VI.Add I.2. General aspects of duplicate cases

Regardless of the system used for collecting and collating ICSRs, there should always be an appropriate mechanism in place for identifying duplicates. The potential causes for duplicates should be carefully taken into account, as well as the appropriate processes to detect and manage them. If duplicates are identified, analysis of the root cause should be performed and corrective action taken, where appropriate.

Examples of common causes of duplicate reports are:

• A consumer and a healthcare professional reporting the same reaction occurrence;
• Multiple health care professionals treating the same patient reporting the same reaction occurrence;
• An reaction occurrence being reported to EV by the original reporter to both the marketing authorisation holder and a competent authority in a Member State;
• Literature reporting of the same reaction occurrence for generics.

Handling duplicate reports typically involves three steps:

(1) searching/detection of duplicates;

(2) confirmation of duplicates; and

(3) management of duplicates.

The identification of potential duplicates in collections of individual cases is a challenge. Duplicates will often either have been submitted by different senders or processed in different reporting systems, and as such case information can be in many instances dissimilar: different terms may have been used to code the same incident, patient information may be of different level of specificity due to differences in the implementation of the personal data protection rules in Member States or the listed medicinal products may be coded differently related to the same incident due to the absence of an international standard on identification of medicinal products. This makes the identification of duplicates a challenging and resource intensive task. It appears to be clear that the problem of duplicate reports is fairly common in spontaneous reporting systems, although there is certainly a lack of published research, both on the extent of the problem and the methods employed to detect them. Even upon the confirmation that reports are indeed duplicates it is not always obvious how to proceed: should the duplicates be maintained in the database or should one of them perhaps be removed from the data set; if so, which one(s)?

Reviewing pharmacovigilance systems for potential duplicates is also considered necessary when evaluating signals e.g. signals of disproportionate reporting (SDR). Such review may be necessary in addition to routine duplicate and data quality checking. Although databases should be screened regularly for potential duplicates, there may be situations when an individual case was reported more than once in the database and may have not appeared initially as a potential duplicate.

All stakeholders are reminded about the duplicate handling provisions laid down in GVP Module VI, whereby the most relevant sections will be also referenced in this document where applicable. The detection and management of duplicates is also an element that needs to be addressed by applicants and marketing authorisation holders as part of their pharmacovigilance system master file (PSMF) (see GVP Module II).