HNZ00203542 Appendix 2
Memo
Myocarditis
Date:
22 July 2021
To:
Dr Ian Town, Chair COVID-19 Vaccine Technical Advisory Group
Copy to:
Dr Ashley Bloomfield, Director-General of Health; Jo Gibbs, National Director -
COVID-19 Vaccine & Immunisation Programme
From:
Mr John Tait, Chair – COVID-19 Vaccine Independent Safety Monitoring Board
For your:
Action
Purpose of report
1.
To request further information from the COVID-19 Vaccine Technical Advisory Group (CV-TAG)
regarding their draft recommendations on the use of the Pfizer-BioNTech mRNA COVID-19 vaccine in
the context of the risk of myocarditis and/or pericarditis fol owing vaccination.
2.
To highlight the need for clarity around the roles and responsibilities of the COVID-19 Vaccine
Independent Safety Monitoring Board (CV-ISMB) and CV-TAG in providing advice to the COVID-19
Vaccine and Immunisation Programme (CVIP).
Background and context
3.
Myocarditis is an adverse event of special interest for the COVID-19 vaccines and is being closely
monitored by the Centre for Adverse Reactions Monitoring (CARM), Medsafe and the CVIP.
4.
Medsafe first presented an overview of myocarditis to the CV-ISMB on 27th May, with further updates
provided at subsequent meetings (24th June and 21st July).
Fol owing advice from the CV-ISMB, an M²
monitoring communication was issued by Medsafe on the 9th June (refer Appendix 1).
5.
An update on myocarditis/pericarditis reports was presented to the Board at their regular monthly
meeting (21st July). Up to the 20th July, CARM have received 9 cases of myocarditis, 7 cases of
pericarditis and 5 cases of myopericarditis. Of the 20 cases (10 females, 10 males), the ages ranged
from 24-73 years, with 3 cases in consumers 20-29 years (2 females, 1 male). Fifteen of the reports
have occurred after the second dose and five reports have been received fol owing the first dose.
6.
The Board was updated that Pfizer had updated their Company Core Data Sheet (CCDS) to include
myocarditis as a rare adverse event with the Comirnaty vaccine based on international data and this
information would be updated in the New Zealand data sheet; with Medsafe to publish an alert
Page 1 of 3
HNZ00203542 Appendix 2
communication to update prescribers and consumers. Medsafe usual process is to update monitoring
communications to either state that the signal was dismissed or that it has been incorporated into the
product information.
7.
New Zealand is currently not seeing the higher than background rates of myocarditis observed
overseas in countries such as the United States, however it is noted that this could be due to our
current rol out strategy. The Board was reassured by the current data and at the present time
myocarditis is not viewed as an issue of concern impacting the balance of benefits and risks of
vaccination with Comirnaty in New Zealand.
8.
The Board was made aware of the draft myocarditis recommendations being prepared by CV-TAG;
however, these haven’t been formal y shared with the CV-ISMB for consideration or comment. As the
fol owing points relate to safety concern(s) and the Board are not aware of the literature that has
guided these recommendations, the Board would like to request copies of the relevant data.
People aged 16-29 years receive their second dose of the Pfizer COVID-19 vaccine 8 weeks
after the first dose. Emerging data suggests that a longer interval between doses may reduce
the severity of some side effects while conferring the maximum protection from COVID-19.
People aged 16-29 years who require regular clinical review by a cardiologist are advised to
discuss the risks and benefits of the first and second doses of the COVID-19 vaccine with their
healthcare team.
9.
The Board expressed concern around ensuring that both the CV-ISMB and the CV-TAG were aligned,
especial y when information/advice is being provided to the Director-General, Ministers and to inform
policy decisions for the CVIP.
Recommendations
10. The CV-ISMB requests the opportunity to review the recommendations proposed by CV-TAG
regarding the risk of myocarditis and/or pericarditis fol owing vaccination, with the relevant
supporting evidence.
11. A review is conducted to delineate the roles of the CV-TAG and the CV-ISMB in providing advice to
inform policy decisions made for the CVIP.
Mr John Tait
Chair of the COVID-19 Vaccine Independent Safety Monitoring Board
Page 2 of 3
HNZ00203542 Appendix 2
Myocarditis Monitoring Communication [link]
Page 3 of 3
HNZ00203542 Appendix 2
Memo
Date:
13 August 2021
To:
Susan Kenyon, Manager Clinical Risk Management, Medsafe
From:
Muireann Walton
Subject:
Comirnaty and thrombocytopenia
DESCRIPTION
As of 31st July 2021, there have been 5 reports of thrombocytopenia following Comirnaty
administration in New Zealand.
The purpose of this memo is to review available information on thrombocytopenia and Comirnaty
and to consider whether any action is required by Medsafe.
NATURE OF THE SAFETY CONCERN
THROMBOCYTOPENIA
Thrombocytopenia is defined by a low platelet count of less than 150 x 109/L [1]. Patients with
platelet counts greater than 50 x 109/L are often asymptomatic, while a platelet count of 10-30 x
109/L may lead to mild symptoms and bleeding with minimal trauma. In rare cases, the number of
platelets can be so low (<10 x 109 per L), that spontaneous and significant bleeding, bruising and
purpura can occur [2]. This can be life threatening and is considered a haematological emergency.
Various disorders and diseases are associated with thrombocytopenia and it is often the first sign of
underlying conditions such as malignancies, infectious diseases, and autoimmune disorders. It can
also be a side effect of several medicines.
Thrombocytopenia can arise from the following:
• decreased platelet production by bone marrow
• increased platelet destruction
• platelet splenic sequestration;
OR • Combination of the above factors [2]
A classification of thrombocytopenia by these mechanisms is shown in table 1.
Notably, immune or idiopathic thrombocytopenia purpura (ITP) has been reported after several
vaccines and it has been linked to the MMR vaccine, with a risk window of six weeks [3] The
aetiology of vaccine-related thrombocytopenia is considered immune related with antibodies
detected on platelets in approximately 79% of cases[3]. The risk of vaccine-related
thrombocytopenia is extremely low, and most cases are self-limiting and resolve with standard
treatment.
Page 1 of 19
HNZ00203542 Appendix 2
Figure 1 below shows the number of people discharged with thrombocytopenia per financial year
since 2015/216. Since 1 July 2015, there have been 13,967 people aged 15 years and older
discharged with a diagnosis of thrombocytopenia and 22,757 hospitalisations. The number is
consistent, with 2,000, -3,000 people discharged from hospital with thrombocytopenia annually.
Figure 1: Number of people (≥15 years) discharged with thrombocytopenia, by financial year, 2015/16 to
2021/22. Source: COVID-19 Vaccination Events Qlik app, updated 1 August 2021 (accessed 1 August 2021).
NEW ZEALAND HOSPITALISATION RATES OF THROMBOCYTOPENIA
The incidence rate of thrombocytopenia per 100,000 persons from 2008 to 2019 was calculated to
understand the prevalence of the conditions in the New Zealand population by age group. This data
was generated by Associate Professor Helen Petousis-Harris and colleagues from the University of
Auckland for the SAFE project and is shown below (figures 2-4).
The background rate for the resident New Zealand population was estimated using health, tax,
education data tables, from the Integrated Data Infrastructure (IDI) which is carefully managed by
Statistics New Zealand (Stats NZ). The counts of thrombocytopenia were selected using the
following ICD10 clinical codes: D693, D694, D695, D696, D820, M311.
As per figure 2, there is a clear upward trend of thrombocytopenia with age, with the highest
incident rate seen in those aged 80+. Conversely, there is no apparent trend of thrombocytopenia
hospitalisation rates by different ethnicities or gender as shown in figures 3 and 4.
Page 3 of 19
HNZ00203542 Appendix 2
Thrombocytopenia hospitalisation rate per 100,000
person years by age group, 2008-2019
140.00
120.00
ears
y 100.00
on
ers
80.00
p
0
60.00
.00
40.00
20.00
er 100
p
0.00
0-9
10-19
20-29
30-39
40-49
50-59
60-69
70-79
80+
Rate
Age group
Figure 2. Thrombocytopenia hospitalisation rate per 100,000 persons by age group from 2008-2019. Source: Preliminary background rates from the UoA SAFE study.
Thrombocytopenia hospitalisation rate per 100,000
person years by prioritised ethnicity, 2008-2019
45.00
40.00
ears
y 35.00
on 30.00
ers
p 25.00
,000 20.00
15.00
er 100 10.00
p
5.00
Rate
0.00
Asian
European
MELAA
Māori
Other
Pasifika
Figure 3. Thrombocytopenia hospitalisation rate per 100,000 persons by ethnicity from 2008-2019.
Source: Preliminary background rates from the UoA SAFE study.
Page 4 of 19
HNZ00203542 Appendix 2
Table 3. Cases of thrombocytopenia after Comirnaty vaccine reported spontaneously to CARM (up to 31 July 2021).
AEFI-A- Vaccine
Date of
TTO
Gender Ethnic
Ag
Do
Event
Seriousness
Reported Outcome
Causality
date
reaction
(hh:mm)
group
e
se
severity
assessed
no.
?
AEFI-A-
22/03/202
7/06/2021 77 days
Male
European
9(2)(a)
2
Thrombocytopenia
Hospitalisation
Severe
Not yet
Unlikely
004921
1
12:50
(2 months,
or other
purpura (ITP)
recovered
1:56 pm
16 days)
AEFI-A-
2/06/2021 5/06/2021 3 days
Female
European
69
1
Thrombocytopenia
Medically
Moderate Not yet
Unlikely
005224
8:00
or other
Significant
recovered
AEFI-A-
26/05/202
5/06/2021 10 days
Male
European
9(2)(a)
1
Thrombocytopenia
Emergency Care Moderate Recovering Probable
005344
1
17:45
or other
AEFI-A-
6/07/2021 12/07/202
6 days
Female
European
76
1
Thrombocytopenia
Hospitalisation
Severe
Recovering Possible
007072
1 11:19
or other
AEFI-A-
24/06/202
21/07/202
27 days
Male
European
9(2)(a)
2
Thrombocytopenia
Hospitalisation
Severe
Recovering Unclassifi
008370
1
1 15:49
or other
purpura (ITP)
ed-
under
investigat
ion
Page 7 of 19
HNZ00203542 Appendix 2
REVIEW OF THE AVAILABLE INFORMATION
The above cases (table 3) reported to CARM are presented below by age, gender, ethnicity, and time
to occurrence (figure 6). Due to the low number of cases reported, it is difficult to detect any trends,
however it appears that thrombocytopenia occurs most commonly after dose 1 than dose 2 and in
older age groups.
Figure 6. Reports of thrombocytopenia received by CARM presented according to age, gender, ethnicity and time
to occurrence.
Source: COVID-19 AEFI Qlik app (accessed 31 July 2021). Terms chosen “thrombocytopenia” or “thrombocytopenia purpura”.
OBSERVED VERSUS EXPECTED (O/E) ANALYSES
Observed versus expected analyses was conducted to determine whether the reported counts of
spontaneously reported thrombocytopenia following Comirnaty vaccination are higher than
expected based on the “natural” background rates of the events in the absence of vaccine exposure
in New Zealand.
This analysis is presented in table 4 and shows that the rates of thrombocytopenia following
vaccinations did not exceed the number of events we would expect in the background.
There are some limitations to this method and caution is needed when interpreting the O/E rates.
The expected rate relies on hospital discharge data from 2008-2019 while the observed counts are
likely to be underestimated due to underreporting that occurs with spontaneous surveillance.
Incomplete reporting and lags in reporting can further underestimate observe counts.
Page 8 of 19
HNZ00203542 Appendix 2
Overall, the clinical presentations and the favourable response to “ITP‐directed” therapies in most of
the treated patients, such as corticosteroids and IVIG suggest an antibody‐mediated platelet
clearance mechanism that is operative in ITP.
Author’s conclusions: the authors concluded that an association between ITP and the mRNA COVID-
19 vaccines could not be ruled out, especially in those with onset 1-2 weeks after vaccination. One
reported case included a patient with a normal platelet count a week prior to vaccination who later
developed symptoms 13 days post vaccination which is compatible with vaccine related secondary
ITP. Further, all but one case reported occurred after the fist dose of the mRNA vaccines. The
authors note that if the vaccine was unrelated to the development of ITP, case occurrences would
likely be divided more evenly between the two doses.
However, they noted that approximately 50,000 adults in the US are diagnosed with ITP per year and
that the incidence of an immune‐mediated thrombocytopenia post vaccination appears either less
than or roughly comparable to what would be seen if the cases were coincidental following
vaccination. Additional surveillance is therefore needed to determine the true incidence of
thrombocytopenia post vaccination.
Welsh, K. L. et al., Thrombocytopenia including immune thrombocytopenia after receipt of mRNA
COVID-19 vaccines reported to the Vaccine Adverse Event Reporting System (VAERS) [7]
Welsh et al., describe a case-series study of thrombocytopenia reported to VAERS after vaccination
with mRNA COVID-19 vaccines (up to 4 February 2021). Fifteen cases of thrombocytopenia were
identified among 18,841,309 doses of Pfizer-BioNTech COVID-19 Vaccine and 13 cases among
16,260,102 doses of Moderna COVID-19 Vaccine. Of these cases, 15 were female, 11 were male and
in two the sex was not reported. The median age of cases was 48.5 years and the reported onset
time ranged from 1-23 days (median 5.5 days) after vaccination. All but two cases were reported
after the first dose of the vaccines. Two patients in this case series died, one of whose death was
attributed to intracranial haemorrhage secondary to ITP. The second patient who died reportedly
experienced acute myocardial infarction, a pulmonary embolism, and thrombocytopenia although
minimal additional case details are available.
The reporting rate of thrombocytopenia was 0.80 per million doses for both vaccines. Based on an
annual incidence rate of 3.3 ITP cases per 100,000 adults in the US, the study found that the
observed number of thrombocytopenia cases following administration of mRNA COVID-19 vaccines
was not greater than the number of ITP cases expected.
Author’s conclusions: the author’s concluded that the number of reported cases of
thrombocytopenia to VAERS does not currently suggest a safety concern attributable to mRNA
COVID-19 vaccines.
King, E. R. et al., A Case of Immune Thrombocytopenia After BNT162b2 mRNA COVID-19
Vaccination [8]
King et al., describe a case of a 39-year-old female who developed a petechial rash on her trunk,
legs and arms, and fatigue and muscle aches 3 days after receiving her second dose of Comirnaty.
She was admitted to hospital and a peripheral smear showed profound thrombocytopenia, with a
platelet count of 1000/µL. Several causes of ITP such as viral hepatitis HIV and H. pylori were tested
during the patient’s hospital stay. All tests returned negative results. Evans syndrome was also ruled
out.
Page 10 of 19
HNZ00203542 Appendix 2
The patient was treated with 2 units of platelets, 2 infusions of IV immunoglobulin, and IV
methylprednisolone. Her platelet count increased to 92,000/µL on the day of discharge and she was
prescribed a tapered dose of oral prednisone. One day later, her rash had resolved, and her platelet
count was 243,000/µL. The patient recovered completely with no complications.
The patient’s medical history included polycystic ovary syndrome, for which she took norgestimate-
ethinyl estradiol. She had no pertinent family or travel history and no history of use of tobacco or
alcohol or of substance abuse. The patient had a complete blood count (CBC) and differential 5
months before, which was within normal limits. At that time, she also was tested for COVID-19
antibodies; the results were negative. The patient did not have any illnesses or known COVID-19
exposures before the incident reported here.
Author’s conclusions: the author’s concluded that due to the lack of medications or conditions that
could have caused the condition in the patient, the development of ITP was likely due to the
Comirnaty vaccine.
Ganzel, C. and E. Ben-Chetrit (2021) – Immune Thrombocytopenia Following the Pfizer-BioNTech
BNT162b2 mRNA COVID-19 Vaccine [9]
Ganzel et al., report a case of a 53-year-old male who was admitted to Shaare Zedek Medical Center
in Jerusalem due to epistaxis and low platelet count 2 weeks after receiving the first dose of the
Pfizer COVID-19 vaccine.
Medical history included, morbid obesity, diabetes, and hypertension for which he was treated with
lercanidipine, losartan, doxazocin, hydrochlorothiazide and aspirin. He took two tablets of
levofloxacin for suspected otitis one week prior to admission. He had previously taken levofloxacin
on several occasions.
Physical examination revealed wet purpura on his palate and petechial and purpuric rash on the
trunk and limbs. His blood count and smear were remarkable for severe thrombocytopenia: 1×103
/μL (normal range 150–400×103 / μL). He was diagnosed with immune thrombocytopenia purpura
(ITP). The patient was treated with dexamethasone 20 mg/d and intravenous immunoglobulins, 1
g/kg, with a gradual increase in platelets. Five days later his platelet counts normalised. Due to the
severity of the thrombocytopenia, the second dose of the vaccine was not given to the patient.
Author’s conclusions: the authors suggest that there may be a temporal relationship between the
development of ITP and receipt of the Comirnaty vaccine in this patient. However, they note that
levofloxacin has been associated with severe thrombocytopenia. They considered this causality
unlikely due to the patient’s prior exposure.
Kragholm, K., et al., (2021). Thrombocytopenia after COVID-19 vaccination [10]
This study reviewed cases of thrombocytopenia reported from the North Denmark Region (capture
population≈600,000 inhabitants) among healthcare professionals ≤65 years of age following
PfizerBioNTech/Moderna (N = 11,689) or the Oxford-AstraZeneca (N = 16,509) COVID-19
vaccinations.
Of 2,130 individuals with post-vaccination platelet measurements available, 50 (40 women and 10
men) had thrombocytopenia (platelet count <145 × 109/L in men and <165 × 109/L in women).
Among 1,873 women, 24/813 (3.0 %) vaccinated with the Oxford-AstraZeneca COVID-19 vaccine
versus 16/1060 (1.5 %) vaccinated with PfizerBioNTech/Moderna COVID-19 vaccines had
thrombocytopenia, odds ratio [95 % confidence interval] for thrombocytopenia of 1.99 [1.05–3.76]
for Oxford-AstraZeneca versus PfizerBioNTech/Moderna. Among 257 men, the corresponding odds
Page 11 of 19
HNZ00203542 Appendix 2
ratio [95 % confidence interval] was 0.49 [0.14–1.79]. Severe thrombocytopenia (platelet count
<50 × 109/L) was seen in three patients vaccinated with Oxford-AstraZeneca (all women between 50
and 60 years of age) versus none among the PfizerBioNTech/Moderna vaccines
Author’s conclusions: the author’s concluded that thrombocytopenia appears to be significantly more
frequent among women vaccinated with Oxford-AstraZeneca COVID-19 when compared to those
vaccinated with the PfizerBioNTech/Moderna COVID-19 vaccines. Cases of severe thrombocytopenia
were seen only among women vaccinated with the Oxford-AstraZeneca COVID-19 vaccine.
PFIZER/BIONTECH COMIRNATY PSUR MAY 2021
The sponsor is currently submitting a monthly Periodic Safety Update Report (PSUR) for Comirnaty
to Medsafe. Their latest reporting interval is 1 June 2021 to 30 June 2021.
Immune Thrombocytopenia (ITP) was previously reviewed by the sponsor in January 2021 and
following further reports in the safety database and a request from other regulators (EMA and FDA),
it has been re-reviewed in the current PSUR.
Table 5 shows the sponsor’s evaluation of new information received related to thrombocytopenia
during the reporting interval.
The sponsor has also provided a tabulated summary (during the interval and cumulatively) of
serious and non-serious adverse reactions from post-market spontaneous data sources (see Table
6).
Table 5. PSUR June 2021 (1 May 2021 to 31 May 2021): thrombocytopenia events
Page 12 of 19
HNZ00203542 Appendix 2
Table 6. Interval and cumulative count of various MedDRA PT related to thrombosis/thromboembolic
events from post-market spontaneous data
MedDRA PT
Serious
Interval
Cumulative
Immune thrombocytopenia
55
210
Thrombocytopenia
158
555
Thrombocytopenia purpura
11
21
Thrombotic thrombocytopenia purpura
14
25
Observed versus expected analysis:
The sponsor performed an observed versus expected analysis (O/E) for ITP (see table 7 for preferred
terms used). O/E is conducted by the sponsor to determine whether the reported counts of
spontaneously reported AESIs are higher than expected based on the background rates of the AESIs
in the absence of vaccine exposure.
The analysis below is reported cumulatively for the period since the vaccine was granted provisional
approval (9 December 2020 in the US) and for the June reporting interval (1 June 2021 to 30 June
2021). The exposure time was calculated using a 21-day risk window and no risk window for both
the reporting interval and the cumulative period. The risk window is defined as the period which one
is expected to be at risk of a given event if there is a causal association between the event and the
vaccine. The 21-day risk window is an estimate assuming a more acute onset, while the no risk
window is a more latent onset.
Note that the limitation to O/E analysis is that the observed counts are likely to be underestimated
due to underreporting that occurs with spontaneous surveillance. Incomplete reporting and lags in
reporting can further underestimate observe counts.
The background rates used in this analysis are derived from US healthcare databases and regional
studies published in the literature. The O/E analysis is presented in tables 8-12.
Page 13 of 19
HNZ00203542 Appendix 2
Table 7. The preferred terms used to identify the spontaneously reported AESI
AESI
PTs
Idiopathic thrombocytopenia purpura,
Immune thrombocytopenia, thrombocytopenia,
autoimmune thrombocytopenia
thrombotic thrombocytopenia purpura,
thrombocytopenia purpura
Table 8. O/E ratios of spontaneously reported events of ITP, autoimmune thrombocytopenia
AESI
21-day risk
No risk window
21-day risk
No risk window
window
(cumulative)
window
(interval)
(cumulative)
(interval)
O/E
95% CI
O/E
95% CI O/E
95% CI O/E
95% CI
Ratio
Ratio
Ratio
Ratio
ITP
0.104
0.097,
0.035
0.033,
0.143
0.125,
0.166
0.145,
0.112
0.038
0.163
0.189
Table 9. O/E analyses of spontaneously reported events of ITP, autoimmune thrombocytopenia (cumulative)
AESI
Background rate
Obs cases
Exp cases, 21-
Exp cases,
per 100,000
(cumulative)
day risk window no risk
Person Years (PY)
(cumulative)
window
(cumulative)
ITP
21.46
764
7,334.8
21,785.6
Table 10. O/E analyses of spontaneously reported events of ITP, autoimmune thrombocytopenia (interval)
AESI
Background rate
Exp cases,
per 100,000
Exp cases, 21-
Obs cases
no risk
Person Years (PY)
day risk window
(interval)
window
(interval)
(interval)
ITP
21.46
255
1,574.0
1,357.2
Table 11. Age-stratified O/E analyses of spontaneously reported events of ITP, autoimmune thrombocytopenia
in European Economic Area Countries and the United States Inputs, 21-Day Risk Window, Interval
AESI
<17 years
18-24 years
25-49 years
50-59 years
60-69 years
70+ years
Obs
Exp
Obs
Exp
Obs Exp
Obs Exp
Obs Exp
Obs Exp
cases
cases cases
cases cases cases cases cases cases cases cases cases
ITP
1
14.4
2
26.4
30
172.8 14
188.4 31
202.2 64
281.7
Page 14 of 19
HNZ00203542 Appendix 2
Table 12. Age-stratified O/E analyses of spontaneously reported events of ITP, autoimmune thrombocytopenia
in European Economic Area Countries and the United States Inputs, 21-Day Risk Window, Cumulative
AESI
<17 years
18-24 years
25-49 years
50-59 years
60-69 years
70+ years
Obs
Exp
Obs Exp
Obs Exp
Obs Exp
Obs Exp
Obs Exp
cases
cases cases cases cases cases cases cases cases cases cases cases
ITP
7
47.5
10
114.6 89
759.8 54
884.9 71
1,331 266
3,537.
.2
8
Sponsor’s conclusion: the sponsor concluded that no new safety signals have emerged based on
this review of the cases, or of the O/E analysis performed. Surveillance will continue.
REGULATORY REVIEW
MEDICINES AND HEALTHCARE PRODUCTS REGULATORY AGENCY (UK) (MHRAUK)
The MHRA publishes a weekly summary of Yellow Card reporting. The most recent report covers the
period 9 December 2020 to 28 July 2021. The report did not note any specific comments or issues
related to Comirnaty vaccinations and thrombocytopenia. An extract of their latest Comirnaty ‘Vaccine
Analysis Print’ is shown below in table 12 and represents all UK spontaneous reports relating to
thrombocytopenia received between 9 December 2020 and 21 July 2021 for the Comirnaty vaccine.
Table 13. All UK spontaneous reports with various MedDRA PT for thrombocytopenia received between 9/12/20
and 21/07/21 for the Comirnaty vaccine.
European Medicines Agency (EMA)
The EMA publishes a monthly COVID-19 vaccine safety update for Comirnaty. In their COVID-19
safety update for Comirnaty dated 29 March 2021, immune thrombocytopenia (ITP) was mentioned.
The EMA stated that “for all COVID-19 vaccines used in the EU, a specific PRAC assessment of
immune thrombocytopenia (ITP, low blood platelet levels that can lead to bruising and bleeding) as
a suspected side effect is ongoing. The PRAC assessment of the ITP cases reported to EudraVigilance
(see section 3) for Comirnaty from vaccination campaigns did not reveal a pattern confirming a
causal relationship of ITP with Comirnaty. In some cases, the time-to-onset of symptoms was
inconsistent with the possibility of a vaccine-mediated immune reaction”.
US Food and Drug Administration (FDA)
Thrombocytopenia is not listed on the 'FDA Comirnaty fact sheet and product labelling '.
The Advisory Committee on Immunisation Practices (ACIP) holds three meetings each year at the
Centres for Disease Control and Prevention (CDC). At a meeting on the 23 June 2021, preliminary
results from their VSD Rapid Cycle Analysis (RCA) were presented. ITP was chosen as an AESI for this
analysis and the results are shown below (see table 13).
Page 15 of 19
HNZ00203542 Appendix 2
Table 14. Outcome events in the 21-day risk interval after either dose of any mRNA vaccine
compared with outcome events in vaccinated comparators on the same calendar day (as of 12 July
2021).
Therapeutic Goods Administration (TGA)(Australia)
Immune thrombocytopenia is continuously monitored as an AESI by the TGA.
The TGA releases weekly COVID-19 vaccine safety report. There is no mention of thrombocytopenia
in their weekly report for Comirnaty dated 29 July 2021.
COMMENTS: International regulatory authorities have not recognised thrombosis/thromboembolic
conditions as adverse events associated with Comirnaty.
According to publicly available information, various regulatory authorities are continually monitoring
for thrombocytopenia as an AESI. To date, they have not seen a higher rate than usual.
PUBLIC INTEREST
There is significant public and media interest in this topic, with accounts of thrombocytopenia
following COVID-19 vaccination published on social media platforms.
For example, public alarm was heightened in the US following the death of a 56-year-old male from
a haemorrhagic stroke, secondary to ITP who had received the Comirnaty vaccine. This was reported
in several US news outlets including the New York Times [11] [12].
EXPERT ADVICE
No expert advice has been sought by external committees or other experts yet.
PROPOSED ACTIONS
Medsafe should continue to monitor for thrombocytopenia through routine pharmacovigilance.
This includes monitoring New Zealand AEFI reports, plus information from the published literature,
company reports (e.g., PSURs), and reports and reviews from other regulatory authorities.
CONCLUSIONS
In conclusion, there is currently insufficient information to confirm a possible signal of
thrombocytopenia with the use of Comirnaty.
As of 31 July 2021, CARM has received 5 cases of thrombocytopenia following administration of
Comirnaty. Two of the cases were reviewed as unlikely related to the vaccine, two were possibly
related and one is unclassified, awaiting further investigation. Although a temporal association
Page 16 of 19
HNZ00203542 Appendix 2
between the vaccine and thrombocytopenia was found in some cases, this does not necessarily
mean there is causality.
Thrombocytopenia is currently being monitored as an AESI by Medsafe and other global regulators
such as the MHRAUK, FDA, EMA and TGA. To date, a causal relationship has not been established
between the AESI and Comirnaty by any global regulatory authority.
Further, the observed versus expected analysis performed using New Zealand data, indicates that
there has not been an increase in the rate of thrombocytopenia compared to the background rate.
This is also supported by the sponsor’s O/E analysis and literature based on global data.
It is recommended that Medsafe continue to monitor this issue through routine pharmacovigilance
activities. This includes monitoring New Zealand case reports, company safety reports, action from
other regulators and information in the literature.
Page 17 of 19
HNZ00203542 Appendix 2
RECOMMENDATIONS
It is recommended that:
1.
This topic is presented to the COVID-19 Vaccine Independent Safety
Yes
Monitoring Board (ISMB).
2.
This topic is monitored through routine pharmacovigilance.
Yes
REFERENCES
1. Gauer, R., & Braun, M. M. (2012). Thrombocytopenia.
American family physician,
85(6), 612-622.
2. Stasi, R. (2012). How to approach thrombocytopenia.
Hematology 2010, the American Society of
Hematology Education Program Book,
2012(1), 191-197.
3. France EK, Glanz J, Xu S, et al. Risk of immune thrombocytopenic purpura after measles‐
mumps‐rubella immunization in children.
Pediatrics. 2008;121:e687‐e692
4. Bhattacharjee, S., & Banerjee, M. (2020). Immune thrombocytopenia secondary to COVID-19: a
systematic review.
SN comprehensive clinical medicine, 1-11.
5. SPEAC D2.5.2.1-Thrombocytopenia-Case-Definition-Companion-Guide V1.0 format12065-1.pdf
(brightoncollaboration.us)
6. Lee, E. J., Cines, D. B., Gernsheimer, T., Kessler, C., Michel, M., Tarantino, M. D., ... & Bussel, J. B. (2021).
Thrombocytopenia following Pfizer and Moderna SARS‐CoV‐2 vaccination. American Journal of
Hematology.
7. Welsh, K. J., Baumblatt, J., Chege, W., Goud, R., & Nair, N. (2021). Thrombocytopenia including immune
thrombocytopenia after receipt of mRNA COVID-19 vaccines reported to the Vaccine Adverse Event
Reporting System (VAERS).
Vaccine,
39(25), 3329-3332.
8. King, E. R. and E. Towner (2021). "A Case of Immune Thrombocytopenia After BNT162b2 mRNA COVID-
19 Vaccination." The American journal of case reports 22: e931478 DOI:
https://dx.doi.org/10.12659/AJCR.931478
9. Ganzel, C. and E. Ben-Chetrit (2021). "Immune Thrombocytopenia Following the Pfizer-BioNTech
BNT162b2 mRNA COVID-19 Vaccine." The Israel Medical Association journal : IMAJ 23(6): 341
10. Kragholm, K., Sessa, M., Mulvad, T., Andersen, M. P., Collatz-Christensen, H., Blomberg, S. N., ... &
Sogaard, P. (2021). Thrombocytopenia after COVID-19 vaccination.
Journal of Autoimmunity,
123,
102712.
11. Weintraub K. Death of Florida doctor after receiving COVID‐19 vaccine under investigation.
USA Today. Published January 6,
2021. https://www.usatoday.com/story/news/health/2021/01/06/death-florida-doctor-
following-pfizer-covid-19-vaccine-under-investigation-gregory-michael/6574414002/
12. 2. Grady D, Mazzei P. Doctor's death after Covid vaccine is being investigated. The New York
Times. Published January 12, 2021. https://www.nytimes.com/2021/01/12/health/covid-
vaccine-death.html
Page 18 of 19
HNZ00203542 Appendix 2
Prepared by: Muireann Walton
12 August 2021
APPENDIX 1.
COVID-CARM CASE NARRATIVES
AEFI-A-004921: is a report of a 25-year-old male who developed symptoms of idiopathic thrombocytopenia
purpurea (ITP) approximately 77 days post 2nd dose of the Comirnaty vaccine (reporter unsure of exact dates).
Platelets <10 - Hospital review reports that had viral URTI week prior to onset. The Medical Assessor classified
the incidence of ITP as unlikely related to the vaccine.
AEFI-A-005224: is a report of a 69-year-old female who started feeling unwell with nausea, weakness, lethargy
and flu-like symptoms 4 days following the first dose of the Comirnaty vaccine. She was seen on 11/06 (Day 10
post vaccination) and identified as having a UTI.
Past medical history; Known myelodysplastic syndrome, on Ruxolitinib 5mg twice daily since October 2020,
which has been holding her blood and platelet count at a much better level. On fortnightly bloods, and sudden
drop in Hb and platelets in last sample, with also worsening neutropenia.
Treated empirically with nitrofurantoin 50mg qds which she took for 3 days until 14/06. Urine sample sent off on
14/06 - no infective organism. 14/06 had bloods taken : Observation date/time: 14/06/2021 16:15
HAEMOGLOBIN: *** 89 g/l (115 - 155) HCT (first: PCV): *** 0.27 L/L (0.35 - 0.46) MCV: 99 fL (80 - 99) MCH: 33 pg
(27 - 33) WBC: *** 1.6 x10E9/L (4.0 - 11.0) DIFFERENTIAL Neut Seg: *** 0.6 x10E9/L (1.9 - 7.5) Lymphocyte: *** 0.8
x10E9/L (1.0 - 4.0) Monocyte: 0.2 x10E9/L (0.2 - 1.0) Eosinophil: < 0.1 x10E9/L. Seriousness details classified as
thrombocytopenia and anaemia with neutropenia. The Medical Assessor classified the incidence of
thrombocytopenia as unlikely related to the vaccine.
AEFI-A-005344: is a report of a 75-year-old male who had symptoms of abdominal pain/discomfort, nausea
and pain in the legs 10 days following the first dose of the Comirnaty vaccine. He also developed neutropenia,
lymphopenia, thrombocytopenia and arthralgia.
Past medical history; 2010 rectal ca, high transverse resection Dukes B stable cyst on the pancreas
The case was classified as ‘non-serious’ and ‘moderate severity’.
The Medical Assessor classified the incidence of thrombocytopenia as probably related to the vaccine.
AEFI-A-007072: is a report of a 76-year-old female who developed gastrointestinal bleeding and bruising due
to severe thrombocytopenia 6 days following the first dose of the Comirnaty vaccine. Patient required high dose
steroids, intravenous immunoglobulin and platelet transfusions. Dry Cough preceding 5/52 (pre-dates
Vaccination). Commenced Prednisone 2/7 before onset of blood-stained stools (subsequently found Platelets
=16). Noted to have had COVID vaccination 6/7 prior to onset of ITP. Found +ve for RSV as cause of cough.
Question mark of whether viral infection as cause of ITP.
The Medical Assessor classified the incidence of thrombocytopenia as probably related to the vaccine.
AEFI-A-008370: is a report of a 92-year-old male who noticed significant bruising with slight knocks. Presented
to ED, found to have thrombocytopenia. Thought to be ITP. Has been discharged from hospital - ongoing
condition unknown. Case still under review by the Medical Assessor - requested follow up information.
Past medical history; AF Sigmoid polyps Glaucoma BPH Falls
Page 19 of 19
HNZ00203542 Appendix 2
USAGE DATA
In New Zealand, Comirnaty is approved for use in individuals aged 12 years and older. Recently the
Comirnaty paediatric vaccine has been approved for usage in those aged 5-11 years. The New
Zealand immunisation programme started on 20 February 2021 with border and MIQ workers and
the people they live with. The programme was expanded in stages to include the general population
over aged 12 years and over. In December 2021 the booster roll-out began in people aged 18 years
and over. From January 2022, the paediatric vaccine roll-out began in children aged 5-11 years.
As at January 21 2022, 8,892,113 doses of Comirnaty have been delivered in New Zealand to eligible
age groups, this includes 51,759 first doses to those aged 5-11 years.
DATA SHEETS
ME/CFS is not mentioned in the Comirnaty data sheet.
This information is consistent with international data sheets.
SOURCE OF SAFETY CONCERN
There has been public interest in the safety of Comirnaty in those with ME/CFS.
REVIEW OF THE AVAILABLE INFORMATION
ME/CFS [1]
ME/CFS is characterised primarily by persistent and debilitating fatigue. ME/CFS can vary in severity
from mild to very severe. About one fifth of patients report severe presentations. There is no
universally accepted definition for ME/CFS. The usage of the term myalgic encephalomyelitis is
questioned as there is limited pathological evidence of brain inflammation in patients with ME/CFS.
Similarly, the term Chronic Fatigue Syndrome is contested by those with ME/CFS as it is viewed as
too broad. ME/CFS is the most commonly accepted name.
The National Institute for Health and Care Excellence guidelines for physicians suggest that
debilitating fatigue that is worsened by any degree of physical, cognitive, or emotional activity, and
is not relieved by rest; post-exertional malaise, unrefreshing sleep and/or disturbed sleep; and
cognitive difficulties must all be present for patients to be diagnosed with the disorder. Diagnosis
must also include the exclusion of an alternative diagnosis and the persistence of symptoms for 3
months.
Research into the aetiology of ME/CFS remains inconclusive. Viral infection has been identified as a
cause in some cases, but ME/CFS is distinct from post-illness fatigue and there is limited
understanding of how infection results in the chronic nature of ME/CFS.
Importantly ME/CFS has a significant impact on the quality of life of an individual suffering from it.
ME/CFS patients report delayed diagnosis and a lack of support from healthcare professionals.
Previous misclassifications of ME/CFS have exacerbated prejudices against those with ME/CFS
leading to inadequate healthcare. The Associated New Zealand ME Society (ANZMES) reports that
the prevalence of ME/CFS in New Zealand is 1 in 250 adults and 1 in 134 children or adolescents.
This totals to 25,000 individuals with ME/CFS in New Zealand.
Page 2 of 7
HNZ00203542 Appendix 2
Reports of ME/CFS in New Zealand following Comirnaty
As at 21 January 2022, there have been 57 reports of suspected ME/CFS exacerbation (relapse or
flare up of symptoms) in people previously diagnosed with ME/CFS. Eighty-eight percent of reports
were in females, the remaining were in males, which mirrors prevalence of ME/CFS in the general
population. The average age in reporters was 48 years of age, the youngest patient was a 22-year-
old female, and the oldest patient was an 80-year-old female. Two of the reports were classified as
persisting disability, one was classified as requiring hospitalisation, one was classified as medically
significant, and seven were classified as serious, by the reporters. All other reports were classified as
non-serious.
In addition, there have been 9 reports of new onset of ME/CFS or ME/CFS-like symptoms following
administration. Of these 9 reports, 3 have been reported by a healthcare professional and denote
that no other viral triggers were noted prior to ME/CFS onset.
Page 3 of 7
HNZ00203542 Appendix 2
Data from international regulators and government departments on ME/CFS
There were no data identified from other regulators and government departments on the subject of
ME/CFS.
Company Summary Monthly Safety Reports
Each month, Medsafe receives summary safety reports from Pfizer for the Comirnaty vaccine. Thirty-
one cases of ME/CFS were identified by the sponsor during the 29 October through 15 December
2021 reporting period.
The sponsor also produced an observed versus expected (O/E) analysis of ME/CFS. The cumulative
results of the analysis can be found in Table 1. O/E analyses are used to determine whether the
numbers of spontaneously reported adverse events are higher than the expected background rates
observed in the absence of vaccine exposure. The background incidence rates for ME/CFS were
based on the incidence of ME/CFS in Olmsted County Minnesota, as estimated using the Rochester
Epidemiology Project population database. [2] It is unclear from the SMSR is the cases reported are
new ME/CFS cases or relapsed/exacerbated cases of ME/CFS.
The O/E ratios for both 21-day and 42-day risk windows were <1, suggesting that the number of
reported cases is not higher than the expected background rate.
Table 1: Summary of O/E analysis ME/CFS in October PSUR.
Review of available information:
There is limited literature examining the exacerbation or onset of ME/CFS following vaccination with
Comirnaty. Anecdotal evidence and surveys regarding the matter have circulated somewhat widely
among the general public but are not peer-reviewed or published in scientific journals.
Page 4 of 7
HNZ00203542 Appendix 2
ANZMES Survey [3]
ANZMES conducted a survey at the request of the New Zealand ME/CFS community, to measure the
effects of ME/CFS patients following their Comirnaty vaccinations.
Within the survey 191 vaccinated respondents answered a question regarding their ME/CFS status
post vaccination. 70 reported a temporary worsening of their symptoms, 48 reported worsening of
their symptoms in a relapse, and 8 reported worsening their symptoms beyond anything previously
experienced. Moreover, 43 reported no change and 20 reported an improvement in symptoms.
Cumulatively this means that 66% of respondents reported a worsening of symptoms.
The survey also asked a question about the state of general illness/wellness in ME/CFS patients
following Comirnaty vaccination. reports that out of 395 respondents with ME/CFS and 144 with
fibromyalgia (comorbid with ME/CFS in an unspecified number of cases), 3.1% reported a significant
worsening of their symptoms, 19.8% reported a worsening of symptoms yet to return to baseline,
and 32.9 reported a worsening of symptoms followed by a return to baseline. Additionally, 6.1%
reported an improvement in symptoms and 38.1% reported no change. Cumulatively, 54.8% of
reporters reported a worsening of their general state of illness/wellness.
Literature:
Vaccination has been suggested as being involved in the aetiology of ME/CFS. Literature examining
the risk of ME/CFS exacerbation following vaccination with Comirnaty or other vaccines was not
found. In addition, despite the lack of a universally accepted definition of long-COVID, there is
significant interest in the potential of post-COVID-19 or long-COVID syndrome fitting under the
definition of ME/CFS. A systematic review and meta-analysis that examined the long-term effects of
COVID-19 has been included in the literature review for completeness.
HPV vaccination and risk of chronic fatigue syndrome/myalgic encephalomyelitis: A
nationwide register-based study from Norway. (2017). [4]
An observational study conducted in the mass vaccination of Norway following the administration of
the HPV Gardasil vaccine to 10–17-year-old girls throughout the six years following the vaccination
programme being made available. The study identified that the incidence rate ratio of ME/CFS in
boys and girls aged 10-17 was 1.15 (respectively) person-years among the 824,133 living in Norway
during the six-year period examined. Further to this, the study identified that the hazard ratio (HR)
for the development of ME/CFS in girls following vaccination was 0.86 during the follow-up period
and 0.94 for the duration of a two-year follow up. As a result, the researchers conclude that
vaccination with Gardasil did not increase the risk of ME/CFS. The study utilised the Norwegian
national immunisation database and population and patient registries.
Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is associated with pandemic
influenza infection, but not with an adjuvanted pandemic influenza vaccine. (2015). [5]
An observational study conducted in Norway following mass vaccinations in response to the 2009
influenza A (H1N1) pandemic, investigated the risk of ME/CFS after infection and vaccination in the
entire Norwegian population. The researchers utilised the Norwegian national immunisation
database to estimate that the HR of ME/CFS in those who received the H1N1 vaccine (Pandemrix,
with AS03 adjuvant) was 0.97. This was contrasted with the HR following infection with Influenza of
2.04. Given that the incidence rate of ME/CFS was 2.08 per 100,000 person-months, the researchers
concluded that symptomatic infection, rather than antigenic stimulation, may trigger ME/CFS.
Page 5 of 7
HNZ00203542 Appendix 2
More than 50 long‑term effects of COVID‑19: a systematic review and meta‑analysis. (2021).
[6]
A systematic review and metanalysis of 15 observational studies examining persistent symptoms
following COVID-19 infection (for a period of 14 to 110 days) found that fatigue is the most
common symptoms of long-COVID-19, present in 58% of people who recovered from acute SARS-
Cov-2 infection. A total of 47,910 patients were included in the analysis and were 17-87 years in age.
Similarities between the symptoms observed in post-COVID-19 patients and ME/CFS patients have
been acknowledged widely. This study notes that key clinical characteristics of ME/CFS, such as post-
exertional malaise, inadequate sleep, and incapacitating fatigue are similar to those in patients with
persistent fatigue following COVID-19 disease.
EXPERT ADVICE
It is recommended that the COVID-19 Vaccine Independent Safety Monitoring Board (CV-ISMB) is
updated on the currently available information about ME/CFS following vaccination with Comirnaty.
CONCLUSIONS AND PROPOSED ACTIONS
Overall, the available data does not highlight safety concerns for the development or exacerbation
of ME/CFS following administration with Comirnaty. There is an absence of literature around the risk
of ME/CFS exacerbation following Comirnaty therefore it is recommended that this continues to be
closely monitored through routine pharmacovigilance activities.
Page 6 of 7
HNZ00203542 Appendix 2
RECOMMENDATIONS
It is recommended that:
1.
This issue is monitored through routine pharmacovigilance activities
Yes
3.
The CV-ISMB is updated on the currently available information
Yes
References
[1] NICE Guidelines, “Myalgic encephalomyelitis (or encephalopathy)/chronic fatigue syndrome:
diagnosis and management,” 2021.
[2] Vincent A., et al., “Prevalence, incidence, and classification of chronic fatigue syndrome in
Olmsted County, Minnesota, as estimated using the Rochester Epidemiology Project,”
Mayo
Clinic Proc., 2021. https://doi.org/10.1016/j.mayocp.2012.08.015.
[3] ANZMES, “Associated New Zealand ME Society (ANZMES) Preliminary Survey Findings,” 2021.
https://anzmes.org.nz/anzmes-preliminary-survey-findings/.
[4] Feiring B., et al., “HPV vaccination and risk of chronic fatigue syndrome/myalgic
encephalomyelitis: A nationwide register-based study from Norway,”
Vaccine, 2017.
https://doi.org/10.1016/j.vaccine.2017.06.031.
[5] Magnus P., et al., “Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is associated
with pandemic influenza infection, but not with an adjuvanted pandemic influenza vaccine,”
Vaccine, 2015. https://doi.org/10.1016/j.vaccine.2015.10.018.
[6] Lopez-Leon S., et al., “More than 50 long-term effects of COVID-19: a systematic review and
meta-analysis,”
Nature, 2021. https://doi.org/10.1038/s41598-021-95565-8.
Page 7 of 7
HNZ00203542 Appendix 2
A Comparison of the risk of death between Comirnaty mRNA
vaccinated and unvaccinated individuals with pre-existing heart
conditions
Vadim Pletzer, NIP Data & Analytics
30/05/2022
Acknowledgments
We’d like to thank Thomas Lumley and Ralph Stewart for their continued support and their fruitful ideas
which we have implemented in this analysis.
Introduction
In New Zealand the primary vaccine used against COVID-19 is Comirnaty mRNA, which has been associated
with an increased risk of myocarditis (Mevorach et al. 2021). The purpose of this analysis is to identify
whether individuals with pre-existing heart conditions are at an increased risk of death post-vaccination. We
tackled this question using a survival analysis.
Methodology
In this study we followed two cohorts, unvaccinated and vaccinated. Our unvaccinated cohort consists of
all individuals who were discharged from hospital to a home setting in 2021, and for which they had a
hospitalization event that was either heart failure or heart attack (see Table 2). We followed these subjects
from their latest hospital discharge date.
In the event the subject had a hospitalization date post vaccination, we followed the subject from the
hospitalization date that was prior to their first vaccination date. Subjects in the unvaccinated cohort were
followed until the 15th of October if they remained unvaccinated, otherwise until, the date of death or date
of their first vaccination. Subjects who received their first vaccination switch to the vaccinated cohort at
their first dose date. For our vaccinated cohort we began following subjects from either the date of their
first or second dose. We stopped following subjects when they received their second dose or until the 15th of
October, whichever date came first.
We made no distinction between doses, that is, if a subject received a second dose we would include the
same person twice in the vaccinated cohort. As a result, we ended up with an unvaccinated cohort of 15,285
subjects and 1,774 unvaccinated deaths. In the vaccinated cohort we had 19,791 observations (this number
includes the same people more than once since we don’t distinguish between dose one and two) and 494
vaccinated deaths.
1
HNZ00203542 Appendix 2
The hazard function
To identify whether there is an increased risk of death due to vaccination, we compared the hazards for each
cohort. The hazard is the instantaneous rate of death at time
t given that the individual survived up to time
t.
1
dS(
t)
λ(
t) = −
(1)
S(
t)
dt
In Equation 1,
S(
t) is the probability to be alive at time
t and is represented as:
Z
t
S(
t) = exp −
λ(
t∗)
dt∗
0
Age-adjusted time dependent survival model
200
150
Cohort
Unvaccinated
100
Count
Vaccinated
50
0
0
100
200
Days from discharge
Figure 1: Distribution of days to death from date of discharge
In Figure 1, we plot the frequency of deaths following discharge for both cohorts. We observe very different
distributions of death between unvaccinated and vaccinated cohorts. For the unvaccinated, most deaths occur
closer to the date of hospital discharge.
In comparison, deaths among vaccinated initially increase, plateau and then decrease with the number of days
since discharge. The increase and decrease are likely due to the small number of subjects being vaccinated
shortly (
< 50 days) and long (
> 200 days) after discharge (see Figure 2).
2
HNZ00203542 Appendix 2
600
400
Count
200
0
0
100
200
300
Days from discharge to vaccination date
Figure 2: Distribution of days to vaccination from date of discharge
3
HNZ00203542 Appendix 2
In response to these differences in the distributions of days to death, we introduce transient effects to the
unvaccinated and vaccinated hazards. These effects can represent the impact of vaccination on mortality, as
well as, potentially, a selection bias. A selection bias may arise if subjects with different health conditions are
prioritized for vaccination.
Since mortality depends on age, we also introduce an effect for age. We model the unvaccinated and vaccinated
hazards as follows:
td
λunvaccinated =
λ0(1 +
a0 exp(−
)) exp(
βz)
(2)
τ0
tv
td
λvaccinated =
λ1(1 +
a1 exp(−
) +
a2 exp(− )) exp(
βz)
.
(3)
τ1
τ0
Equations 2 and 3 model the risk of death for unvaccinated and vaccinated subjects by taking into account
transient effects (1 +
a exp(−
t )) on the baseline hazard (
λ). They also take into account the effect of age (
β).
τ
The parameters to optimize are:
• 0
< λ0
< ∞, long-term hazard of unvaccinated subjects
• 0
< λ1
< ∞, long-term hazard of vaccinated subjects
• −1
< a0
< ∞, selection bias among unvaccinated subjects;
a0
> 0 increases the hazard
• 0
< τ0
< ∞, time it takes for the unvaccinated selection bias to decay
• −1
< a1
< ∞, effect of vaccination on mortality;
a1
> 0 increases the hazard
• 0
< τ1
< ∞, time it takes for the effect of vaccination on mortality to decay
• −∞
< β < ∞, effect of age on mortality; a positive
β means older subjects have a higher hazard
• −1
< a2
< ∞, selection bias among vaccinated subjects;
a2
> 0 increases the hazard.
Here,
td is the time in days since hospital discharge,
tv is the time since vaccination for vaccinated subjects
and
z =
age−
agemin is the standardized age (0 ≤
z ≤ 1).
agemax−
agemin
Results
Parameter
MLE
Lower
Upper
λ0
8.399E-06
5.316E-06
1.148E-05
λ1
7.585E-06
5.987E-06
9.183E-06
a0
3.763E+00
2.368E+00
5.158E+00
τ0
5.448E+01
3.821E+01
7.076E+01
a1
-1.144E+00 -1.198E+00 -1.090E+00
τ1
8.179E+00
4.121E+00
1.224E+01
β
5.745E+00
5.614E+00
5.877E+00
a2
2.264E+00
1.326E+00
3.202E+00
Table 1: Summarizing the maximum likelihood estimates of the parameters from the age-adjusted time
dependent survival model, along with lower and upper bounds corresponding to a 95% confidence interval.
After adjusting for age and transient effects, we found no statistically significant difference in the long-term
hazard between the vaccinated and unvaccinated cohorts. There is no evidence that vaccination increases the
risk of death among individuals with pre-existing cardiac problems.
The period following hospital discharge appears to be associated with a higher risk of mortality. In contrast, we
found vaccination reduces the risk of mortality. This could indicate a strong selection bias in favour of healthy
subjects being chosen for vaccination. In Figure 3, we plot the estimated hazards for the unvaccinated and
vaccinated cohorts. We observe the increased hazard post-discharge and the decreased hazard post-vaccination
(0, 20 and 50 days after discharge).
4
HNZ00203542 Appendix 2
4
xp(beta*z)
3
unvaccinated
vaccinated at 0 days
e to lam0 * e
vaccinated at 20 days
vaccinated at 50 days
2
Hazard relativ
1
0
50
100
150
200
Days from discharge
Figure 3: Hazards from days since hospital discharge
5
HNZ00203542 Appendix 2
In Table 1 we summarize the maximum likelihood estimates for the parameters in the age-adjusted time
dependent models (see Equations 2 and 3). We found the long-term hazards for the unvaccinated (ˆ
λ0 =
0.0000084) and vaccinated (ˆ
λ1 = 0.0000076) cohorts to be very close to each other. That is, among individuals
who have a previous history of hospitalization for heart disease and/or heart attack, the long-term hazard
(instantaneous risk of death) among those that have received at least one dose of Comirnaty is not statistically
different from the long-term hazard for the unvaccinated after accounting for selection bias and age.
The transient effect on the unvaccinated (ˆ
a0 = 3.76) hazard is positive. This means that for the unvaccinated,
immediately after discharge, the risk of death is 1 + 3.76 = 4.76 (see equations 2 and 3) times higher than the
long-term risk of death (ˆ
λ0 = 0.0000084). This transient effect is reduced by 1 or 63% after ˆ
τ
e
0 = 54
.48 days.
Similarly, among subjects that got vaccinated, their risk of death is also highest immediately after discharge
(ˆ
a2 = 2.26); the risk of death immediately after discharge is 3.26 times higher than the long-term risk of
death (ˆ
λ1 = 0.0000076). Again, this increased risk reduces by 63% after about ˆ
τ0 = 54 days. These increased
risks on the unvaccinated and vaccinated hazards immediately after discharge are not statistically different.
Interestingly, the effect of vaccination reduces the hazard (ˆ
a1 = -1.14). This could indicate a selection bias
whereby the healthiest get vaccinated. However, this effect is short lived (ˆ
τ1 = 8
.18 days).
Note that the hazard in our model can be negative immediately after vaccination, which is unphysical. This
can happen when the time of discharge from hospital is larger than ≈ 130 days (see Figure 3). With more
data we expect this issue to disappear.
Lastly, since we standardized age in years at start of follow-up using
age−
agemin , for every additional year
agemax−
agemin
in age, the risk of death increases by a factor exp (5
.75
/(106
.475 − 0
.0345)) = 1
.056 or 5.6%. The minimum
age (
agemin) in our data was 0.0345 years and the maximum (
agemax) was 106.475 years.
Appendix
For each model, we maximized the likelihood of the form:
N
N
Z
t
Y
Y
L =
λ(
ti)
diS(
ti) =
λ(
ti)
di exp(−
λ(
t∗)
dt∗)
i=1
i=1
0
and obtained maximum likelihood estimates for
λ,
a and
τ via the
optim() function in
R. To ensure that the
parameters fall in their range, in the likelihood function, we added a constraint whereby
λ(
t) ≥ 0. In order to
find converged maximum likelihood estimates, we used the “Nelder-Mead” method and set the maximum
number of iterations to 100,000. All other arguments to
optim() were kept as default.
In table 2 we list the ICD10 codes used to define “pre-existing heart condition.” Only patients with these
codes were selected for this study.
References
Mevorach, Dror, Emilia Anis, Noa Cedar, Michal Bromberg, Eric J. Haas, Eyal Nadir, Sharon Olsha-Castell,
et al. 2021. “Myocarditis After BNT162b2 mRNA Vaccine Against Covid-19 in Israel.”
New England
Journal of Medicine 385 (23): 2140–49. https://doi.org/10.1056/NEJMoa2109730.
6
HNZ00203542 Appendix 2
ICD10 Codes
I099
I110
I260
I319
I50
I500
I501
I509
J81
R092
I21
I249
Table 2: ICD10 codes used as inclusion criteria.
7
HNZ00203542 Appendix 2
8.
The Board held three extraordinary meetings in 2021. The first was in April 2021 to test
the process of bringing the Board together at short notice and discuss the emerging issue
of thrombosis with thrombocytopenia syndrome (TTS) with the Janssen and AstraZeneca
COVID-19 vaccines. The other extraordinary meetings were to discuss fatal reports where
there was a sense that the Pfizer vaccine could have contributed to the events leading to
the respective individuals’ deaths. Further details of these meetings are provided in
Appendix 2.
9.
An interim report detailing the Board’s work for 2021 (February-December) was produced
and is published on the Ministry of Health website, along with regular meeting minutes.
This information is available on the COVID-19 Who we're working with page here.
Work of the Board in 2022
10. Changes to the programme including the introduction of different COVID-19 vaccines,
increased eligibility for vaccination and booster doses, require careful monitoring and
messaging around safety. To date in 2022, topics considered by the Board include:
• Use of the AstraZeneca and Novavax COVID-19 vaccines
• Safety of booster doses in New Zealand
• Use of the paediatric Pfizer vaccine
• Reports of persisting disability after the Pfizer vaccine
• COVID-19 vaccination in pregnancy
11. The Board has held one extraordinary meeting to consider two fatal reports of concern.
Further details of this meeting are provided in Appendix 2.
12. The work of the Board has steadily reduced due to the decreased number of vaccine
doses administered and a corresponding decrease in the number of adverse events
reported (including a proportionate fall in the number of serious and fatal cases).
13. The Director, National Immunisation Programme has indicated that the Board should
continue its tenure throughout 2022, due to expected programme changes (new variant
vaccine and potential for infant COVID-19 vaccines).
14. The Board’s tenure is agreed until 23 December 2022, with a review in
September/October. The meeting cadence has been reduced to every six weeks, with the
next meeting scheduled for 3 August 2022.There is stil the provision for extraordinary
meetings to be held at short notice, should an urgent issue arise.
Future
Future governance arrangements for vaccine safety
15. Learnings from the COVID-19 vaccine rollout should be leveraged when considering
future governance arrangements for vaccine safety for the National Immunisation
Programme.
16. The remit of the CV-ISMB currently only covers the safety of the COVID-19 vaccines, and
the members of the Board were recruited on this basis. However, some of the expert
members may wish to transition to a future vaccine governance entity.
17. The World Health Organisation (WHO) recommends that countries establish a National
Immunisation Technical Advisory Group (NITAG) to guide immunisation policies and
programme decisions. The importance of a NITAG was acknowledged by the Strategic
Advisory Group of Experts (SAGE) on Immunisation in April 2017.
Memorandum
Page 2 of 8
HNZ00203542 Appendix 2
18. Broadly, the mandate of a NITAG could include the following:1
•
determination of optimal national immunisation policies
•
provide guidance on the development of strategies for vaccine preventable diseases
through immunisation
•
advise on the monitoring of immunisation programmes, to allow the impact to be
measured and quantified
•
advise the government on collection of disease and vaccine uptake information
•
guidance to organisations, institutions or government agencies around the creation of
policies, plans and strategies for research and development of new vaccines and
vaccine delivery technologies for the future.
19. Further work is needed to understand how the valuable work of the CV-ISMB around
vaccine safety can continue in the future. The function of the CV-ISMB would preferably
be designed into an existing/new group within the Ministry/Health New Zealand.
20. Ideally there wil be a smooth transition between the CV-ISMB ending and a new group
being stood up or an existing group taking over this function. There is a risk if that if this is
not prioritised, there is a delay effecting the support available for CARM, Medsafe and the
Programme.
Recommendation
It is recommended that you:
1.
note
that the COVID-19 Vaccine Safety Monitoring Board tenure has been
agreed until 23 December 2022.
2.
note
further work is needed around the future structure of vaccine safety
governance.
Signature ______________________________________
Date:
Dr Nick Chamberlain
National Director
National Immunisation Programme
1 Duclos P. National Immunization Technical Advisory Groups (NITAGs): guidance for their establishment and
strengthening. Vaccine. 2010 Apr 19;28 Suppl 1:A18-25. doi: 10.1016/j.vaccine.2010.02.027. PMID: 20412991.
Memorandum
Page 3 of 8
HNZ00203542 Appendix 2
Appendix 1 – Members of the COVID-19 Vaccine Independent Safety Monitoring Board (July
2022)
Name
Area of Expertise
Mr John Tait (Chair)
Obstetrics
Honorary Associate Professor Hilary Longhurst
Immunology; Pathology
(Deputy Chair)
Dr Nick Cutfield
Neurology
Associate Professor Matt Doogue
Clinical Pharmacology; Endocrinology
Dr Kyle Eggleton
General Practice
Professor Chris Frampton
Biostatistics
Dr Maryann Heather
General Practice; Pacific Health
Dr Tom Hills
Immunology
Dr Wendy Hunter
Paediatrics
Professor Thomas Lumley
Biostatistics
Ms Saskia Schuitemaker
Lay person – to represent consumer interests
Dr Owen Sinclair, Te Rarawa
Paediatrics, Māori Health
Professor Lisa Stamp
Rheumatology
Dr Anja Werno
Microbiology; Pathology
Dr Enver Yousuf
General Medicine
Professor Ralph Stewart
Cardiology
Dr Laura Young
Haematology
Memorandum
Page 4 of 8
HNZ00203542 Appendix 2
Appendix 2 – COVID-19 Vaccine Independent Safety Monitoring Board extraordinary meetings
Thrombosis with thrombocytopenia syndrome safety concern
An ad-hoc meeting to discuss TTS was held on 22 April 2021. The purpose of the meeting was to
discuss:
• if a similar risk has been identified in New Zealand
• whether the Pfizer/BioNTech vaccine is associated with this concern
• if it would be beneficial to provide information on this clotting/bleeding syndrome for the public,
and if so, what communication would be needed.
At the time, a haematologist was not appointed to the Board, so Dr Laura Young was engaged to
provide expert advice in this capacity. Dr Young was formally appointed as a Board member in
August 2021, following an increase in the number of thrombotic and bleeding events reported for
the Pfizer/BioNTech vaccine and the potential for New Zealand to start using the Janssen or
AstraZeneca COVID-19 vaccines.
The Board considered the available information for TTS and was reassured by the extensive
international experience with the Pfizer/BioNTech vaccine and the local experience to date in New
Zealand. No risk was identified with the Pfizer/BioNTech vaccine. The Board recommended a
Monitoring communication to reassure people that Medsafe is aware of the association between
TTS and the Janssen and AstraZeneca COVID-19 vaccines, and that the safety of the
Pfizer/BioNTech vaccine is being monitored closely for this issue but no such link has been
identified.
Vaccine-mediated myocarditis death
The Board held an ad-hoc meeting on 9 August 2021 to discuss a fatal report of concern in an
individual following COVID-19 vaccination.
On 2 August 2021, CARM received a report from a forensic pathologist for a woman who had
passed away approximately four days after their first dose of the Pfizer/BioNTech vaccine.
Myocarditis was a finding of the post-mortem examination that had not been recognised prior, with
follow-up investigations indicating that the myocarditis could have been temporally associated with
the individual’s vaccination event.
At the 9 August 2021 meeting, the Director of CARM provided an overview of the case followed by
a presentation from the forensic pathologist of their findings to date. The Board had also received
an expert opinion from Dr Ralph Stewart, a cardiologist recently appointed to the Board.
The Board considered the potential causes of the individual’s myocarditis, including the
Pfizer/BioNTech vaccine, and noted the following.
• The Pfizer/BioNTech vaccine and some other COVID-19 vaccines increase the risk of
myocarditis; Medsafe issued an Alert communication on 21 July 2021.
• COVID-19 infection increases the risk of myocarditis substantially more than COVID-19
vaccination.
• There are many possible causes of myocarditis, the most common being viral infection. Over
100 people are discharged from hospital with a principal diagnosis of myocarditis in New
Zealand every year.
• In this case, other factors have been identified that may have potentially caused the
myocarditis or led to a more severe myocarditis.
• The individual had no symptoms prior to the vaccine and the symptoms of myocarditis
developed in the days immediately following the first vaccine dose.
The Board concluded that based on the currently available information, the vaccination event was
the likely cause of the myocarditis. The Board considered that the circumstances of this case do
not impact or change the known information on myocarditis, and the benefits of vaccination with
the Pfizer/BioNTech vaccine for COVID-19 continue to greatly outweigh the risks of this rare side
effect.
Memorandum
Page 5 of 8
HNZ00203542 Appendix 2
The forensic pathologist sent histology slides to cardiac pathologists in the United Kingdom (UK)
and United States (US) for review to confirm the myocarditis type. Feedback received from the UK
cardiac pathologist agreed with the findings of the case. Review from the US was still pending at
the time the Board issued their statement; however, it was considered that this would not change
the viewpoint taken by the Board.
The Board recommended that the Ministry of Health advise clinicians to be aware of myocarditis
and pericarditis symptoms. The Ministry of Health issued a media release on 30 August 2021.
Potential vaccine-mediated myocarditis deaths
The Board met on 8 December 2021 to discuss three fatal reports of concern in individuals
following COVID-19 vaccination.
In the week commencing 29 November 2021, CARM received three fatal reports for individuals
who passed away in the period following vaccination, where vaccine-mediated myocarditis was
proposed as the cause of death.
Two of the reported cases are under investigation by the coroner and were reported to CARM by
the pathologists. The third case was reported to CARM by the district health board (DHB), following
a review by their Adverse Reactions Committee.
High level details of the cases are presented below.
• A young adult man who passed away 12 days after their first dose of the vaccine. The Board
understands he experienced symptoms that could be indicative of myocarditis in the days
preceding his death.
• A young person who passed away 11 days after their second dose of the vaccine.
• A man in his 60s who passed away approximately one month after the second dose of the
vaccine. The individual’s death was not considered to be linked to the vaccine. However,
following a review by the DHB, the death was reported due to the temporality of the vaccination
event.
At the 8 December meeting, the Director of CARM provided an overview of the cases to the Board.
The pathologist investigating the case of the young adult man and the forensic pathologist
investigating the case of the young person both attended the meeting and presented their findings
to date.
The death of the young person was discussed at length, however the Board considered that further
information from pending investigations was needed before a determination on the role of the
Pfizer/BioNTech vaccine could be made. A further ad-hoc meeting to discuss this case will be held
once this information becomes available.
On review of the case of the man in his 60s, the Board considered the myocarditis was unlikely
related to the vaccination event. The time from vaccination to the onset of symptoms and clinical
factors point to other causes and is not consistent with a causal link.
The Board considered the death of the young adult man and noted the symptoms of myocarditis
developed in the days following the first dose.
Based on the available information, the Board concluded that the vaccination event was the likely
cause of the myocarditis in the young adult man. The Board made the following recommendations
to the CVIP around communications.
• Updating communications to the public on symptoms of potential myocarditis and pericarditis
(e.g., is chest pain sufficient or is this better reflected as chest pain, tightness and/or chest
discomfort?).
• Ensuring that information on side effects is detailed at the time of vaccination; individuals need
to be provided with verbal and written information about what to expect after their COVID-19
vaccine. This should include discussion of common and rare side effects and when/where/how
an individual can seek medical advice.
• An update to the healthcare sector, in particular vaccinators, Whakarongorau, general
practitioners and emergency departments, about the risk of myocarditis with the
Pfizer/BioNTech vaccine and myocarditis signs/symptoms.
Memorandum
Page 6 of 8
HNZ00203542 Appendix 2
Myocarditis is a treatable condition, if identified, and outcomes are better the earlier that treatment
is started. The Board considered that the circumstances of these cases did not impact or change
the known information on myocarditis, and the benefits of vaccination with the Pfizer/BioNTech
vaccine for COVID-19 continue to greatly outweigh the risks of this rare side effect.
The Board also noted that Medsafe was actively engaging with other international regulators to
understand whether they have received similar reports.
On 20 December 2021, the Board issued a statement outlining the findings of the 8 December
meeting.
Two fatal reports of concern
The Board met on 2 March 2022 to consider two fatal reports of concern:
A young person who passed away 11 days after their second dose of the
Pfizer/BioNTech vaccine.
An elderly person who passed away eight days after their booster dose of the
Pfizer/BioNTech vaccine. The General Practitioner (GP) for this case had indicated
that the vaccine response (fever and nausea) could have been contributory.
Both cases were previously considered by the Board, with further information sought.
The Board discussed the case of the young person at length in December 2021, however further
tests were still pending, and the Board considered this information necessary before deciding on
the role of the vaccine.
The case was discussed in detail, with expert advice provided by the forensic pathologist. The
case is also with the Coroner, who is investigating. Information from the post-mortem examination
identified myocarditis as the most likely cause of death.
The Board considered the potential causes of myocarditis in this young person, including the
Pfizer/BioNTech vaccine. The Board noted that:
• There were no reported symptoms prior to the vaccine or in the days preceding their
second vaccine dose.
• This young person suffered a sudden cardiac death with no other contributing factors
identified.
• The forensic pathologist described this case as having very few pathological findings but
with a history that could be linked to the vaccination event.
• Children do occasionally die from sudden cardiac death; the annual incidence of sudden
cardiac death in children and young adults in Aotearoa New Zealand and Australia was
found to be 1.3 cases per 100,000 persons. (Bagnall et al. 2016)
• On average 95 people (SAFE study) are discharged from hospital with a principal diagnosis
of myocarditis in New Zealand every year. In most of these cases the cause of myocarditis
is not known but is thought to be a virus. In this individual, testing could not determine
whether the vaccine was or was not the cause of the myocarditis.
Whilst it is acknowledged that some members of the Board felt that the vaccine was the probable
cause of the myocarditis in this case, the majority settled on the vaccine being the possible cause
of the myocarditis.
The Board considers that the circumstances of this case does not impact or change the known
information on myocarditis, and the benefits of vaccination with the Pfizer vaccine for COVID-19
continue to greatly outweigh the risk of such rare side effects.
The Board noted COVID-19 infection can itself be a cause of myocarditis as well as other serious
illnesses and it remains safer to be vaccinated than to be infected with the virus.
Memorandum
Page 7 of 8
HNZ00203542 Appendix 2
The death of the elderly person had been discussed by the Board on 9 February 2022; however, a
decision on the role of the vaccine was not reached and further follow up information from the
GP/Aged Residential Care facility was sought by the Centre for Adverse Reactions Monitoring
(CARM).
The Director of CARM presented details of the case, with the initial report and follow up
information obtained from the individual’s GP. The Coroner was consulted regarding this case,
however, didn’t feel that the death needed to be investigated by them.
The Board considers the role of the Pfizer/BioNTech vaccine in the death of this individual
unclassifiable. The Board felt that there were other factors that could have contributed and/or
caused the events leading to the death and unfortunately these had not been excluded.
The Board did not feel that the circumstances of this case changed the known safety profile of
the Pfizer/BioNTech vaccine. However, reiterated that it was important for the benefit/risk for
vaccination in the frail elderly to be considered on a case-by-case basis and this was reflected
in the Pfizer/BioNTech vaccine data sheet.
Memorandum
Page 8 of 8
HNZ00203542 Appendix 2
9. The expert advice from the Board was invaluable to CARM, Medsafe, and the National
Immunisation Programme as part of the robust COVID-19 vaccine safety monitoring processes
in place.
10. From February 2021 to November 2022, 63,999 adverse events following immunisation were
reported to CARM. 3,676 of these were classified as ‘serious’1. 784 serious cases were
presented to the Board during 28 meetings held throughout 2021 and 2022 (including five
extraordinary meetings).
Final report
11. The Board is required, per their Terms of Reference (Appendix 1), to provide a final report of
their work. An interim report was published July 2022, covering the Board’s findings from
February 2021 to December 2021.
12. The focus of the Board’s interim report was the Pfizer/BioNTech COVID-19 vaccine. In 2022,
although the Pfizer/BioNTech COVID-19 vaccine continued to be the recommended vaccine,
the programme evolved with increased use of different vaccines (AstraZeneca and Novavax),
wider eligibility for booster doses and the introduction of the paediatric vaccine for 5–11-year-
olds.
13. The Board has produced their final report
“Final report 2022: COVID-19 Vaccine Independent
Safety Monitoring Board (CV-ISMB)” covering the period February 2021 to November 2022
(Appendix 2). The content of the final report has been reviewed and agreed by the Secretariat,
Medsafe, all the members of the Board, and the PLG. The family point of contact for the
Programme has also reviewed and approved this report.
14. An overview of the safety information for the AstraZeneca, Novavax and paediatric
Pfizer/BioNTech COVID-19 vaccines primary course and booster doses is provided in the final
report along with an updated overview of the safety signals2 considered for the Pfizer/BioNTech
COVID-19 vaccine.
15. There have been 24 safety signals considered for the Pfizer/BioNTech vaccine which has led
to 40 recommendations made by the Board to either Medsafe or the Programme.
16. Recommendations made by the Board included: suggested communications to the health
sector and public; for companies to update their data sheets; and for Medsafe to continue to
monitor an issue through routine pharmacovigilance activities.
17. Only one safety signal was identified in New Zealand, with myocarditis and pericarditis
identified as very rare adverse reactions to the Pfizer/BioNTech COVID-19 vaccines. A safety
signal for myocarditis and pericarditis was also investigated for the Novavax vaccine and an
alert communication was issued after the company indicated these may be rare adverse
reactions of the vaccine.
18. The Board also reviewed all reported fatal cases for the Pfizer/BioNTech COVID-19 vaccine. In
cases where it was felt there could be a link between the vaccine and the event(s) leading to
the fatal outcome, the Board provided a view on the probability of the association.
19. The experience and learnings from the Board are also invaluable and the report recommends
that these should feed into the development of an expert immunisation advisory group to
provide advice and guidance to the Programme for all vaccines in Aotearoa.
1 An adverse event following immunisation is classified as serious if: it is a medically important event or reaction; requires
hospitalisation or prolongs an existing hospitalisation; causes persistent or significant disability or incapacity; is life
threatening; causes a congenital anomaly/birth defect; results in death.
2 A safety signal is information on a new known adverse event that may be caused by the vaccine and requires further
investigation. Safety signals can be detected from a wide range of sources such as spontaneous reports, clinical studies,
and literature.
Memorandum
Page 2 of 7
HNZ00203542 Appendix 2
20. The full report contains no additional findings of significance from the interim report published
in July 2022.
Publication
21. There has been a lot of interest in the work of the Board from both the healthcare sector and
public. At the beginning of 2022, the decision was made to publish the meeting minutes for the
Board and Interim Report on the Manatū Hauora website.
22. The Board’s meeting minutes for 2022 were published on the Manatū Hauora website (28
March 2023). For transparency and confidence in our safety monitoring systems, it is proposed
that the full final report is also published online.
23. A communications plan has been prepared to support publication (Appendix 3). This will allow
for notification of key internal and external stakeholders prior to publication and provide
reactive lines should there be any queries.
24. Following your approval, the attached report will also progress to the Te Whatu Ora Executive
Leadership Team (ELT) (Appendix 4) and in Te Whatu Ora’s Weekly Report for Ministers, for
noting the intent to publish.
Future of the Board
Governance of the Immunisation System
25. In November 2022, Cabinet agreed to the establishment of a new governance mechanism for
the Immunisation System [CAB-SWC-22-MIN-0227 refers3]. The new governance structure is
an important component for the delivery of a National Immunisation Strategy and will also meet
Aotearoa’s obligations to provide support for the Pacific Realm.
26. The Public Health Agency within Manatū Hauora has progressed this work, with a revised
structure proposed in June 2023 (Figure 1) [Health Report H2023024966 refers]. This seeks
agreement from the Minister of Health to establish the Immunisation Oversight Board and the
Immunisation Outcomes Collective. However, further work is required to develop the
Immunisation Technical Advisory Group.
27. The Immunisation Technical Advisory Group will provide independent, evidence-based
technical advice to support decision making across all aspects of immunisation and ensure that
the programme is able to respond to the latest science and technical information.
28. It is anticipated that the Immunisation Technical Advisory Group will meet the requirements per
the World Health Organization’s Global Vaccine Action Plan (2011-2020), which called for all
countries to establish or have access to a National Immunisation Technical Advisory Group
(NITAG) by 2020. A NITAG is a multidisciplinary body of national experts that provide
evidence-based recommendations to policymakers and immunisation programme managers4.
3 Cabinet Minute: Establishing Strategic Priorities for Immediate COVID-19 Vaccination and Governance for
the Immunisation System
4 Source: National Immunization Technical Advisory Groups (NITAGs) (who.int)
Memorandum
Page 3 of 7
HNZ00203542 Appendix 2
Figure 1. Revised governance structure (June 2023)5
29. Health agencies are working together to consider the Terms of Reference for an Immunisation
Technical Advisory Group (or NITAG). In the interim, the Public Health Agency will continue to
seek expert advice on issues as they emerge.
30. There is an opportunity to consider a new programme advisory group of experts to operate as
the Programme’s lead in the relationship with the NITAG. Further work on the requirements
and establishment of this advisory group will occur following the finalisation of the NITAG
Terms of Reference.
Options
31. The Board’s Terms of Reference state that its tenure is until at least 30 June 2023. The Board
now requires notification as to whether they are required beyond this date. In making this
decision, consideration needs to be given to: the status of the COVID-19 vaccination
programme; Board workload and responsibilities; business-as-usual functions; and the
proposed immunisation governance structure.
32. The following options for the future of the Board have been considered:
Option A: Discontinue the Board (recommended)
Throughout 2021 and 2022 the Board was essential for the success and safety of the COVID-
19 vaccination programme. However, with the COVID-19 vaccination programme now well-
established and the final report completed, the Board has not needed to meet in 2023.
Discontinuing the Board on 30 June 2023 will see the responsibilities of the Board return to
business-as-usual processes, led by Medsafe. Medsafe would be responsible for responding to
any safety signals and will refer any significant safety concerns to the Medicines Adverse
Reactions Committee (MARC). This approach has been agreed with Medsafe.
5 Health Report: H2023024966 Implementing the governance mechanism for the immunisation system
Memorandum
Page 4 of 7
HNZ00203542 Appendix 2
In addition, the new immunisation governance structure may present an opportunity to transfer
vaccine safety responsibilities previously held by the Board, including feedback and learnings
from the Board.
If this option is selected, the Board will conclude on 30 June 2023. Letters thanking the Board
Chair and Board members are proposed to be sent by the Minister of Health. Draft letters are
enclosed in Appendix 5.
Option B: Continue the Board (not recommended)
As an ongoing entity, with the focus on COVID-19, the current structure of the Board is not fit
for purpose to support the broader National Immunisation Programme.
Selecting this option would mean the Board is extended with no change to the terms of
reference, and members are to be available as needed. If any meetings are held this would
incur a cost of $865 per day and $108 per hour for any part day (before tax) for Board
members. With the COVID-19 vaccination programme well-established, it is not anticipated that
further meetings will be required.
The Board would be disestablished at the point when the new immunisation governance
structure is stood up. There is potential that this may result in duplication during the
establishment phase as Board members are likely to be engaged in discussions on the new
structure, and with Medsafe’s responsibility for monitoring vaccine safety.
If this option is selected, this will be communicated to Board members by email.
Risks
Publication of the final report
33. The content of the report could be misinterpreted. This is mitigated by most of the information
in the final report already being available in the public domain through published
communications, the Medsafe Safety Report and Official Information Act (OIA) requests. The
final report provides a consolidated document for information pertaining to the work of the
Board.
34. If the final report is not proactively published, this will likely be requested through the OIA
process. There is already a precedent with the publication of the Board’s interim report. The
availability of the interim report was helpful during 2022 in responding to OIAs and other
queries, especially regarding potential safety signals.
35. Publication of the report could be upsetting for the families of the people who have died from a
potential vaccine-mediated myocarditis, due to the level of information included. This has been
mitigated through only including detail that is already within the public domain. The report was
reviewed by the family point of contact within the Programme and proactive contact with the
families is not needed prior to release of the report. This approach was discussed and agreed
by the PLG.
Changes to the Board
36. If the Board is stood down and the new governance structure has not yet been established,
there is a risk that there will be no group of experts to review any potential adverse events. As
there has been no requirement for meetings in 2023, and Medsafe holds business-as-usual
responsibility for this, the risk of this is low. Relevant experts may need to be reconvened by
Medsafe if required.
Memorandum
Page 5 of 7
HNZ00203542 Appendix 2
Recommendations
It is recommended that you:
1.
Note
The report
“Final report 2022: COVID-19 Vaccine
Independent Safety Monitoring Board (CV-ISMB)” contains
no additional findings of significance from the interim report
published in July 2022.
2.
Approve
The COVID-19 Vaccine Independent Safety Monitoring Board
Yes
(CV-ISMB) Final Report (Appendix 2) for publication online.
3.
Approve
The attached memo (Appendix 4) for the Te Whatu Ora
Yes
Executive Leadership Team (ELT) meeting on 27 June 2023
noting the intention to publish the report.
4.
Note
An update will be included in the weekly report to Ministers
for noting ahead of publication of the final report.
5.
Note
The COVID-19 Vaccine Independent Safety Monitoring Board
was established in February 2021, with terms of reference
through until at least 30 June 2023.
6.
Note
The Board has been available to meet by exception in 2023.
No meetings have been held to date.
7.
Note
The Public Health Agency within Manatū Hauora is leading
work to establish a new governance structure for the
immunisation system, including an Immunisation Technical
Advisory Group.
8.
Note
Further work will be done to consider a new programme
advisory group of experts to operate as the Programme’s
lead in the relationship with the NITAG.
9.
Approve
The next step for the Board, beyond 30 June 2023:
Option A: Discontinue the Board on 30 June 2023
Yes
(recommended)
OR
Option B: Continue the Board beyond 30 June 2023
No
(not recommended)
10.
If Option A: Discontinue the Board on 30 June 2023 selected:
Approve
Letters to the Board Chair and Board members thanking them
Yes
for their service, to be sent from the Minister of Health.
Dr Nick Chamberlain
National Director
National Public Health Service
20 / 06 / 2023
Memorandum
Page 6 of 7
HNZ00203542 Appendix 2
Attachments
Appendix 1: Terms of Reference: COVID-19 Vaccine Independent Safety Monitoring Board (CV-
ISMB)
Appendix 2: Final report 2022: COVID-19 Vaccine Independent Safety Monitoring Board (CV-
ISMB)
Appendix 3: Communications Plan
Appendix 4: Memorandum to the Executive Leadership Team - Publication of the COVID-19
Vaccine Independent Safety Monitoring Board (CV-ISMB) final report
Appendix 5: Draft letters to the Board Chair and Board members
Memorandum
Page 7 of 7
HNZ00203542 Appendix 2
Contents
Summary ........................................................................................................................... 1
Introduction ....................................................................................................................... 2
Results .............................................................................................................................. 2
Conclusions ....................................................................................................................... 5
References ........................................................................................................................ 6
Summary
Overall, the safety profile of JYNNEOS is similar between intradermal and subcutaneous
administration. However, those that receive the JYNNEOS vaccine via intradermal route
may experience syncope as an early onset reaction and local injection site reactions more
frequently than if they were given the vaccine subcutaneously. Clinically, if the local reaction
has not subsided by the time of the second dose, it is recommended to administer the
second dose in the contralateral forearm. For serious adverse reactions, there is currently no
evidence indicating a difference between administration routes. Regarding the vaccine in
general, without considering route of administration, no new or unexpected safety signals
have been identified in post-market surveillance for the JYNNEOS vaccine.
1
HNZ00203542 Appendix 2
Introduction
Due to global supply constraints for the mpox vaccine JYNNEOS, alternate routes of
administration have been implemented to reduce the amount of vaccine consumed per dose.
The two routes for JYNNEOS are intradermal (ID) and subcutaneous (SC). For ID, a 0.1 mL
dose is injected into the space between the epidermis and dermis layers of the skin.
Intradermal vaccination will usually result in a small, bubble on the skin. For SC, a short
needle is used to inject a 0.5 mL dose of the JYNNEOS vaccine into the subcutaneous layer
and is the traditional route of administration.
The Vaccine Safety team in Prevention of the National Public Health Service has performed
a rapid literature and overseas evidence review to determine if ID administration has a
different safety profile from SC. Given the turnaround of less than a week, the review was
not exhaustive. New Zealand based data were also collated on selected adverse events
following immunisation (AEFI) that were identified from the literature review. Records of
JYNNEOS vaccination in the Aotearoa Immunisation Registry (AIR) were cross-referenced
with the National Minimum Dataset (NMDS) to determine the frequency of the specified
AEFIs within a risk window of 42 days after vaccination. However, an extensive formal
analysis was not conducted. As such, there may be additional evidence not included in this
report that could impact the final conclusions.
Of note, the JYNNEOS vaccine was used in 2023 in Aotearoa New Zealand and the Vaccine
Safety team conducted a Post Vaccine Symptom Check (PVSC) survey during that time.
The final report did not have an analysis on route of administration. Access to the PVSC
data has been archived due to migration of IT systems so we are currently unable to do an
analysis retrospectively on PVSC mpox data. Additionally, only 396 survey responses were
received across the entirety of the campaign (combined responses for day 7, 14, and 42
surveys) out of 4,782 vaccination events. No safety concerns were identified in the 2023
PVSC campaign.
Results
Australia
Australia’s AusVaxSafety, their SMS-based active surveillance programme (equivalent to
PVSC), sent surveys in 2022 and 2023 to individuals who received a JYNNEOS vaccine.[1,
2] The surveys were sent 7 days after vaccination. Individuals who received JYNNEOS 0.1
mL intradermally reported slightly higher frequency of reactions (52%) than those who
received the JYNNEOS 0.5 mL subcutaneously (47%), as shown in Figure 1 (ID) and Figure
2 (SC). This slight difference was driven primarily by reports of local reactions to ID, 51%
compared to 44% for SC. Other commonly reported reactions were similar.
2
HNZ00203542 Appendix 2
were not stratified by administration route but there were no concerns with the overall safety
profile and the CDC did not identify any new or unexpected safety signals.
Table 1: The 10 most frequently reported adverse reactions following JYNNEOS vaccine receipt, by
route of administration via Vaccine Adverse Event Reporting System, United States, May 22–October
21, 2022 (reproduced directly from the publication[3])
Intradermal (n=325)
Subcutaneous (n = 212)
Incidence
Incidence
Adverse reaction
N
(95% CI)
Adverse reaction
N
(95% CI)
Injection site erythema
75
150 (118–188) Injection site erythema 36
107 (75–148)
Dizziness
66
132 (102–168) Injection site swelling
36
107 (75–148)
Urticaria
60
120 (91–154) Injection site pain
34
101 (70–141)
Injection site swelling
51
102 (76–134) Pain
29
86 (57–123)
Syncope
43
86 (62–116)
Erythema
28
83 (55–120)
Erythema
42
84 (60–113)
Dizziness
27
80 (53–116)
Loss of consciousness
41
82 (59–111)
Headache
26
77 (50–113)
Injection site pruritus
40
80 (57–109)
Fatigue
25
74 (48–109)
Hyperhidrosis
38
76 (54–104)
Injection site pruritus
23
68 (43–102)
Pruritus
33
66 (45–92)
Pyrexia
23
68 (43–102)
Incidence = reports per million doses
Rapid literature review
Pubmed was search with the following query:
jynneos and (intradermal and subcutaneous) and (safety or adverse or reactions)
A total of 21 articles were returned. Of these, two were relevant to vaccine safety, mode of
administration, used JYNNEOS in the study, and published in the last 10 years.[4, 5] A third
publication had overlapping data from the United States information presented above and
thus was excluded here.[6]
Out of 9,500 vaccinations at a mass vaccination clinic in Sydney, Australia, 11 (0.1%) AEFIs
were recorded, all among ID vaccinations. Eight out of the 11 AEFIs were syncope, two were
injection site reactions, and one mild throat itching.[4]
In a review of the Bavarian Nordic’s global safety database, ‘general disorders and
administration site conditions’ (i.e., local injection site reactions) occurred more frequently in
ID (690/2732, 25.3%) compared to SC (592/2732, 21.7%).[5] Statistical significance was not
calculated by the authors. The authors noted a substantial increase of syncope in ID (70/89,
78.7%) compared to SC (15/89, 16.9%).
New Zealand based data from Vaccine Safety
Based on records pulled from AIR, there were 4,782 doses of JYNNEOS given in New
Zealand from 17 January 2023 through 20 August 2024 where the route of administration
was known. Of these, 2,045 (42.5%) were administered ID and the remaining 2,737 (57.2%)
by SC. Immunisation records were cross-referenced with the NMDS, investigating if any
cases of cellulitis, sepsis, myocarditis/pericarditis, allergic reaction (acute bronchospasms,
anaphylaxis, and urticaria), and syncope were coded in NMDS within 42 days following
JYNNEOS vaccination. There were no events coded in the NMDS for cellulitis, acute
bronchospasms, anaphylaxis, and urticaria. There was one event of sepsis, one event of
pericarditis, and one event of syncope. For all events of sepsis, pericarditis, and syncope,
the route of administration was ID. Given the small sample size and low frequency of events,
tests of associations or effect of administration route cannot be conducted.
4
HNZ00203542 Appendix 2
Conclusions
Overall, the safety profile of JYNNEOS is similar between intradermal and subcutaneous
administration. However, those that receive the JYNNEOS vaccine via intradermal route
may experience syncope as an early onset reaction and local injection site reactions more
frequently than if given subcutaneously. Clinically, if the local reaction has not subsided by
the time of the second dose, it is recommended to administer the second dose in the
contralateral forearm. For serious adverse reactions, there is currently no evidence
indicating a difference between administration routes. Regarding the vaccine in general,
without considering route of administration, no new or unexpected safety signals have been
identified in post-market surveillance for the JYNNEOS vaccine.
The Vaccine Safety team in Prevention technically has the ability to deploy another PVSC
campaign for mpox. However, due to constraints in digital capabilities as of August 2024, it is
not feasible in an outbreak and emergency response situation. Deployment for new PVSC
campaigns need a lead time of at least three months in addition to development costs. As
such, new campaigns outside of influenza and COVID-19 are not currently being considered.
Active surveillance via the hospitalisation dataset can be operationalised quickly within
Vaccine Safety and requires no outside resourcing. Given JYNNEOS is a novel vaccine for
New Zealand, being introduced in the last two years to a population that has never received
the vaccine, it is important to actively monitor the safety of the vaccine for serious AEFIs.
5
HNZ00203542 Appendix 2
References
1.
AusVaxSafety.
Intradermal JYNNEOS mpox (monkeypox) vaccine. 2023 [cited 2024
August 20]; Available from: https://ausvaxsafety.org.au/mpox-monkeypox-
vaccine/intradermal-jynneos-mpox-monkeypox-vaccine.
2.
AusVaxSafety.
Subcutaneous JYNNEOS mpox (monkeypox) vaccine. 2023 [cited
2024 August 20]; Available from: https://ausvaxsafety.org.au/mpox-monkeypox-
vaccine/subcutaneous-jynneos-mpox-monkeypox-vaccine.
3.
Duffy, J.,
Safety monitoring of JYNNEOS vaccine during the 2022 mpox outbreak—
United States, May 22–October 21, 2022. MMWR. Morbidity and Mortality Weekly
Report, 2022.
71.
4.
Pollack, A., et al.,
Mpox vaccination routes: prevalence, correlates and adverse
effects of subcutaneous versus intradermal vaccination at a mass vaccination clinic
in Sydney, Australia. Internal Medicine Journal, 2024.
54(6): p. 1031-1034.
5.
Weidenthaler, H., et al.,
Real-world safety data for MVA-BN: Increased frequency of
syncope following intradermal administration for immunization against mpox disease. Vaccine, 2024.
6.
Duffy, J., et al.,
JYNNEOS Vaccine Safety Surveillance During the 2022 Mpox
Outbreak Using the Vaccine Adverse Event Reporting System and V-safe, United
States, 2022 to 2023. Sexually Transmitted Diseases, 2024.
51(8).
6
HNZ00203542 Appendix 2
Time to onset and characteristics of Brighton level 1–2 anaphylaxis
following COVID-19 vaccination in Aotearoa New Zealand: Implications
for post-vaccination wait times
Vaccine Safety
National Public Health Service
Health New Zealand | Te Whatu Ora
(Dated: 15th July 2025)
Contents
I. Introduction
2
II. Methods
3
III. Results
3
IV. Discussion
4
V. Appendix
7
References
8
HNZ00203542 Appendix 2
2
I.
INTRODUCTION
Post-vaccination wait times are intended to monitor for and promptly respond to serious immediate
adverse events following immunisation (AEFI). The main serious AEFI of concern are anaphylaxis,
syncope, and convulsions.
Anaphylaxis is a severe allergic reaction that can be fatal if not treated immediately, typically with
intramuscular adrenaline. Syncope (fainting) is often related to anxiety or pain and may occasion-
ally present with convulsive movements, leading to diagnostic confusion with seizures. Importantly,
convulsions may also occur during an anaphylactic episode, potentially due to cerebral hypoper-
fusion from vasovagal mechanisms or hypoxia. Convulsions and syncope can lead to subsequent
serious injuries, particularly if driving or operating machinery.
These AEFI vary in frequency. Syncope is the most common, with rates increasing from 15 to 99 per
100,000 vaccinations depending on the number of vaccines co-administered [4]. Convulsions are less
frequent. In Australia, syncopal seizures after HPV vaccination occurred at 2.6 per 100,000 doses
[2]. Anaphylaxis is rare, with Brighton Criteria levels 1–2 rates for non-COVID vaccines generally
low—approximately 1.3 per million doses for influenza, 0.6 for MMR, and 0.1 for tetanus toxoid
vaccines [5]. In contrast, COVID-19 mRNA vaccines show a higher rate of 11.1 per million doses
[6]. This may be due to unique components like polyethylene glycol (PEG), enhanced surveillance
during the pandemic, novel immune responses, and early vaccination of immunocompromised or
high-risk groups with different baseline risk profiles compared to routine pediatric populations.
In Aotearoa New Zealand, post-vaccination wait times differ by vaccine type. For most National
Immunisation Schedule (NIS) vaccines, a 20-minute observation period is recommended. For in-
fluenza vaccines, this may be reduced to 5 minutes in certain eligible individuals (≥ 13 years old,
no history of anaphylaxis, etc.). For COVID-19 vaccines, the standard is 15 minutes.
Inconsistent guidance complicates decisions around observation times, particularly when vaccines
are co-administered.
In 2023, the National Immunisation Programme (NIP) asked the Immunisation Advisory Centre
(IMAC) to review whether current wait times for NIS vaccines should be revised. In collaboration
with the Centre for Adverse Reactions Monitoring (CARM), IMAC reviewed reports of anaphylaxis,
convulsions, and vasovagal syncope occurring within 60 minutes of non-COVID-19 NIS vaccination
[3]. The review concluded that a 15-minute wait period could be considered for NIS vaccines.
Similarly, international bodies have reviewed early-onset AEFI. The Canadian National Advisory
Committee on Immunization (NACI) in 2020 recommended a 15-minute wait post-influenza vac-
cination, with some flexibility for low-risk individuals — specifically, a five to fifteen minute wait
for those who have previously received the influenza vaccine without a history of severe allergic
reactions [1]. Other countries, including the United States, Ireland, and Australia, also recom-
mend a 15-minute wait. The UK has no specific post-vaccine wait time guidance beyond general
monitoring for immediate reactions.
Reducing post-vaccination wait times could offer several benefits: increased vaccination clinic
throughput, lower risk of infection from overcrowded waiting areas, and potentially improved
vaccine uptake by removing wait time as a barrier to access. Furthermore, standardising post-
vaccination wait times would simplify guidance for vaccinators and support consistent compliance.
The key risk of a reduced wait time is that serious AEFI may occur outside clinical settings,
potentially leading to worse health outcomes and reduced confidence in vaccination.
Following international evidence and the IMAC review, we undertook additional analyses to sup-
port evidence-based recommendations for standardising post-vaccination wait times across both
COVID-19 and non-COVID-19 vaccines included in New Zealand’s National Immunisation Sched-
ule (NIS).
This analysis focused specifically on anaphylaxis—the most clinically serious and time-sensitive
AEFI.
HNZ00203542 Appendix 2
3
II.
METHODS
Reports of anaphylaxis following COVID-19 vaccination were obtained from the Centre for Adverse
Reactions Monitoring (CARM), Aotearoa New Zealand’s national passive surveillance system for
adverse events. CARM accepts voluntary reports from both consumers and healthcare profes-
sionals, which are reviewed and validated by clinical assessors. Each report includes a narrative
description and may be linked to vaccination event data through the COVID Immunisation Reg-
ister (CIR). For this analysis, reports were limited to those linked to Comirnaty (Pfizer-BioNTech
COVID-19 vaccine) and occurring between February 2021 and December 2022, when vaccination
was available to individuals aged 12 years and older.
To focus on acute events, only reports of anaphylaxis that were validated by a clinical reviewer
and assigned a Brighton level 1 or 2 rating were included. From the clinical narratives attached
to each report, time-to-onset information was manually extracted by reviewing the free-text notes.
When available, we recorded the earliest time to symptom onset. In cases where symptom onset
time was not specified, but the time to adrenaline administration or oxygenation was recorded,
this was used as a proxy for time to medical intervention. This approach resulted in two groups of
time measurements:
• Time to symptom onset (e.g., rash, throat tightness, difficulty breathing)
• Time to medical intervention (e.g., administration of adrenaline or oxygen)
Demographic characteristics (age, gender, and ethnicity) were summarised for the anaphylaxis
cohort. For comparison, the general vaccinated population during the same study period was
summarised using CIR data.
To describe the distribution of time to onset, we fitted a Gamma distribution to each time category
using maximum likelihood estimation (MLE). The Gamma distribution was parameterised by a
shape parameter α and rate parameter β, with time-to-event τ defined as τ ∼ Γ(α, β). The mean
and variance of the distribution are given by µ = α and σ2 = α , respectively. The log-likelihood
β
β2
function was maximised using the optim function in R (version 4.4.0).
From each fitted distribution, we estimated the average time to onset and estimated cumulative
percentages of anaphylaxis events occurring by given time thresholds (e.g., 5, 10, 15 minutes). This
analysis was performed separately for time to symptom onset and time to medical intervention.
III.
RESULTS
A total of 86 validated reports of Brighton level 1 or 2 anaphylaxis following Comirnaty (Pfizer-
BioNTech) vaccination were included, out of 8,172,770 doses administered between February 2021
and December 2022 (Table 1). This corresponds to a rate of approximately 1.05 anaphylaxis cases
per 100,000 doses administered.
Most cases occurred in females (77 of 86; 89.5%), with a median age of 44 years. In comparison,
the vaccinated population was 51% female, with a similar median age of 45 years. Among the
anaphylaxis cases, 12 individuals (14.0%) had a known history of anaphylaxis, and 23 (26.7%) had
a broader history of hypersensitivity. Hypersensitivity conditions included asthma, severe allergy,
mast cell disorders, or related immune conditions.
Time-to-onset distributions were analysed for two outcomes: symptom onset and medical inter-
vention. The estimated mean time to symptom onset was 15.2 minutes (95% CI: 10.1–20.4), while
the mean time to medical intervention was 42.2 minutes (95% CI: 30.4–54.1).
Accumulated times are summarized in Table 2. We estimate that 25% of symptom onset occurred
within 4.5 minutes, 50% by 10.6 minutes, and 75% by 21.1 minutes. For medical intervention, 25%
HNZ00203542 Appendix 2
4
of cases were treated within 20.9 minutes, 50% by 35.8 minutes, and 75% by 56.7 minutes.
The Gamma distributions provided a reasonable fit to the observed data (Figure 1), capturing the
timing patterns for both symptom onset and treatment events.
Table 1: Demographic characteristics of consumers experiencing anaphylaxis compared to the
vaccinated cohort
Cohort
Gender, n (%)
Median age (years)
No. of reportsa
Female
Male
Anaphylaxis
77 (89.5)
9 (10.5)
44
86
Vaccinatedb
4,185,495 (51.2)
3,987,275 (48.8)
45
-
Ethnic breakdown, n (%)
Ethnicity
Anaphylaxis cohort
Vaccinated cohort
Asian
2 (2.3)
1,299,743 (15.9)
European
60 (69.8)
5,084,820 (62.2)
MELAA
4 (4.7)
141,986 (1.7)
Maori
16 (18.6)
1,034,391 (12.7)
Other
1 (1.2)
26,720 (0.3)
Pacific Peoples
3 (3.5)
548,952 (6.7)
Unspecified
- (-)
36,158 (0.4)
History of allergy, n (%)
Known history of anaphylaxis
12 (14.0)
–
Known history of hypersensitivityc
23 (26.7)
–
a Number of validated BC 1-2 reports following vaccination submitted to CARM.
b Population vaccinated with Comirnaty between February 2021 and December 2022.
c
History of asthma, anaphylaxis, severe allergy, mast cell disorder, or related immune condition.
Table 2: Estimated and observed accumulated times by percentile for symptom onset and medical
intervention
Percentiled
Symptom onset
Medical interventione
Estimated (min)
Observed (min)
Estimated (min)
Observed (min)
25%
4.5
5
20.9
22
50%
10.6
12
35.8
32
75%
21.1
20
56.7
60
90%
34.9
30
81.1
80
95%
45.2
48
98.4
118
99%
69.3
61
136.7
139
d Percentile indicates the percentage of total accumulated reports.
e Medical intervention refers to the earliest reported time of treatment, such as adrenaline administration, oxygenation, etc.
IV.
DISCUSSION
Most reports were from female consumers, with a median age similar to the general vaccinated
population. Anaphylaxis following exposure to LNP-mRNA vaccines, such as Comirnaty, appears
to affect females more than males. This may be linked to polyethylene glycol (PEG), a lipid
conjugate in these vaccines suspected to trigger anaphylaxis. Female predominance is possibly
related to greater exposure to PEG-containing products like cosmetics [7]. In addition, about one
third of anaphylaxis cases had a documented history of hypersensitivity, including asthma, severe
allergies, mast cell disorders, or related immune conditions.
Anaphylaxis rates classified as Brighton Criteria (BC) levels 1–2 following non-COVID vaccines are
generally low—for example, approximately 1.3 per million doses for influenza, 0.6 per million for
HNZ00203542 Appendix 2
5
Figure 1: Observed time to occurrence (blue circles) and estimated cumulative gamma density
function (orange line) for symptom onset (a) and medical intervention (b).
(a) Symptom Onset
100
ts
80
60
ercent of Repor
e P
40
ulativ
20
Cum
Observed
0
Estimated
0
5
10
15
20
30
40
50
60
Time to Symptom Onset (minutes)
(b) Medical Intervention
100
ts
80
60
ercent of Repor
e P
40
ulativ
20
Cum
Observed
0
Estimated
0
5
10
15
20
30
40
50
60
Time to Medical Intervention (minutes)
HNZ00203542 Appendix 2
6
MMR, and 0.1 per million for tetanus toxoid [5]. In contrast, our study found a BC 1–2 anaphylaxis
rate of 10.5 per million doses for Comirnaty, consistent with published COVID-19 mRNA vaccine
rates around 11.1 per million [6]. The higher rate may reflect differences in vaccine composition,
vaccinated populations (adults versus children), and heightened reporting during the pandemic.
Given this rate, approximately 95,000 individuals would need to be observed to detect a single case
of anaphylaxis in the observation period, highlighting the rarity of the event and the importance
of balancing safety with operational efficiency.
The mean time to symptom onset was 15.2 minutes (95% CI: 10.1–20.4), while medical intervention
occurred later, at a mean of 42.2 minutes (95% CI: 30.4–54.1). This delay reflects that symptoms
can begin subtly and progressively worsen before treatment is required.
In Aotearoa, CARM and the Immunisation Advisory Centre (IMAC) audited anaphylaxis, bron-
chospasm, and convulsion reports occurring within 60 minutes of vaccination with New Zealand’s
national schedule vaccines [3]. They found 60% of events occurred within 10 minutes, 69% within 15
minutes, and 92% within 30 minutes. Our analysis of Comirnaty anaphylaxis reports shows 50% of
events occurred within 11 minutes, 75% within 21 minutes, and 90% within 35 minutes. Although
differences in case definitions and event types limit direct comparisons, the overall time-to-onset
distribution for Comirnaty is broadly similar to that for scheduled vaccines.
In relation to post-vaccination wait times, Table 4 shows that extending the observation period
from 15 to 20 minutes increases the estimated capture of symptom onset events from 62.6% to
73.1%, capturing approximately 10.5% more cases.
Apart from the Brighton Criteria assignment by clinical reviewers, CARM currently does not
employ a fully standardised format for reporting anaphylaxis. As a result, medical narratives
vary considerably—some provide detailed timing of symptom onset, others record only the time of
adrenaline administration or intervention, while some reports lack timing information altogether.
Similarly, documentation of patients’ allergy or hypersensitivity history is variable, which may
contribute to an underestimation of individuals with prior hypersensitivity in this analysis.
In conclusion, determining optimal post-vaccination wait times requires balancing the rarity yet
seriousness of early-onset adverse events with the practical need for efficient vaccine delivery. While
longer wait times increase the chances of promptly identifying and managing serious reactions
like anaphylaxis—potentially reducing fatalities and strengthening public confidence—shorter wait
times can improve vaccination throughput, reduce time barriers, and lower infection risks in waiting
areas, all critical during pandemics.
HNZ00203542 Appendix 2
7
V.
APPENDIX
Table 3: Estimated maximum likelihood estimates (MLEs) and mean time to occurrence (minutes)
with 95% confidence intervals, for symptom onset and medical intervention
Parameter
Symptom onset
Medical intervention
ˆ
α
1.021 (0.587, 1.455)
2.122 (0.980, 3.264)
ˆ
β
0.067 (0.031, 0.103)
0.050 (0.020, 0.081)
ˆ
α (Mean)
15.2 (10.1, 20.4)
42.2 (30.4, 54.1)
ˆ
β
Table 4: Observed and estimated cumulative percentages of reports by time to symptom onset
Time to Onset (min)
Observed cumulative %
Estimated cumulative %
0
0.0
0.0
2
24.2
11.9
5
33.3
27.6
10
48.5
47.9
15
63.6
62.6
20
75.8
73.1
30
90.9
86.2
40
90.9
92.9
50
97.0
96.4
60
97.0
98.1
70
100.0
99.0
HNZ00203542 Appendix 2
8
[1] Canada, P. H. A. o. (2020). Government of canada.
[2] Crawford, N. W., Clothier, H. J., Elia, S., Lazzaro, T., Royle, J., and Buttery, J. P. (2011). Syncope
and seizures following human papillomavirus vaccination: a retrospective case series. Med J Aust,
194(1):16–18.
[3] for Adverse Reactions Monitoring (CARM), C. (2023, unpublished). Time to onset analysis for anaphy-
laxis, convulsions and syncopal aefis occurring within the first hour, for routine vaccines administered
in new zealand (data from reports submitted to carm between 1965 and 2023).
[4] Groom, H. C., Brooks, N. B., Weintraub, E. S., Slaughter, M. T., Mittendorf, K. F., and Naleway,
A. L. (2023). Incidence of adolescent syncope and related injuries following vaccination and routine
venipuncture. J Adolesc Health.
[5] McNeil, M. M., Weintraub, E. S., Duffy, J., Sukumaran, L., Jacobsen, S. J., Klein, N. P., Hambidge,
S. J., Lee, G. M., Jackson, L. A., Irving, S. A., King, J. P., Kharbanda, E. O., Bednarczyk, R. A.,
and DeStefano, F. (2015). Risk of anaphylaxis after vaccination in children and adults. J Allergy Clin
Immunol, 137(3):868–878.
[6] Shimabukuro, T. and Nair, N. (2021). Allergic reactions including anaphylaxis after receipt of the first
dose of pfizer-biontech covid-19 vaccine. JAMA, 325(8):780–781.
[7] Somiya, M., Mine, S., Yasukawa, K., and Ikeda, S. (2021). Sex differences in the incidence of anaphylaxis
to LNP-mRNA COVID-19 vaccines. Vaccine, 39(25):3313–3314.
