How to Identify ICSRs in Medical Literature: Step-by-Step Guide for Pharmacovigilance Professionals

Literature screening is a major pillar of pharmacovigilance (PV). It ensures that potential adverse drug reactions reported in scientific publications are captured, evaluated, and reported to regulators. For Marketing Authorization Holders (MAHs), effective literature monitoring is not optional — it is a regulatory obligation. This process, often referred to as literature screening in pharmacovigilance, is essential for timely safety detection.

This guide explains how to identify Individual Case Safety Reports (ICSRs) in literature, based on practical workflow concepts and regulatory expectations, along with real examples and a reviewer checklist.

What is Literature Screening in Pharmacovigilance?

Literature screening in PV involves the systematic review of medical and scientific sources to detect potential safety information related to medicinal products.

Key sources include:

• Published journal articles
• Online medical content
• Conference abstracts
• Unpublished manuscripts (where accessible)

The goal is to identify valid ICSRs and emerging safety signals that may impact the benefit–risk profile of a drug.

Regulatory Expectations for Literature Monitoring

Under global PV requirements:

• MAHs must monitor global and local literature throughout the entire lifecycle of the product.
• Screening should begin as soon as the Marketing Authorization Application (MAA) is submitted and continue while the authorization remains active.
• Searches should primarily use the active substance name rather than brand names.

Common databases reviewed

Regulatory authorities expect routine screening of major biomedical databases such as:

• MEDLINE
• Embase
• Excerpta Medica

Typically, literature surveillance is performed at least weekly.

Why Literature Screening is Important

Effective literature monitoring supports:

• Regulatory compliance (e.g., EMA GVP Module VI)
• Early signal detection
• Benefit–risk evaluation
• Preparation of PSURs/PBRERs
• Ongoing post-marketing surveillance

🔔 Literature findings also contribute to broader safety evaluation and signal detection in pharmacovigilance.

Historically, literature has played a crucial role in safety detection. For example:

• In 1961, a letter published in The Lancet linked thalidomide to severe birth defects.
• In 2005, multiple literature cases identified “floppy iris syndrome” associated with tamsulosin before widespread regulatory reporting.

These examples highlight why literature surveillance remains critical.

Step-by-Step: How to Identify ICSRs in Medical Literature

When screening articles, reviewers should systematically check whether the minimum ICSR criteria are present.

The four minimum criteria

An article may qualify as an ICSR if it contains:

1. Identifiable patient
2. Identifiable reporter
3. Suspected medicinal product
4. Suspected adverse event

If all four are present → potential valid ICSR.

Practical Workflow for Literature Screening

Step 1: Design an Optimal Search Strategy

This is the first major challenge in literature monitoring.

Best practices include:

• Using three or more databases
• Carefully designing search queries
• Applying Boolean operators
• Using proximity searches
• Including relevant abbreviations and synonyms

A poorly designed search is either:

• misses cases (high risk), or
• retrieves too many irrelevant articles (inefficient).

Step 2: Manage High Article Volume

In weekly screening, reviewers often face large numbers of irrelevant articles. The key question becomes:

How can we check effectively without missing important safety data?

Manual screening alone is:

• time-consuming
• overwhelming
• prone to human error

This is why the industry is increasingly shifting toward AI-assisted automation using NLP and machine learning to extract relevant safety data more accurately.

Step 3: Handle Duplicate Data

The second major challenge is duplicate management.

To overcome this:

• Use standardized, well-established deduplication systems
• Compare key identifiers such as:

• author name
• study details
• citation information

Robust deduplication prevents double reporting.

EMA Medical Literature Monitoring (MLM) Service

The EMA operates its own Medical Literature Monitoring (MLM) service.

Important implications

• MAHs are required to submit ICSRs only from literature not already covered by the EMA MLM service.
• However, MAHs must still review literature for their products.

Situations Where Literature Findings May Not Qualify as ICSRs

Certain publications should generally not be submitted as ICSRs, for example:

• The patient cannot be distinctly identified
• The product does not belong to the MAH
• The report originates from a country where the MAH has never marketed the product
• Data comes from competent authority databases or public aggregate databases
• The publication is a meta-analysis or pooled analysis without identifiable individual cases

⚠️ However:
Even if excluded from ICSR submission, these findings must still be discussed in the PSUR/PBRER when relevant.

Handling Valid Cases from Literature

If a valid case is identified:

• The MAH must submit the ICSR electronically according to regulatory timelines.
• Literature citations should follow Vancouver style with DOI where available.

Multiple patients in one article

If a single publication describes multiple patients:

• Create separate ICSRs for each patient
• Submit the article with the first case
• Cross-reference the same article for the remaining cases

This is a commonly tested interview scenario.

✍️ After identifying a valid literature case, the next critical step is narrative development. Learn a structured approach to Pharmacovigilance Narrative Writing with Practice: ICSR Examples for Beginners.

Determining Day Zero

If the minimum criteria are clearly present in the weekly database abstract, then:

• Day 0 = the date the search was conducted

This is a critical compliance point many beginners miss.

Now let’s apply these principles to real screening scenarios.

🔬 Simulated Case Practice for Literature Screening

🧪 Simulated Case 1

Abstract (Simulated):
A 58-year-old female with hypertension developed a persistent dry cough two weeks after starting enalapril. The cough improved after discontinuation. The case was reported by the treating physician.

ICSR Assessment:
✅ Valid ICSR

Justification:

• identifiable patient present
• identifiable reporter present
• suspected drug present
• adverse event present

---

🧪 Simulated Case 2

Abstract (Simulated):
In a study of 2,000 patients receiving Drug A for diabetes, common adverse events included nausea (6%) and dizziness (3%). No individual patient descriptions were provided.

ICSR Assessment:
❌ Not a valid ICSR

Reason:
This is aggregate data without individual case details.

---

🧪 Simulated Case 3 (Tricky)

Abstract (Simulated):
A patient experienced severe headache after initiation of Drug B for migraine prevention. Symptoms resolved after withdrawal. The event was noted in hospital records.

ICSR Assessment:
⚠️ Potential case — requires further assessment

Reason:
Patient is not clearly identifiable (no age or sex mentioned).

 Want to strengthen your case identification skills? Practice more scenarios in our detailed guide on Learning ICSR Processing in Pharmacovigilance: Practical Case Scenarios for Beginners.

📘 You can also explore additional hands-on exercises in My Pharmacovigilance Portfolio – Simulated ICSR Practice to further improve your screening accuracy.

---

🌍 Real-World Literature Example

Article: Chang DF, Campbell JR. Intraoperative floppy iris syndrome associated with tamsulosin. Journal of Cataract & Refractive Surgery, 2005.
PubMed Link: https://pubmed.ncbi.nlm.nih.gov/15899440/

📄 Summary (Paraphrased)

Tamsulosin, an alpha-1 adrenergic antagonist used for benign prostatic hyperplasia, has been associated with intraoperative floppy iris syndrome (IFIS) during cataract surgery. The authors described several elderly male patients who developed IFIS while receiving tamsulosin therapy. The ophthalmologist observed characteristic intraoperative iris behavior that complicated the surgical procedure.

🔍 ICSR Assessment

✅ Contains valid ICSRs

Key Learning Point:
When multiple patients are described in a single publication, separate ICSRs must be created for each identifiable patient, with a cross-reference to the same literature source.

🚨 Common Mistakes in Literature ICSR Identification

• Treating aggregate studies as individual cases
• Missing multiple patients in one article
• Ignoring unclear patient identifiability
• Forgetting Day-0 rules
• Overlooking follow-up requirements
• Confusing review articles with case reports

Avoiding these errors significantly improves literature screening accuracy.

The Future: Shift Toward AI-Driven Literature Screening

Because manual review is resource-intensive, the PV industry is undergoing a major shift toward:

• AI-driven automation
• Natural Language Processing (NLP)
• Machine learning Algorithms

These technologies help:

• Reduce irrelevant article burden
• Improve screening efficiency
• Support accurate data extraction

However, human medical review remains essential for final case validation.

Key Takeaways

• Literature screening is a core pharmacovigilance responsibility.
• MAHs must continuously monitor global and local literature.
• Weekly database searches are standard practice.
• Always check the four minimum ICSR criteria.
• Use strong search strategies and deduplication methods.
• Not all literature findings qualify as ICSRs.
• Multiple patients in one paper require separate case reports.
• AI is increasingly supporting literature surveillance workflows.

Frequently Asked Questions (FAQ)

1. What qualifies as an ICSR in medical literature?

A literature report qualifies as an ICSR when it contains the four minimum criteria: identifiable patient, identifiable reporter, suspected drug, and suspected adverse event.

2. Are aggregate clinical studies considered ICSRs?

No. Aggregate data without individual patient-level information generally do not qualify as valid ICSRs.

3. How many ICSRs should be created from one article?

One ICSR should be created for each identifiable patient described in the publication.

4. What is Day 0 in literature screening?

If the minimum criteria are present in the database abstract, Day 0 is considered the date of the literature search.

5. Is follow-up required for literature cases?

Yes. Follow-up should be attempted when important clinical details are missing and the reporter is contactable.

🔗 Recommended Pharmacovigilance Reads

1. Learning ICSR Processing in Pharmacovigilance
2. My Pharmacovigilance Portfolio – Simulated ICSR Practice
3. Pharmacovigilance Narrative Writing Practice
4. Signal Detection in Pharmacovigilance
5. Medical Coding in Pharmacovigilance (MedDRA Explained)
6. Introduction to Pharmacovigilance (PV)

Final Thoughts

For pharmacovigilance professionals and freshers alike, mastering ICSR identification in literature is a high-value skill. Strong literature screening not only ensures regulatory compliance but also plays a vital role in early signal detection and patient safety.

As the volume of biomedical publications continues to grow, combining structured manual review with intelligent automation will define the future of pharmacovigilance literature monitoring.