VI.App.2.3. Database Searches
A search is more than a collection of terms used to interrogate a database. Decisions about the database selection, approach to records retrieval, term or text selection and the application of limits need to be relevant to the purpose of the search. For searches in pharmacovigilance, some of the considerations for database searching are described below.
VI.App.2.3.1. Precision and recall
Medical and scientific databases are a collection of records relating to a set of publications. For any given record, each database has a structure that facilitates the organisation of records and searching by various means, from simple text to complex indexing terms with associated subheadings. Search terms (text or indexed) can be linked using Boolean operators and proximity codes to combine concepts, increasing or decreasing the specificity of a search. In addition, limits to the output can be set. When searching, the application of search terms means that the output is less than the entire database of the records held. The success of a search can be measured according to precision and recall (also called sensitivity). Recall is the proportion of records retrieved (“hits”) when considering the total number of relevant records that are present in the database. Precision is the proportion of “hits” that are relevant when considering the number of records that were retrieved. In general, the higher recall searches would result in low precision.
VI.App.2.3.2. Search construction
Databases vary in structure, lag time in indexing and indexing policy for new terms. While some database providers give information about the history of a particular indexing term or the application of synonyms, other databases are less sophisticated. In addition, author abstracts are not always consistent in the choice of words relating to pharmacovigilance concepts or medicinal products/active substances names. When constructing a search for pharmacovigilance, the highest recall for a search would be to enter the medicinal product name and active substance name (in all their variants) only. In practice, additional indexing terms and text are added to increase precision and to reduce the search result to return records that are of relevance to pharmacovigilance. There is a balance to be achieved. It is therefore expected that complicated searches are accompanied by initial testing to check that relevant records are not omitted, however, there is no defined acceptable loss of recall when searching for pharmacovigilance purposes. Term selection should be relevant to the database used and the subject of the search.
VI.App.2.3.3. Selection of product terms
Searches should be performed to find records for active substances and not for brand names only. This can also include excipients or adjuvants that may have a pharmacological effect. When choosing search terms for medicinal products, there are a number of considerations.
- Is the active substance an indexed term?
- What spellings might be used by authors (particularly if the active substance is not indexed)?
- What alternative names might apply (numbers or codes used for products newly developed, chemical names, brand names, active metabolites)?
- Is it medically relevant to search only for a particular salt or specific compound for an active substance?
During searches for ICSRs, it may be possible to construct a search that excludes records for pharmaceutical forms or routes of administration different to that of the subject product, however, restrictions should allow for the inclusion of articles where this is not specified. Search construction should also allow for the retrieval of overdose, medication error, abuse, misuse, off-label use or occupational exposure information, which could be poorly indexed. Searches should also not routinely exclude records of unbranded products or records for other company brands.
VI.App.2.3.4. Selection of search terms
As described previously, there is no acceptable loss of recall when searching published literature for pharmacovigilance. The use of search terms (free text or use of indexing) to construct more precise searches may assist in managing the output. Deficiencies that have been found frequently during Competent Authority inspections include:
- the omission of outcome terms, for example “death” as an outcome may be the only indexed term in a case of sudden death;
- the omission of pregnancy terms to find adverse outcomes in pregnancy for submission as ICSR ;
- the omission of terms to include special types of reports which needs to be addressed as well in periodic safety update reports, for example,
- reports of asymptomatic overdose, medication error, misuse, abuse, occupational exposure;
- reports of uneventful pregnancy.
VI.App.2.3.5. Limits to a search
Some databases apply indexing that allows the application of limits to a search, for example by subject age, sex, publication type. The limits applied to a search are not always shown in the “search strategy” or search string.
If limits are applied, they should be relevant to the purpose of the search. When searching a worldwide medical literature database, titles and abstracts are usually in English language. The use of limits that reduce the search result to only those published in the English language is generally not acceptable. Limits applied to patient types, or other aspects of an article, for example human, would need to be justified in the context of the purpose of a search.
Limits can be applied to produce results for date ranges, for example, weekly searches can be obtained by specifying the start and end date for the records to be retrieved. Care should be taken to ensure that the search is inclusive for an entire time period, for example, records that may have been added later in the day for the day of the search should be covered in the next search period. The search should also retrieve all records added in that period, and not just those initially entered or published during the specified period (so that records that have been updated or retrospectively added are retrieved). This should be checked with the database provider if it is not clear.
Although one of the purposes of searching is to identify valid ICSRs which qualify for submission, the use of publication type limits is not robust. ICSRs may be presented within review or study publications, and such records may not be indexed as “case-reports”, resulting in their omission for preparation of periodic safety update reports from search results limited by publication type.
VI.App.2.4. Record keeping
Records of literature searches should be maintained in accordance with the requirements described in Article 12 of the Commission Implementing Regulation (EU) No 520/2012. Marketing authorisation holders should demonstrate due diligence in searching published medical literature. It is always good practice to retain a record of the search construction, the database used and the date the search was run. In addition, it may be useful to retain results of the search for an appropriate period of time, particularly in the event of zero results. If decision making is documented on the results, it is particularly important to retain this information.
VI.App.2.5. Outputs
Databases can show search results in different ways, for example, titles only or title and abstract with or without indexing terms. Some publications are of obvious relevance at first glance, whereas others may be more difficult to identify. Consistent with the requirement to provide the full citation for an article and to identify relevant publications, the title, citation and abstract (if available) should always be retrieved and reviewed.
VI.App.2.6. Review and selection of articles
It is recognised that literature search results are a surrogate for the actual article. Therefore, it is expected that the person reviewing the results of a search is trained to identify the articles of relevance. This may be an information professional trained in pharmacovigilance or a pharmacovigilance professional with knowledge of the database used. Recorded confirmation that the search results have been reviewed will assist in demonstrating that there is a systematic approach to collecting information about suspected adverse reactions from literature sources. It is recommended that quality control checks are performed on a sample of literature reviews / selection of articles to check the primary reviewer is identifying the relevant articles.
A common issue in selecting relevant articles from the results of a search is that often this process is conducted for the purposes of identification of ICSRs only. Whereas the review should also be used as the basis for collating articles for the periodic safety update report production, therefore relevant studies with no ICSRs should also be identified, as well as those reports of events that do not qualify for submission as ICSR (see VI.C.2.2.3.2. for the exclusion criteria in the submission of ICSRs published in the medical literature).
Outputs from searches may contain enough information to be a valid ICSR, in which case the article should be ordered. All articles for search results that are likely to be relevant to pharmacovigilance requirements should be obtained, as they may contain valid ICSRs or relevant safety information. The urgency with which this occurs should be proportionate to the content of the material reviewed and the resulting requirement for action as applicable for the marketing authorisation holder.
Articles can be excluded for the submission of valid ICSRs by the marketing authorisation holder if another company’s branded medicinal product is the suspected medicinal product. In the absence of a specified medicinal product source and/or invented name, ownership of the medicinal product should be assumed for articles about an active substance. Alternative reasons for the exclusion of a published article for the submission of ICSRs are detailed in VI.C.2.2.3.2.
VI.App.2.7. Day zero
As described in VI.B.7., day zero is the date on which an organisation becomes aware of a publication containing the minimum information for an ICSR to qualify for submission. Awareness of a publication includes any personnel of that organisation, or third parties with contractual arrangements with the organisation. It is sometimes possible to identify the date on which a record was available on a database, although with weekly literature searching, day zero for the submission of an adverse reaction present in an abstract is taken to be the date on which the search was conducted. For articles that have been ordered as a result of literature search results, day zero is the date when the minimum information for an ICSR to be valid is available. Organisations should take appropriate measures to obtain articles promptly in order to confirm the validity of a case.
VI.App.2.8. Duplicates
Consistent with the general requirements for the submission of cases of suspected adverse reactions, literature cases should be checked to prevent the submission of duplicates ICSRs. It is, therefore, expected that ICSRs are checked in the organisation database to identify literature articles that have already been submitted. Where applicable, this should include ICSRs resulting from the Agency’s Medical Literature Monitoring activities in accordance with Article 27 of Regulation (EC) No 726/2004.
VI.App.2.9. Contracting out literature search services
It is possible to use the services of another party to conduct searches of the published medical literature. In this event, the responsibility for the performance of the search and subsequent submission of ICSRs still remains with the exception of the provisions set out in Article 27 of Regulation (EC) No 726/2004 and Article 107(3) of Directive 2001/83/EC. The transfer of a pharmacovigilance task or function should be detailed in a contract between the organisation and the service provider. The nature of third party arrangements for literature searching can range from access to a particular database interface only (access to a technology) to full literature searching, review and ICSRs submission (using the professional pharmacovigilance services of another organisation). It is recognised that more than one organisation may share services of a third party to conduct searches for generic active substances. In this instance, each organisation should satisfy itself that the search and service is appropriate to their needs and obligations.
Where an organisation is dependent on a particular service provider for literature searching, it is expected that an assessment of the service(s) is undertaken to determine whether it meets the needs and obligations of the organisation. In any case, the arrangement should be clearly documented.
The clock start for the submission of ICSRs begins with awareness of the minimum information by either the organisation or the contractual partner (whichever is the earliest). This also applies where a third party provides a review or a collated report from the published medical literature, in order to ensure that published literature cases are submitted as required within the correct time frames. That is, day zero is the date the search was run if the minimum criteria are available in the abstract and not the date the information was supplied to the organisation.
VI.App.2.10. Electronic submission of copies of articles on suspected adverse reactions published in the medical literature
In accordance with Article 28(3) of the Commission Implementing Regulation (EU) No 520/2012, upon request of the Agency, the marketing authorisation holder that transmitted the initial report shall provide a copy of the relevant article taking into account copyright restrictions, and a full translation of that article into English.
Table VI.4. Electronic submission of copies of literature articles/translations in line with ICH-E2B
| Reference | E2B(R2)/(R3) requirements |
|---|---|
| ICH-E2B(R2) | 1. Mailing address and format of literature articles: Literature articles reportable to the Agency should be provided in PDF format and sent via e-mail to the following e-mail address: EVLIT@ema.europa.eu. In relation to copies of articles from the published medical literature, marketing authorisation holders are recommended to consider potential copyright issues specifically as regards the electronic submission and handling of electronic copies in the frame of regulatory activities.2. File name of literature articles sent in electronic format to the Agency: The file name of a literature article sent in PDF format should match exactly the data element A.1.10.1 or A.1.10.2 ‘World-Wide Unique Case Identification Number’ assigned to the individual case, which is described in the article and which is provided in the E2B(R2) ICSR format.If there is a follow-up article to the individual case published in the literature, the file name with the World-Wide Unique Case Identification Number must be maintained but should include a sequence number separated with a dash.Examples:
3. Submission of cases described in the medical literature referring to more than one patient: When the literature article refers to the description of more than one patient, the copy of the literature article should be sent only once. The file name of a literature article sent in PDF format should match exactly data element A.1.10.1 or A.1.10.2 as applicable ‘World-Wide Unique Case Identification Number’ assigned to the first submitted individual case described in the article. In addition, all ICSRs which relate to the same literature article should be cross referenced in data element A.1.12 ‘Identification number of the report which is linked to this report’. The data element should be repeated as necessary to cross refer all related cases. |
| ICH-E2B(R3) | Information on document attachments:
“International Committee of Medical Journal Editors. Uniform requirements for manuscripts submitted to biomedical journals. N Engl J Med 1997; 336:309-15. doi:10.1056/NEJM199701233360422.”
|
VI.App.2.11. Examples for the submission as ICSRs of suspected adverse reactions described in the medical literature and referring to more than one patient
Table VI.5. Examples for the submission as ICSRs of suspected adverse reactions described in the medical literature and referring to more than one patient
| Ex. | Scenario | Action |
|---|---|---|
| 1 | A literature article describes suspected adverse reactions that have been experienced by more than one identifiable patient.
ICSRs should be created and submitted for each individual identifiable patient described in the literature article. Each ICSR should contain all the available information on the case. The cross reference with all the linked ICSRs from this literature article should only be provided in the first submitted ICSR, in the data element ICH-E2B(R2) A.1.12 ‘Identification number of the report which is linked to this report’. There is no need to repeat all the cross references in the other ICSRs. |
For the ICSRs which relate to the same literature article, the cross reference in the data element ICH (E2B(R2) A.1.12/ ICH-E2B(R3) C.1.10.r ‘Identification number of the report which is linked to this report’ should be conducted as follows:
Example for the submission of cases originally described in the medical literature referring to a large number of patients: For case 1 described in the literature article:
For Case 2 described in the literature article:
For Case N described in the literature article:
|