What is MedDRA? A plain-English guide for pharmacovigilance and regulatory professionals

MedDRA — the Medical Dictionary for Regulatory Activities — is the international standardised medical terminology used to classify adverse events, diseases, symptoms, diagnoses, and indications in pharmaceutical regulatory submissions and safety reporting.

If you work in pharmacovigilance, regulatory affairs, clinical data management, or medical writing, you will encounter MedDRA. Understanding its structure is not optional — it determines how cases are coded, how signals are detected, and how safety data is communicated to regulators worldwide.

The five-level MedDRA hierarchy

• System Organ Class (SOC): the broadest level — 27 categories organised by body system, aetiology, or purpose. Example: 'Cardiac disorders'. Each PT belongs to a primary SOC.
• High Level Group Term (HLGT): groups related HLTs under a broader concept within a SOC. Example: 'Cardiac arrhythmias'.
• High Level Term (HLT): groups related PTs. Example: 'Ventricular arrhythmias and cardiac arrest'.
• Preferred Term (PT): the primary coding level. Each PT is a distinct medical concept with a precise definition. Example: 'Ventricular tachycardia'. This is what appears on ICSR submissions and regulatory databases.
• Lowest Level Term (LLT): the verbatim-closest level — maps to a PT. Used to capture what the reporter actually said. Example: 'V tach' maps to the PT 'Ventricular tachycardia'.

Why the Preferred Term is the most important level

When pharmacovigilance professionals code an adverse event, they code to the Preferred Term (PT). The PT is the standardised concept used in regulatory submissions, EudraVigilance, FAERS, safety databases, and signal detection queries.

The LLT is what the patient or reporter said — it captures verbatim language but must always map to a PT. If you see 'heart racing' in a case narrative, the coder will map it to the appropriate PT (possibly 'Palpitations' or 'Tachycardia') based on clinical context.

SOC is used for aggregate analysis and signal detection queries — for example, reviewing all cases in the 'Cardiac disorders' SOC. PTs can belong to more than one SOC (multiaxiality), but each PT has one primary SOC.

Common MedDRA coding mistakes to avoid

• Coding at the LLT instead of the PT: always use the PT for ICSR submissions — LLTs are verbatim capture only.
• Choosing the closest-sounding term without reading the PT definition: MedDRA terms have precise scope notes. 'Rash' and 'Dermatitis' are different PTs — check definitions before coding.
• Over-coding: each distinct medical concept needs its own PT, but do not split a single event into multiple PTs unless they genuinely represent different conditions.
• Ignoring multiaxiality: a PT can link to multiple SOCs. Verify the primary SOC is correct for reporting purposes.
• Using outdated MedDRA versions: MedDRA is updated twice yearly (March and September). Use the version current at the time of data lock or as specified by the sponsor.

How MedDRA connects to signal detection

Signal detection relies on MedDRA coding consistency. If the same event is coded differently across cases — for example, 'Myocardial infarction' in some cases and 'Acute coronary syndrome' in others — statistical disproportionality analyses will not aggregate them correctly, and signals may be missed.

Standardised Medical Queries (SMQs) are pre-defined groupings of MedDRA PTs used to search safety databases for specific conditions or drug effects. For example, an SMQ for 'Hepatotoxicity' groups dozens of liver-related PTs so that signal analysts can query for all relevant cases at once.

This is why coding accuracy is not a bureaucratic exercise — it is a patient safety function. The quality of signal detection depends directly on the quality of adverse event coding.