Individual case safety reports
An individual case safety report starts as a single report of a suspected adverse event. The report becomes a valid case when it carries a minimum of information, and the case is then categorised by whose product it involves, the role of each medicine the patient took, who reported it, and whether the event was serious, caused by the product and expected. It reaches the regulator within a timeline counted from the day the company became aware of it, in a format an international standard sets, and in Europe it goes to the EudraVigilance database, whose access policy sets who can see the reports.
What makes a report a valid case
An individual case safety report provides information on an individual case of a suspected adverse reaction to a medicine. A report is a valid case when it has the four pillars of individual case processing: a company product, a patient who took the product, an adverse event, and a reporter who reports it. These four are the mandatory criteria, the minimum safety information a report needs. In pharmacovigilance, the product is anything with medicinal properties that is used to treat or prevent a disease or disorder in a human being, or that takes a corrective action or modifies a physiological function. In the clinical setting, safety reporting begins when there is identifiable patient exposure to the product, an identifiable reporter and an adverse event. The causality of the reported event should then be established if possible, and its severity, its seriousness or medical significance, and its expectedness should also be considered.
Suspicion is enough: even where the patient cannot be identified or the reporter is not known, a suspicion that the company's product caused an adverse event makes the report a valid case, and the case goes into the safety database. A literature article whose author describes five patients who took aspirin and had a headache is a reportable adverse event for the pharmaceutical company, although the patients themselves are not known. Even a suspected product and adverse event alone can be reported to the pharmaceutical company, which is then obliged to report them to the regulatory authority.
Whether the product is the company's
Generic products without brand names are one of the biggest challenges in pharmacovigilance, and authors and reporters often do not record a product's brand name. Paracetamol, for example, may be made by more than a hundred or two hundred companies, so where a case names paracetamol and no brand, the conservative approach is for the company to treat the product as its own and enter the case in its safety system.
A marketing authorisation holder that markets a product across the globe is responsible for entering the cases of that product from across the globe in its system. It is also responsible for collecting the cases of the partners it licenses the product to, so that the safety profile of the product in its system covers its licensed partners. Whether a case involves the company's drug depends on the authorisations and licences it holds across the globe and on its local agreements with partners.
Manufacturers are required to collect, analyse and report adverse events to all countries where the product is marketed, working closely with the regulatory agencies in those countries to identify and understand the safety signals. They are also required to examine reports from the scientific literature and from marketing experience in other countries.
The company also confirms that it markets the product in the country the case comes from. A company that markets paracetamol everywhere except Croatia assumes that a case from Croatia does not involve its product, while it enters a paracetamol case from anywhere else on the suspicion that its product is involved, since a generic has a high probability of being its product. If follow-up with the customer or the reporter shows that the product was another company's, the company nullifies the case, which no longer exists for it.
A literature case from a country where the company does not market the product is not a valid case for it. A company based in France that does not market the product in Germany, and finds a German literature case in a EudraVigilance download, does not enter the case in its safety system as an individual case.
The role of each medicine the patient took
Each medicine the patient took is categorised by when it was taken relative to the adverse event. A suspect drug is any drug taken at the time of the adverse event, during its timeframe or before it, which is believed to be able to cause the reaction. When a headache follows within three hours of taking aspirin, there is a high probability that aspirin is the suspect drug.
A concomitant medication is taken in parallel with the suspect drug, during the time of the side effect, but on medical judgement may not have contributed to it. Whether a drug is suspect or concomitant depends on the product profile, as when a patient with gastrointestinal bleeding took paracetamol and another drug. Without the medical insight to tell them apart, both drugs taken in the same timeframe are kept as suspect, because both may have contributed, while with that knowledge a drug can be categorised as concomitant.
Most of the time, pharmaceutical companies conservatively keep their own product as the suspect drug for any adverse event on it. Another company's product is kept as concomitant unless there is a high probability that it contributed to the side effect, and as suspect if there is.
Past or historical drug therapy is a product taken a long time before the adverse event, and it is judged by the product's half-life. Paracetamol taken ten days before gastrointestinal bleeding is past drug therapy, because there is no probability that it is still in the body and causing the event after ten days.
A drug taken the day before a side effect is always a suspect drug, because there is a high probability that part of its active ingredient is still in the body and can cause the side effect. Whenever there is doubt, the company's drug is best kept as suspect. The half-life is always considered in deciding whether a drug is past therapy, suspect or concomitant, and the product label is checked for the side effects the product is known to cause. A product known to cause the side effect may well have contributed to it, and it is then kept as a suspect drug. By the time a product is launched or is in clinical trials, earlier clinical analysis has defined its half-life, and the label states it, so the drug can be categorised from the label.
Corrective treatment is given for the adverse event and started after it, such as an antibiotic for an infection the patient had on the drug, and it is neither a suspect drug, a concomitant medication nor past therapy. If the patient has diarrhoea and vomiting after the corrective treatment starts, the corrective treatment becomes a suspect drug for these new side effects.
The reporter
The European good pharmacovigilance practice guidelines distinguish reporters who are medically qualified from those who are not. A report from a physician, dentist, pharmacist, nurse, coroner or anyone local regulations specify as a healthcare professional is understood to be a medically confirmed case, because the reporter has medical knowledge and understanding. A report from someone without medical knowledge is not a medically confirmed case, but it still goes into the system.
Who counts as medically qualified depends on local laws and regulations. A dentist reporting from Thailand may not be considered a healthcare professional, because local law does not allow a dentist to be called completely medically qualified, and in some countries nurses are not considered medically qualified.
Some reports are reportable to health authorities only when they come from a healthcare professional. So the pharmaceutical company follows a case up to understand the exact adverse event, its validity and what the reporter is trying to communicate, has it validated by a healthcare professional, confirms whether the case is medically confirmed and updates the system accordingly.
Adverse events and adverse reactions
An adverse event is anything that happens while the patient is on the product, and an adverse reaction is anything related to the product. Causality is the difference between them: an adverse reaction is due to the product, with the reporter confirming a high probability that the patient had it because of the product.
A car accident while a patient is on a product is an adverse event, because it happened while the patient was on the product. Some products can make a patient drowsy or lose consciousness and so lead to an accident, but whether or not the drug caused it, the accident is always an adverse event.
An abnormal laboratory finding is also an adverse event. Low blood pressure in the evening after paracetamol taken in the morning is an adverse event, because it happened while the patient was on the product. The company then assesses whether it is really due to the product, and the reporter's justification feeds the assessment. If the physician reports that the low blood pressure is not due to the product, it is an adverse event and not an adverse reaction. If the physician says it is due to the product, it is an adverse reaction.
What counts as an adverse event
An adverse event is not only a physiological side effect the patient has, but also anything to do with the handling and the quality of the product. Adverse events include:
- an intentional overdose, where the patient purposely took more than they were supposed to, such as 20 grams instead of 10, and an accidental overdose, where the patient took it by mistake
- an intentional overdose to cause harm or to commit suicide, which is captured as more than one adverse event because the drug can also cause suicidal ideation: as a fatal overdose if the patient died, and otherwise as an overdose, a suicidal ideation and an incomplete suicide
- drug misuse, taking the product for an unintended purpose, such as a drug taken because the patient does not sleep well
- abuse, intentional use for an unapproved indication or for recreational purposes
- lack of efficacy, where the product does not give its intended effect
- a product quality issue, as when a patient reports that the drug smelled bad
- drug exposure during pregnancy, where the patient was not supposed to take the drug and took it while pregnant
- hospitalisation, captured as a worst-case approach even when the reason is not known, so that follow-up can find what led to it
- occupational exposure of people who work in a facility where the product is manufactured and inhale its fumes or are otherwise exposed to it, or of physicians and nurses who handle the product and accidentally inject themselves or are exposed to it, captured whether or not a side effect follows, because the exposure can lead to a chain of events with a possible adverse event
- a medication error, any wrong handling, management or administration of the product, which is an adverse event even where no side effect happens
A patient enrolled in a clinical trial can have a side effect before receiving the investigational medicine. Some exceptions require such cases to be entered in the database, but a case is not entered every time a product is not involved. How finely adverse events are categorised depends on how detailed the coding that captures them is.
Seriousness and severity
Severity is the intensity of an adverse event's symptoms, from mild to severe. Seriousness, or medical significance, is whether an adverse event poses a threat to the patient's life or function or results in hospitalisation or prolonged hospitalisation.
An adverse event is serious when it:
- is fatal, the patient dying after taking the product
- is a life-threatening condition
- causes hospitalisation or prolongs it, as when a patient's three-day stay in hospital for another reason is extended by ten days because of a side effect of the product
- causes a disability or permanent damage to body functions, such as blindness, hearing loss or loss of taste
- is a congenital anomaly or birth defect, as when babies are born with side effects of a product their mother took
- is an important medical event, a very serious event that falls in none of the other categories, such as tachycardia or suicidal ideation that the drug is known to cause, which a physician can upgrade to serious on medical judgement
A lack of efficacy is a serious case by default when the product is for a life-threatening condition, a vaccine or an oral contraceptive.
Causality
Causality is the relationship between the adverse event and the product. Different authorities and companies assess it by different methods. Many organisations use algorithms to ensure consistency, such as a set of ten questions whose answers are scored to decide whether an event is an adverse event or an adverse reaction. Sometimes the assessment is purely medical or clinical judgement. For signals and other work that needs statistical methods, statistical methodologies decide whether the drug causes the adverse reaction.
Expectedness and the reference safety information
A product's label records what has been noted as its side effects in clinical trials and in the past, in small and large populations and in documented evidence. The global label is the company core data sheet, which compiles the product's data from across the world. Each country has its own local label on the product's properties, its handling, its expected side effects and their categorisation, such as the summary of product characteristics in Europe, the package insert in the United States and the product monograph in Canada. In a clinical trial the document is the investigator's brochure. Together these are the reference safety documents, which are prepared during clinical trials, updated throughout the product's lifecycle and named differently at different stages, and which state the side effects listed for the product. The side effects found in phase 1, 2 and 3 trials contribute to updating the investigator's brochure and, in turn, the company core data sheet, the summary of product characteristics and the other local reference safety information documents.
A serious adverse reaction is a serious adverse event on the product where causality is suspected. A suspected unexpected serious adverse reaction is a serious adverse reaction that the product is suspected to have caused but that is not in the label, so it is unexpected.
Day zero
Day zero, from which the reporting timelines start, is the day the pharmaceutical company comes to know about the adverse event, and from a regulatory perspective it starts when anybody in the company comes to know of it. An employee who hears at a conference that someone's relative had a side effect on one of the company's products two months earlier has to inform the company within 24 hours of the day they learnt of it. Most companies have an internal rule that an employee who learns of a side effect related to the product reports it within 24 hours.
For a literature case report, day zero is the day the company reviewing the literature read the reference and confirmed that the product was its own, not the day the journal or reference arrived by email or post. That day is the day of awareness.
Reporting timelines
In one pharmacovigilance practitioner's account, the reporting timelines follow the seriousness criteria. Death and life-threatening cases are typically reported within seven days and serious cases within 15 days, for all the health authorities across the world. A 90-day timeline for non-serious cases currently applies only in the European Union. Some regulatory authorities in the Middle East and Latin America ask for non-serious cases without a defined criterion, while Europe and the United States define the 15-day, seven-day and 90-day timelines clearly. An event a physician upgrades to serious on medical judgement is reported within seven or 15 days.
In a journal account of safety monitoring, post-marketing safety reports must be expedited for adverse events that are both serious and unexpected, whatever their source. Non-serious adverse events, expected or not, and serious expected ones go into non-expedited reports. The two accounts differ on whether a serious expected case is expedited.
The report's international format
In January 2015 the European Medicines Agency published a guide, developed jointly with the Heads of Medicines Agencies, supporting implementation in the EU of a new international standard for the safety monitoring of medicines, the ISO standard for individual case safety reports, whose use was to take effect on 1 July 2016. The standard aimed to establish the same format for reports of individual cases of suspected side effects worldwide, was expected to carry better information about the medicines that might be associated with a reaction and about their therapeutic uses, and strengthened the protection of personal data in the records that companies and regulators hold. The expected effects were better data quality, better ability to search and analyse the reports, and faster detection and handling of safety issues by regulators, and so better protection of patients.
The guide defines the electronic transmission process, the format and content of the report, the business rules for validating one, and the classification and data quality principles, and it was to assist software providers and IT developers as pharmacovigilance databases were developed as well as the companies and regulators preparing to use the standard. It is based on the ICH E2B(R3) guideline and the corresponding ISO standard as referred to in Article 26 of Commission Implementing Regulation (EU) No 520/2012, and it applies with the use of the standard from 1 July 2016.
EudraVigilance
EudraVigilance is the European database of all suspected adverse reactions reported for medicines authorised in the European Economic Area. The European Medicines Agency manages it on behalf of the EU medicines regulatory network, and in 2015 it received over one million adverse drug reaction reports a year. Its large datasets provide the backbone for the continuous safety monitoring of medicines in the EU.
The finalisation of the guide to the ISO standard was a major step in the agency's preparation for an enhanced EudraVigilance system, as the EU pharmacovigilance legislation requires. The agency announced in May 2017 that a new and improved version of EudraVigilance, the European information system of suspected adverse reactions to medicines authorised or under study in the European Economic Area, would go live on 22 November 2017 with enhanced functions for reporting and analysing suspected adverse reactions, supporting better safety monitoring and a more efficient reporting process. National competent authorities, marketing authorisation holders and clinical trial sponsors had to make final preparations to ensure their processes and local IT infrastructure were compatible with the new system and the internationally agreed format, and the agency would support them through targeted e-learning, face-to-face training, webinars and information days. The agency expected these benefits from the new system:
- Reporting of individual case safety reports, and their re-routing to member states, would be simplified, because holders would send them directly to EudraVigilance rather than to the national authorities, which would ultimately reduce duplication of effort.
- Detection of new or changing safety issues would improve, enabling rapid action to protect public health.
- Transparency would increase through broader access to reports of suspected adverse reactions by healthcare professionals and the general public through the database's public interface.
- Search would be enhanced and data analysis more efficient, and system capacity and performance larger, to support large volumes of users and reports, including non-serious reactions originating in the Economic Area.
- Collaboration with the World Health Organization would become more efficient, because the agency would make reports of individual cases within the Economic Area available to the Uppsala Monitoring Centre directly, a task member states would no longer need to carry out.
Two things were to stay unchanged: reporting of adverse reactions by patients and healthcare professionals to national authorities through local spontaneous reporting systems, and the reporting of suspected unexpected serious adverse reactions during clinical trials, until the new Clinical Trial Regulation applied.
Access to the reports in EudraVigilance
The agency has made EudraVigilance data publicly available since 2011, with levels of access to information on adverse reaction reports defined per stakeholder group: European regulators, healthcare professionals, consumers and patients, marketing authorisation holders, and academia. Information on centrally authorised products and on substances commonly used in medicines is available through a dedicated public website.
In December 2015 the agency's management board adopted a revision of the EudraVigilance access policy, under which the agency would give increased access to reports on suspected adverse reactions to medicines authorised in the EU while guaranteeing that personal data would be fully protected. The adoption followed a broad public consultation that generated close to 400 comments, which were taken into account in the final policy. The revised policy takes into account the changes the pharmacovigilance legislation introduced to the system of safety monitoring, such as new transparency provisions, direct patient reporting across all EU member states and simpler reporting of adverse reaction reports for pharmaceutical companies. Its key changes were these:
- The public would have access to more information, including line listings of the side effect reports and summary presentations for individual adverse reaction reports received in EudraVigilance, while patients and those who send in reports of suspected side effects would not be identifiable, a significant increase in transparency for the users of medicines.
- Academia could get extended access to data sets on request in support of its research.
- The World Health Organization's Uppsala Monitoring Centre would be added as a new stakeholder group, to be provided with individual case safety reports originating within the European Economic Area.
- Medicines regulatory authorities in countries outside the European Economic Area would be provided with data in line with the World Health Organization's data set on request.
- Marketing authorisation holders of medicines authorised in the EU would be given enhanced access to reports on their medicines in support of their signal detection and other pharmacovigilance obligations.
These changes were to come into effect in the third quarter of 2017, in parallel with a series of technical improvements to the EudraVigilance system.
The agency and the World Health Organization concluded an agreement to transfer data on suspected adverse reactions occurring in the European Economic Area electronically to the Uppsala Monitoring Centre every day. The transfer was to start in 2017, after the new reporting rules within the European Economic Area took effect following a successful audit of the improved EudraVigilance system. The transferred reports contribute to VigiBase, the World Health Organization's global individual case safety report database, on behalf of its Programme on International Drug Monitoring. Better global knowledge of the safety of medicines helps promote their safe use for the benefit of patients worldwide.
What individual case reports make possible
Reporting of adverse events is the crux on which pharmacovigilance runs. When adverse events are reported, individual case reports, aggregate reports and signals of new side effects never reported before all become possible.