Head-to-Head Claims: Inside Pharma Evidence Standards

The weakness is usually structural: evidence has been gathered in one workflow, medical review completed in another, and regulatory substantiation treated as a final approval step rather than as the operating framework for the claim itself.

That model creates a predictable bottleneck. A promotional statement may be technically accurate in isolation yet misleading in context because the comparator is poorly defined, the endpoint is not clinically meaningful, the statistical result is overstated, or a relative risk reduction is presented without its absolute context. When a healthcare professional or decision-maker asks for the supporting evidence, the organization must then retrieve, reconcile, and defend the claim under pressure.

This is the real challenge in comparative claims substantiation pharma: not producing more promotional language, but building a workflow in which every claim is proportionate to the evidence available to support it.

The evidentiary threshold: beyond the head-to-head trial

A head-to-head clinical trial is often treated as the default answer to any comparative claim. That is too simplistic for both compliance and evidence strategy.

Under the ABPI Code, a direct head-to-head trial is not strictly required for every comparative statement. A company may rely on other robust and relevant evidence where the claim remains accurate, fair, balanced, and non-misleading. However, direct comparative data can materially strengthen the substantiation file because they reduce the number of inferential steps between the evidence and the promotional conclusion.

The operational question is not simply whether two products have been studied in the same trial. It is whether the evidence supports the precise proposition being communicated.

A claim that one medicine produces a greater reduction in a defined endpoint than another requires more than two separate efficacy studies placed side by side. The populations, treatment settings, endpoints, assessment periods, dosing conditions, and analytical methods must be sufficiently comparable for the conclusion to remain scientifically defensible. If they are not, the promotional material may imply a comparison the studies never actually made.

Your medical signatory review should therefore begin with a claim-evidence map, not with the finished artwork or presentation deck. For each proposed statement, record:

• The exact wording of the claim, including modifiers such as “greater,” “faster,” “more effective,” or “better tolerated.”
• The comparator and the basis on which it has been selected.
• The patient population and treatment setting represented by the evidence.
• The primary and secondary endpoints relevant to the claim.
• The statistical result and its confidence interval where available.
• The clinical interpretation of the result.
• The limitations, exclusions, and conditions that must appear alongside the claim.
• The source documents that can be produced if substantiation is requested.

This is not administrative overhead. It is the minimum framework required to prevent a broad promotional conclusion from outrunning a narrow clinical result.

A comparative claim is only as strong as the shortest inferential step between the study result and the words placed in front of the prescriber.

What the evidence can support

A well-structured evidence review distinguishes at least three levels of comparative language.

The first is a direct numerical description of an observed result. This may report a measured difference between treatments under the conditions of a defined study.

The second is a statistical conclusion. This addresses whether the observed difference meets the prespecified statistical threshold and how precisely the result has been estimated.

The third is a clinical or promotional conclusion. This is where the organization states that one treatment is superior, faster, safer, more convenient, or otherwise preferable.

These levels are related, but they are not interchangeable. A statistically detectable difference does not automatically establish a clinically important advantage. A numerically larger result does not establish superiority when the studies were not designed for an appropriate comparison. A favourable secondary endpoint does not justify a broad claim about overall treatment performance.

The promotional wording must remain inside the boundary created by the evidence. If the study demonstrates a difference in one endpoint over one period, the claim should not silently expand into a statement about overall efficacy, long-term outcomes, or the entire patient population.

PMCPA Clause 6 and Clause 18.2: substantiation must be operational

The PMCPA framework places substantiation at the centre of pharmaceutical promotion. Under Clause 6 and Clause 18.2 of the PMCPA Code, information, claims, and comparisons must be capable of substantiation. If a healthcare professional or decision-maker requests supporting evidence, the company must provide it within 10 working days.

That response period changes the way your organization should design its review process. The question is not whether the evidence exists somewhere in the medical affairs archive. The question is whether the organization can identify the relevant evidence, confirm its applicability, and provide a coherent response within the required timeframe.

A defensible substantiation process should have clear ownership at each stage:

1. Claim owner — defines the commercial or educational purpose of the statement and prevents uncontrolled wording changes during production.

2. Clinical evidence lead — identifies the relevant studies, endpoints, population data, and statistical outputs.

3. Medical signatory — assesses whether the claim accurately reflects the evidence and whether its balance is adequate.

4. Regulatory or compliance reviewer — tests the claim against the applicable code, local requirements, and promotional context.

5. Document control owner — maintains the approved wording, evidence package, version history, and retrieval pathway.

Without this allocation, the same claim can be modified repeatedly by brand, medical, legal, and agency teams until its final wording no longer matches the original review. That is a governance failure, not a copy-editing problem.

The updated ABPI Code of Conduct came into operation on 1 October 2024, with a transitional implementation period for relevant UK pharmaceutical promotion rules until 31 December 2024. Organizations should not treat code transitions as isolated legal events. They should convert changes into operational controls: revised review templates, refreshed training, updated approval matrices, and explicit ownership for claims already in circulation.

Build the evidence file before launch

A comparative claim should not be considered ready because the final visual asset has been approved. It is ready when the supporting evidence can be retrieved and interpreted without reconstructing the rationale from fragmented emails.

Your substantiation file should contain, at minimum:

• The final approved claim and all material variants.
• The study reports or authoritative evidence supporting the statement.
• The comparator definition and rationale.
• The relevant statistical analyses.
• The clinical interpretation of the result.
• The required qualifiers and limitations.
• The review history and approval dates.
• The intended audience, channel, and promotional context.
• A response owner for any subsequent evidence request.

This approach improves scalability. It also prevents a common failure mode: a company can defend the underlying science but cannot demonstrate why the selected wording was proportionate to that science.

Statistical significance is not clinical superiority

Statistical significance is one of the most frequently mishandled elements in promotional review. The error is not usually a failure to understand the p-value. It is a failure to understand what the p-value does not establish.

A statistically significant difference indicates that the observed result is unlikely to be explained by random variation under the assumptions of the analysis. It does not, by itself, establish that the difference matters to patients, changes clinical decision-making, or justifies a superiority claim in promotional material.

The reverse is also important. A result that does not reach statistical significance must not be presented as clinically meaningful superiority. Under PMCPA guidance, a non-significant difference cannot be converted into a promotional advantage merely because the numerical direction appears favourable.

A robust review separates the following questions:

• Was the analysis prespecified?
• Was the study adequately designed to test the comparison?
• Did the result meet the relevant statistical threshold?
• What is the size of the observed effect?
• How precise is the estimate?
• Does the result cross a clinically meaningful threshold?
• Were multiplicity, missing data, and subgroup analyses handled appropriately?
• Is the conclusion being applied beyond the population and endpoint actually studied?

The final question is where many materials fail. A favourable result in a selected subgroup may be relevant, but it does not automatically support a claim covering all patients. A statistically significant difference in a surrogate endpoint does not automatically support a statement about a patient-important outcome. A post hoc analysis may generate a hypothesis without providing a stable foundation for a broad comparative assertion.

Promotional wording must reflect the analytical design

The language of superiority carries a high evidentiary burden. Terms such as “superior,” “more effective,” “better control,” or “faster response” should be used only where the study design and analysis support that exact conclusion.

Where the evidence is descriptive, the wording should remain descriptive. Where the analysis is exploratory, the promotional treatment should not present it as a definitive finding. Where the comparison is indirect, the material must not create the impression that the products were tested against each other in a controlled head-to-head trial.

This is the point at which medical signatory review must be more than a signature exercise. The signatory is not merely confirming that a citation exists. The review must test whether the claim, the statistical interpretation, and the clinical implication remain aligned.

Statistical significance can validate a result. It cannot perform the clinical interpretation for you.

Absolute risk context: the discipline missing from relative reductions

Relative risk reduction is persuasive because it produces a larger-looking number. That is precisely why it requires disciplined presentation.

A relative reduction without absolute risk context can distort the practical meaning of a treatment effect. The reader may understand the direction of the result while misunderstanding its scale. Under PMCPA guidance, relative risk reductions must be accompanied by absolute risk context.

This requirement is not a formatting preference. It is a protection against a predictable communication failure.

Suppose a treatment reduces an event from a low baseline risk to a lower risk. The relative reduction may appear substantial, while the absolute difference remains modest. The promotional material must allow the healthcare professional to understand both dimensions. Presenting only the relative figure can inflate perceived benefit and obstruct meaningful clinical judgement.

The evidence presentation should therefore define:

• The baseline event rate.
• The event rate with each treatment.
• The absolute risk difference.
• The relative risk reduction, if relevant.
• The timeframe over which the events were measured.
• The population in which the result was observed.
• Any uncertainty around the estimate.

The same discipline applies to safety comparisons. A lower relative rate of an adverse event does not communicate the full risk profile if the baseline incidence, exposure duration, or event definition is omitted.

Avoiding visual distortion

Risk communication can become misleading through design even when the underlying numbers are accurate. Scale differences, selective emphasis, large percentage callouts, and footnotes that carry essential limitations all affect interpretation.

The review process should test the complete asset, not just the headline claim. Ask:

• Does the visual hierarchy give more prominence to the relative figure than to the absolute result?
• Is the denominator clear?
• Can the audience identify the timeframe without searching through small print?
• Are the treatment groups and populations defined?
• Is the uncertainty visible where it affects interpretation?
• Does the claim remain accurate when removed from the full study citation?

This is where promotional material review pharma teams often need stronger integration with design and agency workflows. The evidence review cannot stop at the manuscript. A layout can change the implied meaning of a statement without changing a single word.

Comparative claims across UK and U.S. frameworks

Global pharmaceutical organizations face a second structural risk: assuming that one approved claim can move across jurisdictions without modification.

The PMCPA and ABPI frameworks require claims and comparisons to be capable of substantiation, accurate, fair, balanced, and non-misleading. In the United States, prescription drug comparative advertising is assessed under FDA requirements, with substantial evidence typically requiring two adequate and well-controlled studies. The FDA may accept one such study with confirmatory evidence under the statutory framework of 21 USC § 355(d).

These standards overlap in their concern for scientific reliability, but they should not be collapsed into one global approval rule. The evidentiary threshold, regulatory terminology, presentation expectations, and enforcement environment may differ.

A global claim governance model should distinguish between:

This table is not a substitute for jurisdiction-specific review. It is a control against false harmonization.

The correct strategy is to create a global evidence core and local claim layers. The evidence core contains the validated study results, statistical analyses, clinical interpretations, and limitations. The local layer governs wording, formatting, required qualifiers, approval pathways, and channel-specific restrictions.

That structure improves scalability without pretending that regulatory frameworks are identical.

The workflow bottleneck is usually upstream

Organizations often attempt to solve compliance failures by adding another approval meeting. That is rarely sufficient. If the claim was poorly defined at intake, late-stage review becomes a negotiation between commercial urgency and regulatory caution.

A stronger workflow begins with claim classification.

Classify the claim before drafting

Every proposed comparative claim should be assigned a risk level based on its wording and evidentiary implications. A factual statement describing a defined trial result creates a different review burden from an unqualified superiority claim. A safety comparison creates a different risk profile from a convenience statement. A claim based on indirect evidence requires a different substantiation pathway from a direct randomized comparison.

The intake process should identify:

• Whether the claim is comparative, superiority-based, non-inferiority-based, or descriptive.
• Whether the comparator is a named product, treatment class, standard of care, or historical benchmark.
• Whether the endpoint is clinical, surrogate, patient-reported, or operational.
• Whether the result is primary, secondary, subgroup, or exploratory.
• Whether the claim is intended for healthcare professionals, decision-makers, or another audience.
• Whether the claim will appear in a static, digital, oral, or interactive format.

This classification allows the organization to route the claim to the correct evidence and review pathway before resources are committed to production.

Create a single source of truth

Version control is a compliance control. Your organization should maintain one authoritative record for approved comparative claims, their supporting references, mandatory qualifiers, expiry or review dates, and permitted adaptations.

Without a single source of truth, the same claim can appear with different denominators, inconsistent timeframes, or omitted limitations across channels. These discrepancies are particularly dangerous when a field team adapts centrally approved material for local use.

The record should also make clear what the evidence does not support. Negative boundaries are operationally valuable. If a study supports a difference in a defined endpoint but not an overall superiority claim, that restriction should be visible to everyone handling the material.

Test retrieval, not just approval

A compliant process must be tested under realistic conditions. Give the responsible team a claim and ask them to retrieve the evidence package, identify the relevant endpoint, explain the statistical result, and state the limitations. If this cannot be done efficiently, the process has not been operationalized.

The PMCPA 10-working-day response timeframe should be treated as a service-level requirement for evidence governance. It exposes weak document control quickly. A company that cannot retrieve the substantiation file under normal conditions will struggle when the request arrives during a product launch, field escalation, or regulatory inquiry.

A strategic mandate for compliant comparative promotion

Your organization does not need more generic review activity. It needs a controlled system that connects evidence, wording, approval, deployment, and retrieval.

Implement the following mandate:

1. Define the claim before commissioning the asset.

Specify the comparator, endpoint, population, timeframe, and level of intended conclusion.

2. Build the evidence map at the start of development.

Identify whether the support is direct, indirect, descriptive, exploratory, or confirmatory.

3. Separate statistical findings from clinical conclusions.

Do not allow a p-value or numerical direction to become a superiority statement without clinical justification.

4. Present relative and absolute risk together.

Include the baseline event rate, timeframe, and population so the audience can interpret scale rather than impression.

5. Make the medical signatory review analytical, not ceremonial.

Require explicit confirmation that the final wording remains within the evidence boundary.

6. Localize global claims by jurisdiction.

Use a common evidence core, but apply UK, U.S., and other local requirements through controlled claim layers.

7. Maintain a retrievable substantiation file.

The response process must be capable of meeting the 10-working-day PMCPA timeframe when applicable.

8. Audit the final asset as seen by the audience.

Review design hierarchy, footnotes, risk presentation, and channel adaptations—not only the approved sentence.

Comparative advertising compliance in healthcare is not achieved by attaching more references to a claim. It is achieved when the organization can demonstrate a continuous chain from evidence to wording to audience interpretation.

That is the standard your promotional review framework should enforce. Not because every comparative claim requires a head-to-head trial, but because every comparative claim requires disciplined proportionality. The message must never outrun the study, the statistic must never outrun the clinical meaning, and the approval process must never outrun the organization’s ability to defend what it published.