How to Perform Safety Literature Monitoring
Run a documented literature search that finds reportable cases before an authority does.
What a how-to is not
A how-to is SPEQ’s practitioner method, not a procedure. It does not replace your own SOP, it is not a validated approach, and the judgement calls in it belong to your quality unit.
Systematic literature review is a recognised source of post-approval safety information, and awareness of a case in the literature starts the same reporting clock as a spontaneous report. The work is not reading papers — it is designing a search that can be repeated and defended, and evidencing that it ran on schedule. The usual finding is not a missed case but an unreproducible search.
- 1
Define the scope by substance, not by brand
Search on the active substance and its synonyms, including salts, codes and former names, because the literature will not use your product name. A search built on the brand name alone will systematically miss the academic literature where most case reports appear.
- 2
Specify the databases and the search strategy
Record the databases searched, the full search string, the fields covered and the date range. The strategy is part of the record: a result set that cannot be regenerated from the documented strategy cannot be shown to have been complete.
- 3
Fix the frequency and hold to it
Set a search frequency proportionate to the product and its obligations, and run it on schedule. Awareness dates from the day the search is performed, so an overdue search does not pause the clock — it compresses it.
- 4
Screen against explicit inclusion criteria
Screen abstracts against written criteria for validity and relevance, and record the decision for each. The screening record is what demonstrates that an excluded article was considered rather than never seen.
- 5
Process qualifying articles as cases
For articles meeting the minimum criteria, create the case with the publication as the source, and seek the full text where the abstract is insufficient to assess seriousness or causality.
- 6
Retain the evidence of each cycle
Keep the strategy, the run date, the retrieved set, the screening decisions and any cases created. This bundle is the answer to the only question an inspector actually asks about literature monitoring: show me that it ran, and show me what it found.
- !Searching the brand name rather than the active substance and its synonyms.
- !A search strategy documented so loosely that the result set cannot be regenerated.
- !Treating an overdue search as though it delays the reporting clock, when awareness dates from the day it runs.
- !Excluded abstracts discarded with no record, leaving no evidence they were screened at all.
How to Perform Safety Literature Monitoring: frequently asked questions
Common questions on perform safety literature monitoring.
When does the clock start for a case found in the literature?
On the day the search is performed and the relevant article is identified — that is the day of awareness. Running a scheduled search late therefore does not buy time; it consumes the reporting window before anyone has read the article.
Is a database search enough, or is manual review needed?
The database search is the systematic backbone, but abstracts still need screening against written criteria by someone competent to judge validity and seriousness, and the full text is often needed where the abstract does not support the assessment.
What has to be retained from each cycle?
The search strategy, the date it ran, the retrieved set, the screening decision for each record, and any cases created. Together they demonstrate both that monitoring occurred on schedule and that its scope was wide enough to find what it should have found.