Monday, November 1, 2010

Surveillance for vaccine adverse effects is inadequte/ Medical Journal of Australia editorial

From Sydney Morning Herald and Medical Journal of Australia:
The [following] editorial calls for the urgent establishment of a new national body to provide uniform monitoring and "active surveillance", looking for adverse reaction cases in the community.
"This is critical to ensure that the benefits of vaccination outweigh the risks," Dr Gold said.
Passive surveillance cannot be relied on as the sole means of surveillance

Michael S Gold, Paul Effler, Heath Kelly, Peter C Richmond and Jim P Buttery

492 MJA• Volume 193 Number 9 • 1 November 2010


On 22 April 2010, use of seasonal trivalent influenza
vaccine in children aged 5 years and under was suspended
across Australia, pending an investigation into an appar-
ent increase in reports of adverse events following immunisation
(AEFI).1 This unprecedented halt to a national immunisation
initiative followed Western Australia’s decision to place a mora-
torium on the use of this vaccine in young children after observing
a spike in emergency department presentations for high fever and
febrile convulsions after vaccination.2 A subsequent investigation
by the Therapeutic Goods Administration indicated that febrile
convulsions related to the vaccine were reported from all jurisdic-
tions except the Northern Territory.2 The apparent rate of febrile
convulsions following vaccination was 5–9 per 1000 doses admin-
istered, about 50 times higher than that reported following
measles–mumps–rubella vaccination
.2,3 A recent review, requested
by the Minister for Health in WA, has highlighted significant
deficiencies in AEFI surveillance
In Australia, the current mechanism for identifying AEFI nation-
ally is passive surveillance. Passive surveillance relies on health
providers and the public recognising and reporting suspected
AEFI to state or federal health authorities. The constraints that are
inherent to passive surveillance, including under-reporting and
biased reporting, are compounded by the diverse approaches to
surveillance that are employed throughout Australia, as illustrated
by a fourfold difference in AEFI reporting rates per 100000
population between jurisdictions.
5,6 Adding to concerns about
variable sensitivity across the state systems is the inevitable delay
in collection, aggregation and analysis of AEFI reports forwarded
to the national authority.
A number of the issues evident during the response to the
vaccine-associated reactions were recognised 5 years earlier during
the National Vaccine Safety Workshop.7 A clear set of recommen-
dations for improving adverse event surveillance was identified at
the time, but many of the recommendations have not been
adequately addressed.

Robust postmarketing surveillance is vital for influenza vaccines
because seasonal trivalent influenza vaccine does not require
clinical trial data to demonstrate safety before release
— it is
assumed that safety is not altered by the annual change in the
combination of vaccine strains.
While past experience suggests
that this is true, history also indicates that future vaccine scares are
inevitable and we should plan accordingly.8 Trivalent influenza
vaccine, in particular, highlights the need for postmarketing
surveillance to be linked with the capacity for rapid review and
response, because a large proportion of the vaccine is administered
over a short period before the onset of the influenza season each
The way forward is to establish a coordinated, uniform approach
to AEFI reporting, coding, collation and analysis. A standing
vaccine safety monitoring group which includes key stakeholders
— representing the regulators, state and national immunisation
programs and vaccine safety and epidemiology experts — needs to
be urgently established.

The inability of the existing surveillance systems to detect the
early signal of an increased incidence of febrile convulsions, within
24 hours of receiving 2010 seasonal trivalent influenza vaccine,
demonstrates that passive surveillance cannot be relied on as the
sole means of surveillance. Complementary active surveillance
systems which can methodically detect potential AEFI signals,
quickly establish rates and establish causality should be devel-
oped. The Australian Childhood Immunisation Register is
uniquely placed to contribute to vaccine safety surveillance
through data linkage with hospital morbidity and emergency
department datasets, as demonstrated by a recent study from
South Australia.3 Sentinel surveillance in four tertiary care Austral-
ian paediatric hospitals has been shown to be an effective mech-
anism of surveillance for specific AEFI.9 Implementing active AEFI
surveillance systems will require sustainable funding, but this will
be a small fraction of the cost expended on vaccines and vaccine
delivery and could be resourced by levying a surcharge per vaccine
dose sold, similar to methods adopted elsewhere to support
compensation for vaccine-associated injuries.
Central to any system of vaccine safety monitoring are issues of
governance; specifically, transparency in decision making.
countries currently provide full disclosure and web access to de-
identified AEFI reports and open access to the deliberations of
expert committees.11,12 This engenders public trust in immunisa-
tion programs, and similar strategies should be considered in
The vast majority of Australian parents, vaccine recipients and
health care providers trust public health authorities to assess and
monitor vaccine safety. This is critical to ensure that the benefits of
vaccination outweigh any potential risks. In the aftermath of the
2010 seasonal trivalent influenza vaccine experience, maintaining
the public’s trust requires that we get started on building the fully
functional, standard-of-care AEFI surveillance system that Aus-
tralia deserves. Vaccine safety should be an integral component of
the National Immunisation Strategy, which should include strat-
egies for comprehensive and complementary passive and active
systems of surveillance.

For disclosures, footnotes and author information go here.

No comments: