The Process of Pharmacovigilance

The goal is not just to collect data on harms, but to actively use that data to minimize risks and promote the safe use of medicines. The process can be broken down into five key stages:

1. Signal Detection
2. Evaluation & Investigation
3. Taking Action
4. Communication
5. Crisis Management

1. Signal Detection

• What is a Signal?
  A signal is an alert or a piece of information that suggests a potential new causal relationship between a drug and an adverse event, or that a known risk is more serious or more frequent than previously thought. It is a hypothesis for further investigation, not a confirmed fact.
• How are Signals Detected?
  The most common source for initial signals is spontaneous ADR reporting databases. To find signals in these large databases, a method called “data mining” is used. This involves using statistical measures to find “disproportionality” – that is, drug-event combinations that are reported more often than you would expect by chance alone.
• Key Tool: Proportional Reporting Ratio (PRR)
  • The PRR compares the proportion of all reports for a specific drug that are for a particular ADR, against the proportion of that same ADR for all other drugs in the database.
  • Example: rifabutin and the eye condition uveitis. While uveitis made up only 0.13% of all ADRs in the database, it accounted for 75% of all ADRs reported for rifabutin. This massive disproportionality (PRR = 556) was a clear signal.
  • A common rule of thumb for a potential signal is: PRR >= 3, Chi-squared >= 4, and at least 3 reported cases.

2. Evaluation and Investigation

Once a signal is identified, it must be prioritized and evaluated.

• Prioritization: Not all signals are equally urgent. Systems like “triage” (WHO) or “impact analysis” (UK’s MHRA) are used to score signals based on:
  • Evidence Score: Strength of the statistical signal, plausibility, and quality of case reports.
  • Public Health Score: Seriousness of the event, number of people using the drug, and potential health impact.
• Signal Evaluation: This is a detailed clinical review asking key questions:
  • Causality: Is it likely the drug actually causes this effect? (Using principles from Chapter 2, like temporality, alternative causes, etc.).
  • Frequency: How common might the reaction be? (e.g., very common >1/10, rare <1/10,000).
  • Seriousness: Is the reaction life-threatening, fatal, or disabling?
  • Preventability: Are there identifiable risk factors? Could it be prevented with dose adjustment or monitoring?
• Investigation: If the signal is strong enough, formal studies are launched:
  • Pharmacoepidemiological Studies: To quantify the risk and confirm (or refute) the association in a large population (e.g., cohort or case-control studies).
  • Mechanistic Studies: Laboratory research to understand how the drug might be causing the effect.

3. Taking Action

The ultimate goal of pharmacovigilance is to prevent harm. If evaluation confirms a risk, actions are taken to manage it.

• Common Actions (Risk Minimization):
  • Updating the Summary of Product Characteristics (SPC or “label”):
    • Adding new Contraindications or Warnings.
    • Changing Dosage instructions for specific groups.
    • Adding advice for Monitoring patients.
    • Listing the new ADR in the Undesirable Effects section.
  • Issuing “Dear Healthcare Professional” letters to inform prescribers directly.
• More Drastic Actions:
  • Restricted Access Programs: For high-risk drugs, supply can be linked to mandatory monitoring (e.g., the clozapine monitoring system where patients must have regular blood tests to get their prescription).
  • Market Withdrawal: Removing the drug from the market entirely. This is a last resort, used only when the risks outweigh the benefits and the risks cannot be adequately minimized.

4. Communication

Communicating drug safety information effectively is crucial but challenging. Poor communication can cause public scares and lead to patients stopping necessary treatments.

• The “ABOUT” Principles:
  • Accurate: Facts and numbers must be correct.
  • Balanced: Both risks and benefits must be presented.
  • Open: Be honest and don’t hide the hazard.
  • Understandable: Use clear, simple language.
  • Targeted: Tailor the message to the audience (doctors vs. patients).
• Basic Communication Model:
  1. State the drug and the hazard.
  2. Summarize the evidence.
  3. Explain what is being done (e.g., label change, new study).
  4. State the implications for doctors and patients.
  5. Provide a balanced view of risks and benefits.
  6. Give contact details for more information.

5. Crisis Management

When a major, newly identified hazard suggests that a drug’s risks may outweigh its benefits, it becomes a crisis.

• This requires a pre-planned Crisis Management Plan.
• A dedicated crisis team is formed to:
  • Rapidly evaluate the evidence.
  • Make urgent decisions (often about suspension or withdrawal).
  • Manage practical implementation.
  • Develop and deliver clear external communications.