Looking for potential drug safety issues in Yellow Card data

Pills and prescription



Voluntary reports of reactions on Yellow Cards constitute a vital source of information on adverse effects of medicines. Reports from healthcare professionals, patients and the pharmaceutical industry are all entered onto the MHRA’s adverse drug reaction (ADR) database and used to identify new potential drug safety issues (‘signal detection’).

 

Life cycle of a Yellow Card


What is signal detection?
A signal is generally defined as reported information on a possible causal relationship between an adverse event and a drug, the relationship being unknown or incompletely documented previously. Usually more than a single report is required to generate a signal, depending upon the quality of the information. A signal is not a confirmed adverse reaction.

The aim of signal detection is to promptly detect any possible unwanted effect associated with a medicine or to detect a change in the pattern or frequency of ADRs already known to be associated with a drug.


Data mining
Every new ADR report received could potentially contribute to a new signal. In the past, each report was individually assessed to decide if it did indeed contribute to a new signal, but due to the large volume of ADR reports received an automated process is used to help identify a signal. Automated signal detection applies statistical methods that help to focus resources on drug-reaction combinations which require fuller investigation. The ultimate aim is to detect possible drug safety issues as soon as possible so that prompt action can be taken to protect public health.

At the MHRA, ADRs reported to the Yellow Card Scheme are rapidly entered onto a database so they are available for signal detection as early as possible. For more information on this please see what happens to a Yellow Card.

Each week, all drug-reaction combinations in the database are statistically analysed and each combination is assigned a ‘disproportionality score’. This disproportionality score is used to identify drug-reaction combinations that have been reported unusually frequently compared to the background of other drug-reaction combinations in the database. Drug-reaction combinations which meet defined criteria are then assessed by a group of scientists, physicians and pharmacists to determine if the signal is likely to be a real medicine-related concern.


Signal detection criteria at the MHRA
Our criteria for selecting signals for further assessment with established medicines are different from those for medicines under intensive monitoring (black triangle medicines).

For medicines assigned a black triangle, all serious reports are assessed regardless of the disproportionality score as the adverse reaction profiles for these medicines have not been fully established. Just a single case report might represent an important safety signal.

For established (or non-black triangle) medicines we select signals for further analysis which meet a set of criteria. These criteria include, for example, all ADRs which occur in children or any which involve a fatality.

The criteria are constantly refined and validated to ensure that important signals are selected for further analysis.


Signal assessment
Once a signal has been identified, it is assessed further by a group of scientists and physicians for the likelihood of a causal relationship between the drug and reaction, and to identify possible risk factors contributing to the reaction, for example, age, underlying disease, and genetic susceptibility.

We consider the risk of a newly-identified adverse reaction to a particular medicine in the context of its overall side effect profile, and compare this profile with those of other medicines used to treat the same conditions. We also consider the benefit of the medicine in terms of its efficacy, the condition it is used to treat, and the target patient population.

In addition to data from the Yellow Card Scheme, we evaluate data from a variety of other sources such including:

  • epidemiological studies
  • published and unpublished data, including regular safety updates supplied by pharmaceutical companies
  • information from other regulatory authorities and worldwide databases
  • the General Practice Research Database.


Signal prioritisation
Once a signal has been identified and assessed to be a true adverse effect, it is prioritised using two further statistical methods which help recommend the course of action to take. Impact analysis indicates the impact of the signal through the strength of evidence and the implications for public health. A second tool that takes into account factors that include public perception of the issue and Agency obligations is used to prioritise the signal and define the timeline for completing a full investigation and take necessary action.

Signals that warrant further action are discussed with experts across the Agency including epidemiologists and clinicians. Expert advice is also sought from the Commission on Human Medicines, Expert Advisory Groups and from Europe through the Committee for Medicinal Products for Human Use (external link).


Regulatory action
Once a signal has been confirmed to be a true ADR, regulatory action is taken to minimise risk to the patient. Such action may include:

  • updating patient information leaflets (PILs) and Summaries of Product Characteristics (SmPCs) to take account of newly identified risks. In addition, patient information leaflets (PILs) are provided with all medicines. The Summary of Product Characteristics and patient information leaflets for a drug can be found in:
  • restricting indications, or introducing new contra-indications, or changing the recommended dose or duration of treatment
  • changing the legal status of a medicine, for example, from over the counter to prescription only
  • publishing warnings in ‘Drug Safety Update’ which is produced monthly by the MHRA and CHM and is aimed at all healthcare professionals who can subscribe to be notified by email each month when it is published (email registration@mhradrugsafety.org.uk)
  • issuing 'Dear Healthcare professional' letters. Letters are sent to all doctors and pharmacists by post or electronic cascade to provide urgent warnings about drug hazards. This may be because of the seriousness of a newly identified ADR, because the drug is being withdrawn on safety grounds, or because a quality defect has occurred. How this information is disseminated will depend upon the nature of the problem. For urgent safety issues – health professionals may receive a letter from the chairman of the CHM which outlines the issue, the action being taken and provides advice on the implications for patient treatment. These letters are usually distributed via the MHRA’s direct fax systems for drug alerts or via the Chief Medical Officer’s Public Health link and are also published on our website. Alternatively, health professionals may receive a letter from the pharmaceutical company which is responsible for the medicine.
  • publishing safety alerts on our website
  • fact sheets are produced for major safety issues for both healthcare professionals and patients
  • in rare circumstances, removal of the medicine from the market, if the risks of a medicine are found to outweigh the benefits.

For more information please see How is Yellow Card data used to improve patient safety?

Patients and health professionals reporting suspected adverse drug reactions to the Yellow Card Scheme help contribute to these important processes. Suspected reactions can be reported online at www.mhra.gov.uk/yellowcard


Page last modified: 12 August 2011