorsed by the
30 World Health Organization, of potential adverse events relevant to COVID-19 vaccines. We

iew
31 leveraged the Brighton Collaboration list to evaluate serious adverse events of special interest
32 observed in phase III randomized trials of mRNA COVID-19 vaccines.
33
34 Methods. Secondary analysis of serious adverse events reported in the placebo-controlled,
35 phase III randomized clinical trials of Pfizer and Moderna mRNA COVID-19 vaccines
36 (NCT04368728 and NCT04470427), focusing analysis on potential adverse events of special

ev
37 interest identified by the Brighton Collaboration.
38
39 Results. Pfizer and Moderna mRNA COVID-19 vaccines were associated with an increased
40 risk of serious adverse events of special interest, with an absolute risk increase of 10.1 and 15.1

r
41 per 10,000 vaccinated over placebo baselines of 17.6 and 42.2 (95% CI -0.4 to 20.6 and -3.6 to
42 33.8), respectively. Combined, the mRNA vaccines were associated with an absolute risk
43
44
45
46
er
increase of serious adverse events of special interest of 12.5 per 10,000 (95% CI 2.1 to 22.9).
The excess risk of serious adverse events of special interest surpassed the risk reduction for
COVID-19 hospitalization relative to the placebo group in both Pfizer and Moderna trials (2.3
and 6.4 per 10,000 participants, respectively).
pe
47
48 Discussion. The excess risk of serious adverse events found in our study points to the need for
49 formal harm-benefit analyses, particularly those that are stratified according to risk of serious
50 COVID-19 outcomes such as hospitalization or death.
51
ot

52 Funding. This study had no funding support.

53
54 Keywords: SARS-CoV-2; COVID-19; vaccines; COVID-19 vaccines; mRNA vaccines; Pfizer-
tn

55 BioNTech COVID-19 vaccine BNT162b2; Moderna COVID-19 vaccine mRNA-1273;

56 NCT04368728; NCT04470427; serious adverse events; adverse events of special interest;
57 Brighton Collaboration; Coalition for Epidemic Preparedness Innovations; Safety Platform for
58 Emergency vACcines
59
rin

60 Conflicts of interest:
61 JF, JE, MJ, SG, PW, RK: none to declare. PD has received travel funds from the European
62 Respiratory Society (2012) and Uppsala Monitoring Center (2018); grants from the FDA
63 (through University of Maryland M-CERSI; 2020), Laura and John Arnold Foundation (2017-22),
ep

64 American Association of Colleges of Pharmacy (2015), Patient-Centered Outcomes Research

65 Institute (2014-16), Cochrane Methods Innovations Fund (2016-18), and UK National Institute
66 for Health Research (2011-14); was an unpaid IMEDS steering committee member at the
67 Reagan-Udall Foundation for the FDA (2016-2020) and is an editor at The BMJ. The views
Pr

68 expressed here are those of the authors and do not necessarily reflect those of their employers.
69

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 70 INTRODUCTION

ed
71
72 In March 2020, the Brighton Collaboration and the Coalition for Epidemic Preparedness
73 Innovations partnership, Safety Platform for Emergency vACcines (SPEAC), created and
74 subsequently updated a “priority list of potential adverse events of special interest relevant to

iew
75 COVID-19 vaccine trials.”1 The list comprises adverse events of special interest (AESIs) based
76 on the specific vaccine platform, adverse events associated with prior vaccines in general,
77 theoretical associations based on animal models, and COVID-19 specific
78 immunopathogenesis.1 The World Health Organization’s Global Advisory Committee on
79 Vaccine Safety endorsed and recommended the reporting of AESIs based on this priority list. To
80 our knowledge, however, the list has not been applied to serious adverse events in randomized

ev
81 trial data.
82
83 We sought to investigate the association between FDA-authorized mRNA COVID-19 vaccines
84 and serious adverse events identified by the Brighton Collaboration, using data from the phase

r
85 III randomized, placebo-controlled clinical trials on which authorization was based. We then use
86 the results to illustrate the need for formal harm-benefit analyses of the vaccines that are
87
88
89
90
against post-authorization observational data.

METHODS
er
stratified according to risk of serious COVID-19 outcomes, as well as contextualize the findings
pe
91
92 Pfizer and Moderna each submitted the results of one phase III randomized trial in support of
93 the FDA’s emergency use authorization of their vaccines. Two methodologist reviewers
94 searched journal publications and trial data on the FDA’s and Health Canada’s websites to
95 locate serious adverse event results tables for these trials. The Pfizer and Moderna trials are
ot

96 expected to follow participants for two years. Within weeks of the emergency authorization,
97 however, the sponsors began a process of unblinding all participants who elected to be
98 unblinded. In addition, those who received placebo were offered the vaccine. These self-
tn

99 selection processes may have introduced nonrandom differences between the vaccine and
100 unvaccinated participants, thus rendering the post-authorization data less reliable. Therefore, to
101 preserve randomization, we used the interim datasets that were the basis for emergency
102 authorization in December 2020, approximately 4 months after trials commenced.
103
rin

104 The definition of a serious adverse event (SAE) was provided in each trial’s study protocol and
105 included in the supplemental material of the trial’s publication.2–4 Pfizer and Moderna used
106 nearly identical definitions, consistent with regulatory expectations. An SAE was defined as an
107 adverse event that results in any of the following conditions: death; life-threatening at the time of
ep

108 the event; inpatient hospitalization or prolongation of existing hospitalization; persistent or

109 significant disability/incapacity; a congenital anomaly/birth defect; medically important event,
110 based on medical judgment.
111
Pr

112 In addition to journal publications, we searched the websites of the FDA (for advisory committee
113 meeting materials) and Health Canada (for sections of the dossier submitted by sponsors to the

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
114 regulator).5 For the FDA website, we considered presentations by both the FDA and the

ed
115 sponsors.6 Within each of these sources, we searched for SAE results tables that presented
116 information by specific SAE type; we chose the most recent SAE table corresponding to the
117 FDA’s requirement for a safety median follow-up time of at least 2 months after dose 2.
118

iew
119 For each trial, blinded SAE tables (containing SAE types without results data) were prepared.
120 Using the blinded SAE tables, two clinician reviewers (JF and JE) independently judged whether
121 each SAE type was an AESI.
122
123 Our project used an AESI list derived from the work of Brighton Collaboration’s Safety Platform
124 for Emergency vACcines (SPEAC) Project. This effort created an AESI list which categorizes

ev
125 AESIs into three categories: those included because they are seen with COVID-19, those with a
126 proven or theoretical association with vaccines in general, and those with proven or theoretical
127 associations with specific vaccine platforms. The first version was produced in March 2020
128 based on experience from China. Following the second update (May 2020), the WHO Global

r
129 Advisory Committee on Vaccine Safety (GACVS) adopted the list, and Brighton commenced a
130 systematic review process “to ensure an ongoing understanding of the full spectrum of COVID-
131
132
133
134
er
19 disease and modification of the AESI list accordingly.”7 This resulted in three additional
AESIs being added to the list in December 2020. The subsequent (and most recent fourth)
update did not result in any additional AESIs being added to the list.
pe
135 We matched SAEs recorded in the trial against an expanded list of AESIs created by combining
136 Brighton’s SPEAC COVID-19 AESI list with a list of 29 clinical diagnoses Brighton identified as
137 “known to have been reported but not in sufficient numbers to merit inclusion on the AESI list.”7
138 Sensitivity analysis was used to determine whether the original versus expanded list had an
139 effect on identifying a safety concern. For SAEs that described symptoms, not diagnoses, the
ot

140 clinician reviewers independently judged whether each SAE type was likely to have been
141 caused by an AESI. For example, the SAE “abdominal pain” is a symptom based diagnosis,
142 which was judged as fitting within the SPEAC clinical diagnosis of "colitis/enteritis.”
tn

143 Disagreements were resolved through consensus; in two cases, consensus could not be
144 reached and were resolved by the judgment of a third clinician reviewer (PW) to create a
145 majority opinion. For each included SAE, we recorded the corresponding Brighton
146 Collaboration AESI category and organ system.
147
rin

148 Risk ratios and risk differences between vaccine and placebo groups were calculated for the
149 incidence of SAEs. We excluded SAEs that were efficacy outcomes (i.e. COVID-19). Because
150 we did not have access to individual participant data, to account for the occasional multiple
151 SAEs within single participants, we reduced the effective sample size by multiplying standard
ep

152 errors in the combined SAE analyses by the square root of the ratio of the number of SAEs to
153 the number of patients with an SAE. This adjustment increased standard errors by 10% (Pfizer)
154 and 18% (Moderna), thus expanding the interval estimates. We estimated combined risk ratios
155 and risk differences for the two mRNA vaccines by averaging over the risks using logistic
Pr

156 regression models.

157

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
158 We used a simple harm-benefit framework to place our results in context. The analysis

ed
159 compared risks of excess serious AESIs against reductions in serious complications of COVID-
160 19.
161
162

iew
163 RESULTS
164
165 Serious adverse event tables were located for each of the vaccine trials submitted for EUA in
166 the United States: Pfizer-BioNTech COVID-19 vaccine BNT162b2 (NCT04368728)2,8,9 and
167 Moderna COVID-19 vaccine mRNA-1273 (NCT04470427).3,10,11 (Table 1)
168

ev
169 Reporting windows and all-cause serious adverse events
170
171 Moderna reported SAEs from dose 1 whereas Pfizer limited reporting from dose 1 to 1 month
172 after dose 2. Both studies reported all data at the time of data cutoff.

r
173
174 The Pfizer trial reported a 36% higher risk of serious adverse events unrelated to COVID-19 in
175
176
177
178
er
vaccinated participants than placebo recipients: 67.5 per 10,000 versus 49.5 per 10,000; risk
ratio 1.36 (95% compatibility1 interval, CI 1.02 to 1.83). The Moderna trial reported a 5% higher
risk of SAEs unrelated to COVID-19 in vaccinated individuals compared to those receiving
placebo: 136 per 10,000 versus 129 per 10,000; risk ratio 1.05 (95% CI 0.83 to 1.32).
pe
179 Combined, there was a 15% higher risk of SAEs unrelated to COVID-19 in mRNA vaccine
180 recipients than placebo recipients: 98 per 10,000 versus 85 per 10,000; risk ratio 1.15 (95% CI
181 0.96 to 1.38). (Table 2)
182
183 Serious adverse events of special interest
ot

184
185 Regarding whether each SAE type was included on the SPEAC derived AESI list, agreement
186 between the two independent clinician reviewers was 86% (281/325); 40 of the 44
tn

187 disagreements were resolved through consensus, and only four disagreements necessitated a
188 third clinician reviewer. Supplemental Table 1 includes a full list of included and excluded
189 SAEs across both trials.
190
191 In the Pfizer trial, 52 serious AESI (27.7 per 10,000) were reported in the vaccine group and 33
rin

192 (17.6 per 10,000) in the placebo group. This difference corresponds to a 57% increased risk of
193 serious AESI (RR 1.57 95% CI 0.98 to 2.54) and an absolute risk increase of 10.1 serious AESI
194 per 10,000 vaccinated participants (95% CI -0.4 to 20.6). In the Moderna trial, 87 serious AESI
195 (57.3 per 10,000) were reported in the vaccine group and 64 (42.2 per 10,000) in the placebo
ep

196 group. This difference corresponds to a 36% increased risk of serious AESI (RR 1.36 95% CI
197 0.93 to 1.99) and an absolute risk increase of 15.1 serious AESI per 10,000 vaccinated
198 participants (95% CI -3.6 to 33.8). Combining the trials, there was a 43% increased risk of
Pr

1A compatibility interval is identical to a confidence interval, but relabeled to emphasize that it is not a
Bayesian posterior interval (as is improperly suggested by the “confidence” label).12,13

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
199 serious AESI (RR 1.43; 95% CI 1.07 to 1.92) and an absolute risk increase of 12.5 serious AESI

ed
200 per 10,000 vaccinated participants (95% CI 2.1 to 22.9). (Table 2)
201
202 Of the 236 serious AESIs occurring across the Pfizer and Moderna trials, 97% (230/236) were
203 adverse event types included as AESIs because they are seen with COVID-19. In both Pfizer

iew
204 and Moderna trials, the largest increase in absolute risk occurred amongst the Brighton
205 category of coagulation disorders. Cardiac disorders have been of central concern for mRNA
206 vaccines; more cardiovascular AESIs occurred in the vaccine group in the Pfizer trial, but
207 cardiovascular AESI events were balanced in the Moderna trial. (Tables 3 and 4)
208
209 Sensitivity analysis

ev
210
211 In a sensitivity analysis, we restricted the serious AESI analysis to those AESIs listed in
212 SPEAC’s COVID-19 AESI list (i.e. separating out Brighton’s list of 29 clinical diagnoses “known
213 to have been reported but not in sufficient numbers to merit inclusion on the AESI list.”) This

r
214 reduced the total number of AESIs across the two trials by 48 (35 vaccine group, 13 placebo
215 group). There was still a higher risk of serious AESI when limited to the SPEAC COVID-19 AESI
216
217
218
219
er
list, but the magnitude of the increase (in both relative and absolute terms) was smaller than
when using the larger AESI list. (Supplemental Table 2).

Harm-benefit considerations
pe
220
221 In the Moderna trial, the excess risk of serious AESIs (15.1 per 10,000 participants) surpassed
222 the risk reduction for COVID-19 hospitalization relative to the placebo group (6.4 per 10,000
223 participants).3 In the Pfizer trial, the excess risk of serious AESIs (10.1 per 10,000) surpassed
224 the risk reduction for COVID-19 hospitalization relative to the placebo group (2.3 per 10,000
ot

225 participants).
226
227 Comparison with FDA reviews and post-authorization studies
tn

228
229 In their review of SAEs that supported the authorization of the Pfizer and Moderna vaccines, the
230 FDA concluded that SAEs were, for Pfizer, “balanced between treatment groups,”14 and for
231 Moderna, were “without meaningful imbalances between study arms.”15 In contrast to the FDA
232 analysis, we found an increased risk of all cause SAEs in the Pfizer trial. While our analysis
rin

233 excluded SAEs related to COVID-19 (because it is an efficacy outcome), this exclusion did not
234 explain the difference given the low risk of SAEs attributed to COVID-19 (0 in the vaccine arm, 1
235 in the placebo arm). Instead, the difference in findings may in part be explained by the fact that
236 the FDA analyzed the total number of participants experiencing any SAE, whereas our analysis
ep

237 was based on the total number of SAE events. Given that approximately twice as many
238 individuals in the vaccine group experienced multiple SAEs than the placebo group (there were
239 24 more events than participants in the vaccine group, compared to 13 in the placebo group),
240 FDA’s analysis of only the incidence of participants experiencing any SAE would not reflect the
Pr

241 observed increase in multiple SAEs in the vaccine group.

242

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
243 A more important factor, however, may be that FDA’s review of non-fatal SAEs used a different

ed
244 analysis population with different follow-up windows. The FDA reported 126 of 21621 (0.6%) of
245 vaccinated participants experienced at least one SAE at data cutoff compared to 111 of 21631
246 (0.5%) of placebo participants. In contrast, our analysis found 127 SAEs among 18,801 vaccine
247 recipients versus 93 SAEs among 18,785 placebo recipients.14 While summary results for the

iew
248 population we analyzed was provided in a table, FDA did not report an analysis of them. The
249 substantially larger denominators in FDA’s analysis (5,666 more participants) reflect the fact that
250 their analysis included all individuals receiving at least one dose (minus 196 HIV-positive
251 participants), irrespective of the duration of post-injection follow-up time. In contrast, our
252 analysis was based on the study population with median follow-up ≥2 months after dose 2
253 (minus 120 HIV-positive participants), of which 98.1% had received both doses.2,16 The FDA’s

ev
254 analysis of SAEs thus included thousands of additional participants with very little follow-up, of
255 which the large majority had only received 1 dose.
256
257 Although the randomized trials offer high level evidence for a causal association, the sparsity of

r
258 their data necessitates that harm-benefit analyses also consider observational data. Since their
259 emergency authorization in December 2020, hundreds of millions of doses of Pfizer and
260
261
262
263
er
Moderna COVID-19 vaccines have been administered and post-authorization observational
data offer a complementary opportunity to study AESIs. Post-authorization observational safety
studies include cohort studies (which make use of medical claims or electronic health records)
and disproportionality analyses (which leverage spontaneous adverse event reporting systems).
pe
264 In July 2021, the FDA reported detecting four potential adverse events of interest: pulmonary
265 embolism, acute myocardial infarction, immune thrombocytopenia, and disseminated
266 intravascular coagulation following Pfizer’s vaccine based on medical claims data in older
267 Americans.17 Three of these four serious adverse event types would be categorized as
268 coagulation disorders, which is the Brighton AESI category which showed the largest absolute
ot

269 increase in the vaccine group in both the Pfizer and Moderna trials. FDA stated it would further
270 investigate the findings but at the time of our writing has not issued an update. Similarly,
271 spontaneous-reporting systems have registered serious adverse reactions including
tn

272 anaphylaxis (all COVID-19 vaccines), thrombocytopenia syndrome among premenopausal

273 females (Janssen vaccine), and myocarditis and pericarditis among younger males (Pfizer and
274 Moderna vaccines).18,19
275
276 Using data from three postmarketing safety databases for vaccines (VAERS, EudraVigilance,
rin

277 and VigiBase), disproportionality studies have reported an increase in many of the same SAE
278 types found in the present study.20–22 For example, a study using VAERS and EudraVigilance
279 comparing the disproportionality of adverse event reports between the influenza vaccine versus
280 the mRNA COVID-19 vaccines reported increased relative risk of the following Brighton AESIs:
ep

281 cardiovascular events, coagulation events, hemorrhages, gastrointestinal events, and

282 thromboses. While CDC published a protocol23 in early 2021 for using proportional reporting
283 ratios for signal detection in the VAERS database, the agency has not yet reported such a
284 study.24 Among self-controlled case series, one reported an incidence rate ratio of 1.38 (95% CI
Pr

285 1.12-1.71) for hemorrhagic stroke following Pfizer vaccine,25 another reported 0.97 (95% CI
286 0.81-1.15),26 while a cohort study27 reported 0.84 (95% CI 0.54-1.27).

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
287

ed
288 DISCUSSION
289
290 Using a prespecified list of AESI identified by the Brighton Collaboration, an increase in serious
291 AESI was found in the mRNA COVID-19 vaccine group in both the Pfizer and Moderna adult

iew
292 phase III trials, from 10.1 (Pfizer) to 15.1 (Moderna) additional events for every 10,000
293 individuals vaccinated.
294
295 Comparing the excess of serious AESI against the reduction of serious complications of COVID-
296 19 among the vaccinated is essential for harm-benefit analyses. The results show an excess
297 risk of serious AESIs greater than the reduction in COVID-19 hospitalizations in both Pfizer and

ev
298 Moderna trials. These results are compatible with a recent preprint analysis of COVID-19
299 vaccine trials by Benn et al., which found no evidence of a reduction in overall mortality in the
300 mRNA vaccine trials based on data from the later, March 2021 BLA (Biologics License
301 Application) timepoints that underpinned subsequent regulatory approval (31 deaths in the

r
302 vaccine arms versus 30 events in the placebo arms; RR 1.03, 95% CI 0.63 to 1.71).28 Our
303 analysis as well as Benn et al. point to the need for formal harm-benefit analyses especially in
304
305
306
307
er
individuals at low risk of COVID-19 hospitalization or death. Using VAERS data, Krug et al.
attempted such an analysis, albeit focused on just one SAE (myocarditis)19 Individual participant
data for all SAEs is not publicly available at present, but would help identify factors (e.g. age
and comorbidities) that may elevate the risk of serious AESIs. It would also be essential to
pe
308 compare long-term outcomes of vaccinated and unvaccinated groups, e.g., for symptoms
309 identified with “long covid.”
310
311 Adverse events detected in the post-marketing period have led to the withdrawal of several past
312 vaccines. An example is intussusception following one brand of rotavirus vaccine: around 1
ot

313 million children were vaccinated before identification of intussusception, which occurred in
314 around 1 per 10,000 vaccinees.29 Despite the unprecedented scale of COVID-19 vaccine
315 administration, the AESI types identified in our study may still be challenging to detect with
tn

316 observational methods. Most cohort study designs crucially depend upon comparing the risks
317 of adverse events “observed” against a background (or “expected”) risk. However, background
318 incidence risks display great variation, by database, age group, and sex.30 If the risk ratio of 1.4
319 estimated in our analysis were the actual effect size, it could be quite difficult to unambiguously
320 replicate it with observational data given concerns about systematic as well as random errors.31–
rin

321 33

322
323 In addition, disproportionality analyses following COVID-19 vaccination also have limitations,
324 particularly with respect to the type of adverse events seen in our study. The majority of SAE
ep

325 types that contributed to our results are relatively common events, such as ischemic stroke,
326 acute coronary syndrome, and brain hemorrhage. This complicates signal detection because
327 clinical suspicion of an adverse vaccine reaction following an event commonly seen in clinical
328 practice will be lower than for less commonly observed SAEs like myocarditis. For this reason,
Pr

329 the basic ingredient for effective pharmacovigilance--clinical suspicion leading to the filing of an
330 individual case safety report--may be far less common in the post-authorization setting. At the

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
331 same time, heightened awareness about COVID-19 vaccines can result in over- and under-

ed
332 reporting. Public health messages assuring vaccine safety may lower clinical suspicion of
333 potential causal relationships, whereas messages about potential harms can conversely
334 stimulate reports that otherwise may not have been made. There are thus factors that can lead
335 to bias in either direction, further complicating analysis and interpretation. In contrast to these

iew
336 problems, in the randomized clinical trials used in this analysis, all SAEs were to be recorded,
337 irrespective of clinical judgment regarding potential causality.
338
339 Although our analysis is secondary, reanalyses of clinical trial data have led to the detection of
340 adverse events well after the market entry of major drugs such as rofecoxib and
341 rosiglitazone.34,35 Our analysis has an advantage over postmarketing observational studies in

ev
342 that the data are from blinded, placebo-controlled randomized trials vetted by the FDA, and
343 uses the Brighton Collaboration AESI list, which was pre-specified, endorsed by WHO, and
344 established well before the availability of the clinical-trial results, and designed for use in
345 COVID-19 vaccine trials.

r
346
347 Limitations of our study include that Pfizer’s SAE table did not include SAEs occurring past 1
348
349
350
351
er
month after dose 2. This reporting threshold may have led to an undercounting of serious AESIs
in the Pfizer study, and for both studies, the limited follow up time prevented an analysis of
harm-benefit over a longer time period. It should also be recognized that all SAEs in our
analysis are those that met the regulatory definition of a serious adverse event. However, many
pe
352 adverse event types which a patient may themselves judge as serious may not meet this
353 regulatory threshold.
354
355 Another limitation is our lack of access to individual participant data, which forced us to use a
356 conservative adjustment to the standard errors. The 95% CI12,13 calculated are therefore only
ot

357 approximate because we do not know which patients had multiple events. Furthermore, despite
358 our attempt to remove efficacy endpoints from our analysis (i.e., SAEs labeled as COVID-19,
359 COVID-19 pneumonia, and “SARS-CoV-2 test positive”), it was not possible to identify and
tn

360 remove SAEs that occurred in patients with serious complications of COVID-19 (e.g., acute
361 respiratory failure, cardiac arrest, and acute kidney injury), which are common. Of 18 total
362 efficacy SAEs removed from our analysis, 17 were in the Moderna trial, and of these, 16 were in
363 the placebo arm. This suggests the possibility that SAEs were overcounted in the placebo arm
364 of our analyses, primarily for Moderna’s vaccine, due to our inability to remove COVID-19-
rin

365 related SAEs. These study limitations all stem from the fact that the raw data from COVID-19
366 vaccine clinical trials are not publicly available.36,37 Given the global public health implications,
367 there is an urgency to make all COVID-19 trial data public, particularly regarding serious
368 adverse events, without any further delay.
ep

369
370 Finally, we emphasize that the elevated risk of serious AESIs in the vaccine group represents
371 an average across the group. SAEs may not be distributed equally across the demographic
372 subgroups enrolled in the trial, and the risks may be substantially less in some groups
Pr

373 compared to others. Thus, knowing the actual demographics of those who experienced an
374 increase in serious AESI in the vaccine group is necessary for a proper harm-benefit analysis.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
375 A systematic review and meta-analysis using individual participant data should be undertaken to

ed
376 address questions of harm-benefit in various demographic subgroups. Full transparency of the
377 COVID-19 vaccine clinical trial data is needed to properly evaluate these questions.
378 Unfortunately, well over a year after widespread use of COVID-19 vaccines, participant level
379 data remain inaccessible.36,37

iew
380
381
382 Author Contributions: All authors had full access to all of the data in the study (available at
383 https://doi.org/10.5281/zenodo.6564403), and take responsibility for the integrity of the data and
384 the accuracy of the data analysis.
385

ev
386 Study concept and design: All authors
387 Acquisition of data: Doshi
388 Analysis and interpretation: All authors
389 Statistical analysis: Jones, Greenland

r
390 Drafting of the manuscript: Fraiman, Doshi
391 Critical revision of the manuscript for important intellectual content: All authors
392
393
394
395
er
Acknowledgements: We thank Jean Rees for help identifying sources of data.

Funding: This study had no funding support.

pe
396
397 Conflicts of interest:
398 JF, JE, MJ, SG, PW, RK: none to declare. PD has received travel funds from the European
399 Respiratory Society (2012) and Uppsala Monitoring Center (2018); grants from the FDA
400 (through University of Maryland M-CERSI; 2020), Laura and John Arnold Foundation (2017-22),
ot

401 American Association of Colleges of Pharmacy (2015), Patient-Centered Outcomes Research

402 Institute (2014-16), Cochrane Methods Innovations Fund (2016-18), and UK National Institute
403 for Health Research (2011-14); was an unpaid IMEDS steering committee member at the
tn

404 Reagan-Udall Foundation for the FDA (2016-2020) and is an editor at The BMJ. The views
405 expressed here are those of the authors and do not necessarily reflect those of their employers.
406
407 References:
rin

408 1. Law B, Pim C. SO2-D2.1.3 Priority List of COVID-19 Adverse events of special interest
409 [Internet]. 2021 Oct [cited 2022 Feb 17]. Available from: https://brightoncollaboration.us/wp-
410 content/uploads/2021/11/SO2_D2.1.3_COVID-19_AESI-update_V1.0_Part-
411 2_09Nov2021.pdf
ep

412 2. Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, et al. Safety and
413 Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. N Engl J Med [Internet]. 2020 Dec
414 31;383(27):2603–15. Available from: http://dx.doi.org/10.1056/NEJMoa2034577

415 3. Baden LR, El Sahly HM, Essink B, Kotloff K, Frey S, Novak R, et al. Efficacy and Safety of
Pr

416 the mRNA-1273 SARS-CoV-2 Vaccine. N Engl J Med [Internet]. 2021 Feb 4;384(5):403–
417 16. Available from: http://dx.doi.org/10.1056/NEJMoa2035389

10

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
418 4. Sadoff J, Gray G, Vandebosch A, Cárdenas V, Shukarev G, Grinsztejn B, et al. Safety and

ed
419 Efficacy of Single-Dose Ad26.COV2.S Vaccine against Covid-19. N Engl J Med [Internet].
420 2021 Jun 10;384(23):2187–201. Available from: http://dx.doi.org/10.1056/NEJMoa2101544

421 5. Health Canada. Search for clinical information on drugs and medical devices [Internet].
422 2019 [cited 2021 Nov 9]. Available from: https://clinical-information.canada.ca/

iew
423 6. Food and Drug Administration. Meeting Materials, Vaccines and Related Biological
424 Products Advisory Committee [Internet]. U.S. Food and Drug Administration. 2022 [cited
425 2022 Feb 18]. Available from: https://www.fda.gov/advisory-committees/vaccines-and-
426 related-biological-products-advisory-committee/meeting-materials-vaccines-and-related-
427 biological-products-advisory-committee

ev
428 7. Law B. SO2-D2.1.2 Priority List of COVID-19 Adverse events of special interest: Quarterly
429 update December 2020 [Internet]. 2020 Dec [cited 2020 Dec 20]. Available from:
430 https://brightoncollaboration.us/wp-content/uploads/2021/01/SO2_D2.1.2_V1.2_COVID-
431 19_AESI-update-23Dec2020-review_final.pdf

r
432 8. Pfizer-BioNTech. PFIZER-BIONTECH COVID-19 VACCINE (BNT162, PF-07302048)
433 VACCINES AND RELATED BIOLOGICAL PRODUCTS ADVISORY COMMITTEE
434 BRIEFING DOCUMENT. [cited 2021 Dec 20]; Available from:
435

436
437
9.
er
https://www.fda.gov/media/144246/download#page=87

Pfizer. Final Analysis Interim Report: A Phase 1/2/3, Placebo-Controlled, Randomized,

Observer-Blind, Dose-Finding Study to Evaluate the Safety, Tolerability, Immunogenicity,
pe
438 and Efficacy of SARS-COV-2 RNA Vaccine Candidates Against COVID-19 in Healthy
439 Individuals (Protocol C4591001) [Internet]. [cited 2022 May 3]. Available from:
440 https://clinical-information.canada.ca/ci-rc/item/244906; https://clinical-
441 information.canada.ca/ci-rc-vu.pdf?file=m5/c45/c4591001-fa-interim-report-
442 body_Unblinded_Redacted.pdf&id=244906
ot

443 10. Moderna. Sponsor briefing document [Internet]. 2020 Dec [cited 2022 Feb 21]. Available
444 from: https://www.fda.gov/media/144452/download

445 11. Moderna. Unblinded Safety Tables Batch 1 (DS2) [Internet]. [cited 2022 May 3]. Available
tn

446 from: https://clinical-information.canada.ca/ci-rc/item/244946; https://clinical-

447 information.canada.ca/ci-rc-vu.pdf?file=m5/5.3.5.1/m5351-mrna-1273-p301-p-unblinded-
448 safety-tables-batch-1.pdf&id=244946

449 12. Amrhein V, Greenland S, McShane B. Scientists rise up against statistical significance.
rin

450 Nature [Internet]. 2019 Mar;567(7748):305–7. Available from:

451 http://dx.doi.org/10.1038/d41586-019-00857-9

452 13. Rafi Z, Greenland S. Semantic and cognitive tools to aid statistical science: replace
453 confidence and significance by compatibility and surprise. BMC Med Res Methodol
ep

454 [Internet]. 2020 Sep 30;20(1):244. Available from: http://dx.doi.org/10.1186/s12874-020-

455 01105-9

456 14. Food and Drug Administration. Emergency Use Authorization for Pfizer-BioNTech COVID-
457 19 Vaccine Review Memo [Internet]. 2020 Dec [cited 2022 Feb 21]. Available from:
Pr

458 https://www.fda.gov/media/144416/download

11

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
459 15. Food and Drug Administration. Moderna COVID-19 Vaccine EUA FDA review

ed
460 memorandum [Internet]. 2020 Dec [cited 2022 Feb 21]. Available from:
461 https://www.fda.gov/media/144673/download

462 16. Food and Drug Administration. Pfizer-BioNTech COVID-19 vaccine EUA review
463 memorandum [Internet]. 2020 Dec [cited 2022 Mar 30]. Available from:

iew
464 https://www.fda.gov/media/144416/download

465 17. Food and Drug Administration. Initial Results of Near Real-Time Safety Monitoring COVID-
466 19 Vaccines [Internet]. 2021 [cited 2022 Mar 30]. Available from:
467 https://www.fda.gov/vaccines-blood-biologics/safety-availability-biologics/initial-results-near-
468 real-time-safety-monitoring-covid-19-vaccines-persons-aged-65-years-and-older

ev
469 18. CDC. Selected adverse events reported after COVID-19 vaccination [Internet]. 2021 [cited
470 2021 May 28]. Available from: https://www.cdc.gov/coronavirus/2019-
471 ncov/vaccines/safety/adverse-events.html

472 19. Krug A, Stevenson J, Høeg TB. BNT162b2 Vaccine-Associated Myo/Pericarditis in

r
473 Adolescents: A Stratified Risk-Benefit Analysis. Eur J Clin Invest [Internet]. 2022
474 May;52(5):e13759. Available from: http://dx.doi.org/10.1111/eci.13759

475
476
477
er
20. Dutta S, Kaur R, Charan J, Bhardwaj P, Ambwani SR, Babu S, et al. Analysis of
Neurological Adverse Events Reported in VigiBase From COVID-19 Vaccines. Cureus
[Internet]. 2022 Jan;14(1):e21376. Available from: http://dx.doi.org/10.7759/cureus.21376
pe
478 21. Montano D. Frequency and Associations of Adverse Reactions of COVID-19 Vaccines
479 Reported to Pharmacovigilance Systems in the European Union and the United States.
480 Front Public Health [Internet]. 2021;9:756633. Available from:
481 http://dx.doi.org/10.3389/fpubh.2021.756633

482 22. Jeet Kaur R, Dutta S, Charan J, Bhardwaj P, Tandon A, Yadav D, et al. Cardiovascular
ot

483 Adverse Events Reported from COVID-19 Vaccines: A Study Based on WHO Database. Int
484 J Gen Med [Internet]. 2021 Jul 27;14:3909–27. Available from:
485 http://dx.doi.org/10.2147/IJGM.S324349
tn

486 23. Centers for Disease Control and Prevention. Vaccine Adverse Event Reporting System
487 (VAERS) Standard Operating Procedures for COVID-19 (as of 29 January 2021) [Internet].
488 2021 Jan [cited 2022 Mar 30]. Available from:
489 https://www.cdc.gov/vaccinesafety/pdf/VAERS-v2-SOP.pdf
rin

490 24. Centers for Disease Control and Prevention. Vaccine safety publications [Internet]. 2022
491 [cited 2022 Mar 31]. Available from:
492 https://www.cdc.gov/vaccinesafety/research/publications/index.html

493 25. Patone M, Handunnetthi L, Saatci D, Pan J, Katikireddi SV, Razvi S, et al. Neurological
ep

494 complications after first dose of COVID-19 vaccines and SARS-CoV-2 infection. Nat Med
495 [Internet]. 2021 Dec;27(12):2144–53. Available from: http://dx.doi.org/10.1038/s41591-021-
496 01556-7

497 26. Jabagi MJ, Botton J, Bertrand M, Weill A, Farrington P, Zureik M, et al. Myocardial
Pr

498 Infarction, Stroke, and Pulmonary Embolism After BNT162b2 mRNA COVID-19 Vaccine in
499 People Aged 75 Years or Older. JAMA [Internet]. 2022 Jan 4;327(1):80–2. Available from:

12

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
500 http://dx.doi.org/10.1001/jama.2021.21699

ed
501 27. Barda N, Dagan N, Ben-Shlomo Y, Kepten E, Waxman J, Ohana R, et al. Safety of the
502 BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Setting. N Engl J Med [Internet]. 2021
503 Sep 16;385(12):1078–90. Available from: http://dx.doi.org/10.1056/NEJMoa2110475

iew
504 28. Benn CS, Schaltz-Buchholzer F, Nielsen S, Netea MG, Aaby P. Randomised Clinical Trials
505 of COVID-19 Vaccines: Do Adenovirus-Vector Vaccines Have Beneficial Non-Specific
506 Effects? [Internet]. 2022 [cited 2022 May 9]. Available from:
507 https://papers.ssrn.com/abstract=4072489

508 29. Hampton LM, Aggarwal R, Evans SJW, Law B. General determination of causation
509 between Covid-19 vaccines and possible adverse events. Vaccine [Internet]. 2021 Mar

ev
510 5;39(10):1478–80. Available from: http://dx.doi.org/10.1016/j.vaccine.2021.01.057

511 30. Li X, Ostropolets A, Makadia R, Shoaibi A, Rao G, Sena AG, et al. Characterising the
512 background incidence rates of adverse events of special interest for covid-19 vaccines in
513 eight countries: multinational network cohort study. BMJ [Internet]. 2021 Jun 14 [cited 2022

r
514 Mar 28];373. Available from: https://www.bmj.com/content/373/bmj.n1435

515 31. Greenland S. Relation of probability of causation to relative risk and doubling dose: a
516
517

518
er
methodologic error that has become a social problem. Am J Public Health [Internet]. 1999
Aug;89(8):1166–9. Available from: http://dx.doi.org/10.2105/ajph.89.8.1166

32. MacLehose RF, Ahern TP, Lash TL, Poole C, Greenland S. The Importance of Making
pe
519 Assumptions in Bias Analysis. Epidemiology [Internet]. 2021 Sep 1;32(5):617–24. Available
520 from: http://dx.doi.org/10.1097/EDE.0000000000001381

521 33. Greenland S. Invited Commentary: Dealing With the Inevitable Deficiencies of Bias
522 Analysis-and All Analyses. Am J Epidemiol [Internet]. 2021 Aug 1;190(8):1617–21.
523 Available from: http://dx.doi.org/10.1093/aje/kwab069
ot

524 34. Krumholz HM, Ross JS, Presler AH, Egilman DS. What have we learnt from Vioxx? BMJ
525 [Internet]. 2007 Jan 20;334(7585):120–3. Available from:
526 http://dx.doi.org/10.1136/bmj.39024.487720.68
tn

527 35. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death
528 from cardiovascular causes. N Engl J Med [Internet]. 2007 Jun 14;356(24):2457–71.
529 Available from: http://dx.doi.org/10.1056/NEJMoa072761
rin

530 36. Tanveer S, Rowhani-Farid A, Hong K, Jefferson T, Doshi P. Transparency of COVID-19

531 vaccine trials: decisions without data. BMJ Evid Based Med [Internet]. 2021 Aug 9;
532 Available from: http://dx.doi.org/10.1136/bmjebm-2021-111735

533 37. Doshi P, Godlee F, Abbasi K. Covid-19 vaccines and treatments: we must have raw data,
ep

534 now. BMJ [Internet]. 2022 Jan 19;376:o102. Available from:

535 http://dx.doi.org/10.1136/bmj.o102

536
537
Pr

13

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 Table 1. Data sources for phase III trials

ed
Trial Data Journal FDA sources Health Canada
cutoff date articles sources

iew
Pfizer trial in 14 Nov Aggregate Table 23 in sponsor Table 55 in sponsor
ages 16 and 2020 data only briefing document document C4591001
above (supported Final Analysis
Dec 2020 Interim Report Body
(NCT04368728)
EUA)

Moderna trial in 25 Nov Table S11 in Table 27 in sponsor Table 14.3.1.13.3 in

ev
ages 18 and 2020 publication briefing document sponsor document
above (supported mRNA-1273-P301
Dec 2020 Unblinded Safety
(NCT04470427)
EUA) Tables Batch 1
(DS2)

r
Note: bolded font indicates dataset chosen for analysis; EUA = Emergency Use Authorization

538
539
er
pe
ot
tn
rin
ep
Pr

14

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
540

ed
Table 2. Serious adverse events

Eventsa Risk difference Risk ratio

per 10,000 participants (95% CI)

iew
(95% CI)
Trial Vaccine Placebo

All serious adverse eventsb

Pfizer 127 93 18.0 (1.2 to 34.9) 1.36 (1.02 to 1.83)

ev
Moderna 206 196 6.4 (-23.9 to 36.8) 1.05 (0.83 to 1.32)

Combined 333 289 12.9 (-0.4 to 29.3) 1.15 (0.96 to 1.38)

r
Serious adverse events of special interestc

Pfizer 52 33 10.1 (-0.4 to 20.6) 1.57 (0.98 to 2.54)

Moderna

Combined
87

139
64

97
er
15.1 (-3.6 to 33.8)

12.5 (2.1 to 22.9)

1.36 (0.93 to 1.99)

1.43 (1.07 to 1.92)

pe
a Denominators for Pfizer were 18,801 in the vaccine group and 18,785 in the placebo group,
and for Moderna were 15,185 in the vaccine group and 15,166 in the placebo group.
b All SAEs are included in the calculations except for efficacy outcomes which were included in

certain SAE tables: “COVID-19” and “COVID-19 pneumonia” (Moderna) and “SARS-CoV-2 test
positive” (Pfizer). “All SAEs” for Moderna was calculated using the “Number of serious AEs” row
ot

in Moderna’s submission to FDA.10

c Standard errors used to estimate 95% CIs were inflated by the factor

√[#SAE]/[#patients with SAE] to account for multiple SAE within patients.

tn

541
542
543
544
rin
ep
Pr

15

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
Table 3. Serious AESIs, Pfizer trial

Vaccine events Placebo events Difference in

e d
Brighton category
Association with immunization in general
Anaphylaxis
Vaccine

1
Placebo

1
per 10,000

0.5
per 10,000

0.5
events per 10,000

ie
0.0
w Risk ratio

1.00
Association with specific vaccine platform(s)
Encephalitis/encephalomyelitis 0 2 0.0 1.1

e v -1.1 0.00
Seen with COVID-19
Acute kidney injury
Acute liver injury
2
0
0
1
r
1.1
0.0
r 0.0
0.5
1.1
-0.5
N/A
0.00
Acute respiratory distress syndrome
Coagulation disorder
2
16
1
10

ee
1.1
8.5
0.5
5.3
0.5
3.2
2.00
1.60

p
Myocarditis/pericarditis 2 1 1.1 0.5 0.5 2.00
Other forms of acute cardiac injury 16 12 8.5 6.4 2.1 1.33
Subtotal 39 28

o t
Brighton list of 29 clinical diagnoses seen with COVID-19
20.7 14.9 5.8 1.39

n
Abscess 4 1 2.1 0.5 1.6 4.00
Cholecystitis 4 2 2.1 1.1 1.1 2.00
Colitis/Enteritis
Diarrhea
Hyperglycemia
Pancreatitis
ir n t 1
1
1
1
1
0
1
0
0.5
0.5
0.5
0.5
0.5
0.0
0.5
0.0
0.0
0.5
0.0
0.5
1.00
N/A
1.00
N/A
Psychosis
Subtotal

e p 1
13
0
5
0.5
6.9
0.0
2.7
0.5
4.3
N/A
2.60

Pr
Total 52 33 27.7 17.6 10.1 1.57

16

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
Table 4. Serious AESIs, Moderna trial
e d
Brighton category
Association with specific vaccine platform(s)
Vaccine Placebo
Vaccine events Placebo events
per 10,000
Difference in

ie w
per 10,000 events per 10,000 Risk ratio

Bell's Palsy
Encephalitis/encephalomyelitis
1
1
0
0
0.7
0.7
0.0

e
0.0
v 0.7
0.7
N/A
N/A
Seen with COVID-19
Acute kidney injury
Acute liver injury
1
1
3
0
r r 0.7
0.7
2.0
0.0
-1.3
0.7
0.33
N/A
Acute respiratory distress syndrome
Angioedema
Coagulation disorder
7
0
20
4
2
13
ee 4.6
0.0
13.2
2.6
1.3
8.6
2.0
-1.3
4.6
1.75
0.00
1.54
Generalized Convulsions
Myelitis
2
0

t
0
1 p 1.3
0.0
0.0
0.7
1.3
-0.7
N/A
0.00
Myocarditis/pericarditis
Other forms of acute cardiac injury
Other rash
n
4
26
1 o 5
26
1
2.6
17.1
0.7
3.3
17.1
0.7
-0.7
0.0
0.0
0.80
1.00
1.00
Rhabdomyolysis

Subtotal

ir n
Single Organ Cutaneous Vasculitis
t 0
1
65
Brighton list of 29 clinical diagnoses seen with COVID-19
1
0
56
0.0
0.7
42.8
0.7
0.0
36.9
-0.7
0.7
5.9
0.00
N/A
1.16

Abscess
Arthritis

e p 1
3
0
1
0.7
2.0
0.0
0.7
0.7
1.3
N/A
3.00

r
Cholecystitis
Colitis/Enteritis

P
4
6
0
3
2.6
4.0
0.0
2.0
2.6
2.0
N/A
2.00

17

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
Diarrhea
Hyperglycemia
2
1
1
0
1.3
0.7
0.7
0.0
0.7
0.7
e d
2.00
N/A
Hyponatremia
Pancreatitis
Pneumothorax
1
2
0
1
0
1
0.7
1.3
0.0
0.7
0.0
0.7

ie w 0.0
1.3
-0.7
1.00
N/A
0.00
Psychosis
Thyroiditis
1
1
1
0
0.7
0.7
0.7

e
0.0
v 0.0
0.7
1.00
N/A
Subtotal
Total
22
87
8
64

r r 14.5
57.3
5.3
42.2
9.2
15.1
2.75
1.36

ee
t p
n o
ir n t
e p
Pr 18

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 ed
Supplemental Table 1. Included and excluded SAE types across both trials

Included SAE types (matching AESI list): Abdominal pain, Abdominal pain upper, Abscess,
Abscess intestinal, Acute coronary syndrome, Acute kidney injury, Acute left ventricular
failure, Acute myocardial infarction, Acute respiratory failure, Anaemia, Anaphylactic reaction,

iew
Anaphylactic shock, Angina pectoris, Angina unstable, Angioedema, Aortic aneurysm, Aortic
valve incompetence, Arrhythmia supraventricular, Arteriospasm coronary, Arthritis, Atrial
fibrillation, Atrial flutter, Axillary vein thrombosis, Basal ganglia haemorrhage, Bile duct stone,
Blood loss anaemia, Bradycardia, Brain abscess, Cardiac failure, Cardiac failure acute,
Cardiac failure congestive, Cardiac stress test abnormal, Cardio-respiratory arrest, Cerebral
infarction, Cerebrovascular accident, Chest pain, Cholecystitis, Cholecystitis acute,

ev
Cholelithiasis, Colitis, Coronary artery disease, Coronary artery dissection, Coronary artery
occlusion, Coronary artery thrombosis, Deep vein thrombosis, Dermatitis bullous, Diabetic
ketoacidosis, Diarrhoea, Diplegia, Dyspnoea, Embolic stroke, Empyema, Facial paralysis,
Fluid retention, Gastroenteritis, Gastrointestinal haemorrhage, Haematoma, Haemorrhagic
stroke, Hemiplegic migraine, Hepatic enzyme increased, Hyperglycaemia, Hyponatraemia,

r
Hypoxia, Ischaemic stroke, Laryngeal oedema, Multiple sclerosis, Myocardial infarction, Non-
cardiac chest pain, Oedema peripheral, Pancreatitis, Pancreatitis acute, Pericarditis,
Peripheral artery aneurysm, Peritoneal abscess, Pleuritic pain, Pneumothorax, Post
er
procedural haematoma, Post procedural haemorrhage, Postoperative abscess, Procedural
haemorrhage, Psychotic disorder, Pulmonary embolism, Rash, Rash vesicular, Respiratory
failure, Retinal artery occlusion, Rhabdomyolysis, Rheumatoid arthritis, Schizoaffective
pe
disorder, Seizure, Subarachnoid haemorrhage, Subcapsular renal haematoma, Subdural
haematoma, Tachyarrhythmia, Tachycardia, Thrombocytopenia, Thyroid disorder, Toxic
encephalopathy, Transaminases increased, Transient ischaemic attack, Traumatic intracranial
haemorrhage, Type 2 diabetes mellitus, Uraemic encephalopathy, Uterine haemorrhage,
Vascular stent occlusion, Ventricular arrhythmia
ot

Excluded SAE types (not matching AESI list): Abdominal adhesions, Abortion
spontaneous, Abortion spontaneous incomplete, Accelerated hypertension, Adenocarcinoma
gastric, Adrenal gland cancer, Alcohol abuse, Alcohol poisoning, Alcohol withdrawal
syndrome, Animal bite, Ankle arthroplasty, Ankle fracture, Anxiety, Anxiety disorder, Aortic
tn

stenosis, Appendicitis, Appendicitis perforated, Arteriosclerosis, Asthma, Atelectasis,

Autonomic nervous system imbalance, B-cell small lymphocytic lymphoma, Back injury, Back
pain, Benign prostatic hyperplasia, Bipolar disorder, Breast cancer, Breast cancer stage I,
Breast hyperplasia, Bronchitis, Cartilage injury, Cellulitis, Cervical radiculopathy, Cervical
spinal stenosis, Cervical vertebral fracture, Choroidal neovascularisation, Chronic kidney
rin

disease, Chronic lymphocytic leukaemia, Chronic myeloid leukaemia, Chronic obstructive

pulmonary disease, Clostridium difficile colitis, Clostridium difficile infection, Colon cancer
stage III, Colon injury, Colorectal cancer, Completed suicide, Complicated appendicitis,
Concussion, Confusional state, Constipation, Cough, Craniocerebral injury, Dehydration,
Depression, Diplopia, Diverticular perforation, Diverticulitis, Dizziness, Drug hypersensitivity,
ep

Duodenal ulcer, Duodenal ulcer haemorrhage, Emphysema, Facial bones fracture, Fall,
Feeling hot, Femoral neck fracture, Femur fracture, Fibromuscular dysplasia, Flail chest,
Flank pain, Food poisoning, Foot fracture, Foot operation, Forearm fracture, Fracture
nonunion, Gastric cancer, Gastric perforation, Gastrooesophageal reflux disease, Gout, Gun
shot wound, Head injury, Heart disease congenital, Hepatic cancer metastatic, Hepatic mass,
Pr

Hepatitis A, Hernia, Hiatus hernia, Hip arthroplasty, Hip fracture, Humerus fracture,
Hypertension, Hypertensive emergency, Hypertensive urgency, Hypoglycaemia,

19

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 Hypokalaemia, Hypomagnesaemia, Hypotension, Idiopathic intracranial hypertension,

ed
Immunisation anxiety related reaction, Incarcerated hernia, Incision site pain, Influenza like
illness, Intentional self-injury, Interstitial lung disease, Intervertebral disc degeneration,
Intervertebral disc protrusion, Intestinal obstruction, Intestinal perforation, Intraductal
proliferative breast lesion, Invasive ductal breast carcinoma, Invasive lobular breast
carcinoma, JAMMED RIGHT INGUINAL HERNIA@@, Jaw operation, Joint injury, Knee

iew
arthroplasty, Large intestine perforation, Lead dislodgement, Leiomyosarcoma metastatic,
Leydig cell tumour of the testis, Ligament rupture, Loss of consciousness, Lower limb fracture,
Lung cancer metastatic, Lymphadenopathy, Major depression, Malignant melanoma,
Meningioma, Mental disorder, Metabolic acidosis, Metastases to central nervous system,
Migraine, Multiple injuries, Musculoskeletal chest pain, Nausea, Neck pain, Nephrolithiasis,
Neutropenia, Obstructive pancreatitis, Oesophageal carcinoma, Oesophageal food impaction,

ev
Organising pneumonia, Orthostatic hypotension, Osteoarthritis, Osteochondritis,
Osteomyelitis, Ovarian cyst, Ovarian mass, Overdose, Pancreatic mass, Papillary thyroid
cancer, Paraesthesia, Pelvic neoplasm, Penile cancer, Penile neoplasm, Peritonitis,
Pharyngitis streptococcal, Pleural effusion, Pneumonia, Pneumonia aspiration, Pneumonia
staphylococcal, Pneumonitis, Polymyalgia rheumatica, Postoperative wound infection,

r
Precancerous condition, Prostate cancer, Prostate cancer metastatic, Pulmonary mass,
Pyelonephritis, Pyelonephritis acute, Rectal prolapse, Renal cancer, Renal cell carcinoma,
Renal colic, Retinal detachment, Retinal tear, Rib fracture, Road traffic accident, Salivary
er
gland calculus, Salpingitis, Sepsis, Septic shock, Sexual abuse, Shoulder injury related to
vaccine administration, Skin laceration, Small intestinal obstruction, Speech disorder, Spinal
cord injury cervical, Spinal fusion surgery, Spinal stenosis, Staphylococcal infection,
pe
Streptococcal sepsis, Suicidal ideation, Suicide attempt, Suspected COVID-19, Swelling face,
Syncope, Systemic inflammatory response syndrome, Tendon rupture, Thoracic vertebral
fracture, Thyroidectomy, Toxic shock syndrome, Toxicity to various agents, Transient global
amnesia, Traumatic liver injury, Ulna fracture, Umbilical hernia, Unevaluable event, Urinary
bladder polyp, Urinary tract infection, Urosepsis, Uterine leiomyoma, Uterine prolapse,
Vertigo, Viral pharyngitis, Volvulus, Vomiting, Wound infection, Wrist fracture
ot

Excluded SAE types (efficacy-related endpoints): COVID-19, COVID-19 pneumonia,

SARS-CoV-2 test positive.
tn

Note: Pfizer and Moderna coded all SAEs using the MedDRA coding dictionary; terms here
are reproduced verbatim from the SAE tables. Preferred terms with @@ denote uncoded
terms.
rin
ep
Pr

20

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239
 ed
Supplemental Table 2. Sensitivity analysis

Eventsa Risk difference Risk ratio

per 10,000 participants (95% CI)

iew
(95% CI)
Trial Vaccine Placebo

Serious adverse events of special interestb

Pfizer 52 33 10.1 (-0.4 to 20.6) 1.57 (0.98 to 2.54)

ev
Moderna 87 64 15.1 (-3.6 to 33.8) 1.36 (0.93 to 1.99)

Combined 139 97 12.5 (2.1 to 22.9) 1.43 (1.07 to 1.92)

r
SAEs matching Brighton’s SPEAC COVID-19 AESI listc,d

Pfizer 39 28 5.8 (-3.5 to 15.2) 1.39 (0.82 to 2.37)

Moderna

Combined
65

104
56

84
er5.9 (-10.9 to 22.6)

5.9 (-3.2 to 15.0)

1.16 (0.76 to 1.77)

1.24 (0.89 to 1.72)

pe
a Denominators for Pfizer were 18,801 in the vaccine group and 18,785 in the placebo group,
and for Moderna were 15,185 in the vaccine group and 15,166 in the placebo group.
b This analysis, presented in the main paper, is reproduced here for ease of interpreting the

sensitivity analysis.
c This list does not include the 29 clinical diagnoses Brighton identified as “known to have
ot

been reported [in conjunction with COVID-19] but not in sufficient numbers to merit inclusion
on the AESI list.”
d Standard errors used to estimate 95% CIs were inflated by the factor √[#SAE]/[#patients

with SAE] to account for multiple SAE within patients.

tn
rin
ep
Pr

21

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4125239