Rethinking Side Effect Frequencies Through Real-World Evidence

In the prescribing information, adverse reaction rates are often taken as the definitive measure of a drug’s safety. However, these data are based on clinical trials and only tell part of the story. In real-world clinical practice, Real-World Evidence complements this knowledge, refining the safety profile and expanding understanding of risks in different patient populations. Such real-world data sources complement the results of clinical trials, clarify the safety profile of drugs, and facilitate continuous monitoring of the benefit-risk balance throughout the product life cycle.

What “Common” and “Rare” Really Mean in Drug Safety Data

When healthcare professionals, pharmacists, or patients review a medication’s package insert, one of the most important sections is usually the list of adverse reactions and their frequency. Terms such as “very common,” “common,” “uncommon,” and “rare” help assess the potential risks of treatment and facilitate more informed healthcare decisions.

Because these data are included in the official product information, many people consider them to be the final, unchangeable characteristics of the drug. However, in reality, the situation is more complex.

The adverse reaction frequencies listed in the package insert are typically based on results from clinical trials conducted before the drug was approved for marketing. These studies provide an initial understanding of the drug’s safety profile. Still, they do not always fully reflect how the drug will perform after it reaches the market and is used by millions of patients.

How Side Effect Frequency Information Is Established

Before receiving marketing authorisation, every medicinal product undergoes several stages of clinical development. During randomised controlled trials, researchers collect information not only about the drug’s effectiveness but also about all adverse events experienced by participants.

After the results are analysed, adverse reactions are classified by frequency of occurrence. In regulatory practice, the following classification system is commonly used:

Frequency Category	Incidence
Very common	≥1/10
Common	≥1/100 to <1/10
Uncommon	≥1/1,000 to <1/100
Rare	≥1/10,000 to <1/1,000
Very rare	<1/10,000

For example, if headache is classified as a common adverse reaction, it means that it was observed in at least 1 out of 100 patients but in fewer than 1 out of 10 participants in a clinical study. These estimates are extremely valuable for healthcare professionals and regulatory authorities. However, they reflect outcomes observed under specific clinical trial conditions rather than in real-world medical practice.

Why Clinical Trials Do Not Show the Complete Picture

Clinical trials remain the gold standard for assessing the safety and efficacy of drugs. They are essential for demonstrating that a drug’s potential benefits outweigh its potential risks. At the same time, every clinical trial has limitations that can affect the accuracy with which it reflects the incidence of adverse reactions in real-world clinical practice.

Limited Number of Participants

Even large Phase III studies typically include only hundreds or thousands of patients. While these numbers are sufficient to detect common adverse reactions, they may be insufficient to detect very rare events. If a particular reaction occurs in one in every 20,000 or 50,000 patients, it may go undetected until the drug is widely available after market launch.

Selected Patient Populations

Clinical trials typically enrol patients who meet strict inclusion and exclusion criteria. Patients’ health status with severe comorbidities, polypharmacy, significant renal or hepatic impairment, and other complicating factors are often excluded or underrepresented. As a result, clinical trial participants may differ significantly from the broader population that will ultimately receive the drug in routine practice.

Controlled Treatment Conditions

During clinical trials, patients are monitored by healthcare professionals. They attend regular follow-up visits, receive detailed treatment instructions, and are carefully assessed for potential side effects. However, in real-world clinical settings, patients may skip doses, change their treatment schedule, take multiple medications simultaneously, or fail to report symptoms to their doctors.

Limited Duration of Follow-Up for Data Collected

Many clinical trials last only a few months or a few years. However, some adverse reactions may only appear after prolonged treatment or cumulative exposure to the drug. As a result, long-term safety risks often become apparent only after the drug is used in routine clinical practice.

What Happens After a Medicine Reaches the Market?

After receiving marketing approval, a drug enters a new stage in its life cycle. While clinical trials may have involved only a few thousand participants, after approval, it may be used by hundreds of thousands or even millions of patients worldwide.

These patients vary significantly in age, comorbidities, genetic background, treatment adherence, and concomitant medication use. At this stage, it becomes possible to evaluate the drug’s safety in real-world settings. This is where pharmacovigilance plays a crucial role. For example, an analysis of 55 large post-marketing studies involving 402,444 patients found that 69% of the studies led to updates in the drug’s safety section. In contrast, 35% resulted in changes to the description of adverse drug reactions.

What Are Real-World Data and Real-World Evidence?

In recent years, the terms “real-world data” (RWD) and “real-world evidence” (RWE) have become increasingly used in pharmacovigilance and regulatory settings. However, these concepts are not identical.

Real-World Data

Real-World Data refers to information on patient health and medication use collected outside traditional randomised clinical trials. Sources of RWD may include:

• Patient-reported outcomes

• Spontaneous adverse event reports

• Electronic health records

• Insurance and administrative databases

• Patient registries and disease registries (e.g., rare diseases)

• Patient support programs and observational studies

• Mobile health applications and digital monitoring tools

Real-World Evidence

And what is Real-World Evidence (RWE)? Real-world evidence is clinical evidence obtained through the analysis and interpretation of data from real-world clinical practice. In other words, RWD is the raw information, while RWE is the conclusions drawn from it. Unlike clinical trials with strict eligibility criteria, RWE reflects the real-world experience of drug use in broad patient populations. This allows for a better assessment of the safety, effectiveness and long-term effects of treatment in everyday medical practice.

How Real-World Evidence Supports Pharmacovigilance

Modern pharmacovigilance increasingly relies on real-world evidence to evaluate the safety of medicines.

Detection of New Safety Signals

One of the primary goals of pharmacovigilance is the identification of safety signals. A safety signal is information suggesting a possible new causal relationship between a medicinal product and an adverse event. By analysing real-world data, safety professionals can identify rare or unexpected adverse reactions that may not have been detected during clinical development.

Identification of Risk Factors

Not all patients face the same risk of experiencing adverse reactions. Real-world evidence helps identify patient groups that may be more vulnerable to specific safety concerns. For example, increased risk may be observed among:

• Elderly patients
• Individuals with renal impairment
• Patients with liver disease
• People receiving multiple medications simultaneously

Such findings can improve prescribing recommendations and support more personalised risk management strategies.

Assessment of Long-Term Safety

Once a medication is widely used, safety data accumulate over many years. This allows pharmacovigilance professionals to evaluate risks that could not be fully assessed during clinical trials. Long-term safety monitoring is particularly important for medicines intended for chronic or lifelong use.

Evaluation of Risk Minimisation Measures

Pharmacovigilance is not limited to identifying risks. It is equally important to determine whether risk minimisation measures are effective. Using real-world evidence, companies and regulatory authorities can assess whether the frequency of specific adverse reactions decreases following the introduction of educational materials, updated warnings, or changes to product labelling.

Why the Definition of a “Common” Side Effect Can Change

One of the most common misconceptions is the belief that the frequency of adverse reactions is a fixed characteristic of a drug. In reality, a medical product’s safety profile constantly changes throughout its lifecycle.

As additional data becomes available, it may become clear that a particular reaction occurs more frequently or less frequently than anticipated during clinical trials. Previously unknown adverse reactions may also be identified.

This is why product information and instructions for use are regularly updated. Therefore, the term “common adverse reaction” should not be considered a fixed characteristic of a drug, but rather a reflection of the best available data at a given time.

The Growing Role of Real-World Evidence in Pharmacovigilance

Modern pharmacovigilance increasingly incorporates real-world evidence to support regulatory decision-making. Data from routine clinical practice are used for:

• Signal detection and evaluation
• Benefit-risk reassessment
• Post-authorisation safety studies
• Evaluation of risk minimisation measures
• Updates to risk management plans

As healthcare systems become increasingly digitalised, the importance of RWE in pharmacovigilance is expected to continue growing.

Technology as a Foundation of Modern Pharmacovigilance

The volume of available drug safety information is growing at an unprecedented rate. Every day, pharmacovigilance specialists must review adverse event reports, scientific publications, clinical trial results, regulatory updates, and data from numerous external sources.

Managing this volume of information manually is becoming increasingly challenging. As a result, pharmaceutical companies are implementing digital solutions, automation technologies, and artificial intelligence tools to support pharmacovigilance activities. These technologies help them more quickly identify potential safety signals, analyse large datasets, and improve decision-making in risk management.

Conclusion

Clinical trials remain the foundation of drug safety assessment. They provide an initial understanding of a product’s safety profile and support regulatory approval decisions. However, no clinical trial can fully replicate the diversity and complexity of real-world clinical practice.

This is why drug safety assessment continues even after marketing authorisation. Real-world clinical practice data complements clinical trial results, helps identify new risks, refines adverse reaction rates, and supports ongoing benefit-risk assessments.

In modern pharmacovigilance, clinical trials and real-world clinical practice data are not competing sources of information. Instead, they form a complementary evidence base that enables a more complete understanding of drug safety throughout the entire product lifecycle. As new data become available after a drug’s launch, healthcare providers gain a clearer understanding of its true safety profile. This continuous improvement in knowledge is one of the primary goals of modern pharmacovigilance and a key element in protecting patient care and safety.