Skip to main content
hpa logo
Topics A-Z:
Search the site:
Home Topics Infectious Diseases Infections A-Z Post Licensure vaccine safety surveillance

Post Licensure vaccine safety surveillance


Vaccination is one of the most effective public health interventions but with the decline in vaccine-preventable diseases concern surrounding vaccine safety threatens to eclipse the public’s fear of the disease. Common adverse events are seen at the clinical trial stage but rare reactions are not seen as the numbers of subjects in the trials are too small. Even the largest pre-licensure trials are inadequate to assess rare but serious side effects and events may only occur in a sub-group of the population not significantly represented in pre-marketing studies

It is important to have systems in place to investigate and respond rapidly to vaccine safety concerns as prolonged uncertainty may result in loss of confidence in the vaccine and a precipitous decline in coverage. The UK has a "Yellow Card" scheme which identifies possible adverse events after vaccination. Although such reports are essential for identifying novel or unusual adverse events, they are inevitably incomplete and cannot be used to estimate true risks or to establish causal relationships for events that are not exclusively vaccine-associated.


Once a possible adverse event is identified there is a need to supplement this existing system using large population based studies. Since 1995 we have linked hospital admission data and immunisation data using NHS number or an algorithm of sex, date of birth and postcode. This linkage provides us with a complete record for the child with the date and ICD code of the admission to hospital and the date of immunisation. After linkage, validation of the diagnosis coding of all or a subset of the data is performed by reviewing hospital case notes. This record linkage system has been used to establish an active system for post-marketing surveillance of vaccine safety which covers all vaccinations given during the first five years of life with sufficient power to rapidly identify rare reactions. The study regions together have a population of approximately 15 million.

In order to have access to these data approvals need to be sort on a national and local level. We have specific high level approval through the National Information Governance Board for Health and Social Care (NIGB) to use personal identifiers such as NHS number and Sex, date of birth and postcode for monitoring vaccine safety. We also have MREC approval for the study and also local permissions from the research and development offices at hospital and PCT level.


This method has demonstrated consistent results over the past 14 years.

Febrile convulsions

In 1995 a risk of febrile convulsions was seen in the 6 to 11 day period following MMR vaccine RI 3.77 (95% confidence interval 1.95-7.3)(1) and subsequently confirmed using the same methods in 2007 RI 3.64 (95% confidence interval 2.44-5.44) (2).

Aseptic meningitis

In the same study in 1995 a risk of aseptic meningitis was seen in the period 15-35 days after MMR 38.1 (95% confidence interval 4.3- 336) (1) and was subsequently corroborated in 2007 with a risk of 25.9 (95% confidence interval 2.8 - 233) (2). This increased risk was found to be attributed to the Urabe mumps component of the vaccine.

ITP (Idiopathic thrombocytopenic purpura)

These methods have also been used to assess the risk of ITP (Idiopathic thrombocytopenic purpura) in the 15-35 days after the 1st dose of MMR. Firstly in 1995 a increased risk of 6.44 (95% confidence interval 1.94 – 21.4) was seen (1) and again in 2001 with an increased risk of 5.8 (95% confidence interval 2.3 – 14.6) (3) with a risk per dose of 1 in 32,000. The risk after the 2nd dose of MMR was also investigated but no increased risk was seen in the 0-42 days period after vaccination RI 1.04 (95% confidence interval 0.37-2.92) (4)

Bacterial/Viral infections

The hypothesis that vaccination can overload the immune system was first tested using these methods in 2002 and rather than an increased risk an apparent reduced risk was seen in the three months following MMR vaccine RI 0.76 (95% confidence interval  0.58-0.99) (5). This study was repeated in 2009 and the findings confirmed this reduction in risk RI 0.81 (95% confidence interval 0.70 – 0.95) (6).


Since 1995 we have been able to respond rapidly to vaccine safety concerns using the robust epidemiological methods outlined here. Even rare outcomes such as ITP with a risk per dose of 1in 32,000 has been ascertained using these methods.

It is important that the public have confidence in the routine vaccination programme as a successful vaccination programme does have the capacity to protect the unvaccinated though herd immunity and so on an individual level  the risk-benefit balance may be questioned if there is any perceived risk from receiving the vaccine.


1. Farrington P, Pugh S, Colville A, Flower A, Nash J, Morgan-Capner P, Rush M, Miller E. A new method for active surveillance of adverse events from diphtheria /tetanus/pertussis and measles/mumps/rubella vaccines.  Lancet 1995; 345: 567-9

2. Miller E, Andrews N, Stowe J, Grant A, Waight P, Taylor B. Risks of Convulsion and Aspectic Meningits following Measles-Mumps-Rubella Vaccination in the United Kingdom Am J Epidemiol 2007; 165: 704-709

3. Miller E, Waight P, Farrington CP, Andrews N, Stowe J, Taylor B. Idiopathic thrombocytopenic purpura and MMR. Arch Dis Child. 2001 Mar;84(3):227-9

4. Stowe J, Kafatos G, Andrews NJ, Miller E. Idiopathic Thrombocytopenic Purpura and the second dose of MMR.Arch Dis Child. 2007 Oct 25

5. Miller E, Andrews N, Waight P, Taylor B. Bacterial infections, immune overload and MMR vaccine. Arch Dis Child 2002;88:222-223

6. Julia Stowe, Nick Andrews, Brent Taylor, Elizabeth Miller. No evidence of an increase of bacterial and viral infections following Measles, Mumps and Rubella vaccine Vaccine. 2009 Feb 25;27(9):1422-5