Patient Preference Studies: What They Are and Why Sponsors Can No Longer Ignore Them

For decades, the dominant model of drug and device development has been straightforward: demonstrate safety and efficacy through well-controlled clinical studies, submit the evidence to regulators, and obtain approval. Patient input, to the extent it existed at all, was largely informal — a background consideration rather than a structured data source.

That model is changing. The November 2025 release of ICH E22, the draft international harmonised guideline on Patient Preference Studies (PPS), is among the clearest signals yet that regulators worldwide expect sponsors to systematically measure, document, and submit evidence about what patients actually value. For medical device and drug sponsors who have not yet integrated patient preference methods into their development programs, the time to start is now — not at the end of a development program, but at the beginning.

What Patient Preference Studies Actually Measure

A patient preference study is not a satisfaction survey. It is not a focus group about brand perception. Done properly, it is a rigorous empirical study designed to quantify the relative importance patients assign to specific attributes of a treatment — and the trade-offs they are willing to make among those attributes.

The ICH E22 guideline focuses specifically on stated-preference methods: structured surveys or interviews in which participants are asked to express their choices or acceptable thresholds for trade-offs across defined treatment scenarios. This is distinct from revealed-preference methods, which infer preferences from observed real-world behavior. Stated-preference methods are more directly applicable to regulatory contexts because they can be designed prospectively, tied to specific endpoints, and analyzed with pre-specified statistical methods.

The outputs of a well-designed quantitative PPS are genuinely informative. They can tell you, with numerical precision, how much additional efficacy patients in a specific disease population require before they are willing to accept a given incremental risk. They can tell you whether patients view two components of a composite endpoint as equally important or whether one dominates the other. They can characterize preference heterogeneity — the distribution of trade-off thresholds within a population — which matters enormously when a single trial result masks meaningfully different patient subgroups.

These are not soft findings. They are structured, analyzable data that can be submitted to regulatory authorities as part of a benefit-risk dossier.

Why Regulators Are Paying Attention

The ICH E22 guideline did not emerge in a vacuum. It reflects a sustained and deliberate effort across FDA, EMA, and other global regulators to formalize the role of patient input in regulatory decision-making.

FDA has been explicit about this direction for years. The 21st Century Cures Act mandated that FDA develop methods for systematically collecting patient experience data. The FDA's benefit-risk framework, referenced directly in the ICH E22 guideline through ICH M4E(R2), already contemplates the use of patient preference data in the Clinical Overview and Benefits and Risks Conclusions of a regulatory submission. The E22 guideline operationalizes how that data should be generated.

The practical implication is this: a benefit-risk assessment that relies solely on sponsor-defined clinical endpoints, without any structured evidence about whether patients consider those endpoints meaningful or how they weigh them against risks, is increasingly seen as incomplete. It answers the question regulators asked twenty years ago. It does not fully answer the question they are asking today.

The Strategic Case for Early Integration

The most important message in ICH E22 is also the one most likely to be overlooked: patient preference studies need to be planned early. The guideline is explicit that the timing of a PPS is typically influenced by the objective of the study, when sufficient information is available to design it appropriately, and when the results are needed.

This is not a documentation formality. It reflects a fundamental truth about how PPS data is used. A preference study designed to inform endpoint selection must be completed before the pivotal trial is designed. A study designed to characterize benefit-risk trade-offs must be aligned with the efficacy and safety data expected from clinical studies — which means the attributes and their levels must be calibrated against the anticipated range of clinical outcomes.

Sponsors who wait until late-stage development to commission a patient preference study often find themselves with data that cannot be used for its intended purpose. The attribute levels in the PPS do not cover the actual outcomes observed in the trial. The patient population studied in the PPS does not match the enrolled trial population. The analysis plan was not pre-specified. These are not minor methodological issues — they are the kinds of limitations that FDA reviewers will identify and that can undermine the evidentiary value of the entire exercise.

The regulatory and competitive advantages of early PPS integration are real. At early development stages, PPS can identify which efficacy outcomes matter most to the target population — information that can sharpen endpoint selection and reduce the risk of designing a trial around a measure patients do not find meaningful. It can reveal whether patients would consider a proposed treatment's benefit-risk profile acceptable even before pivotal data exist, providing an early signal about commercial and regulatory viability. And it can inform protocol design in ways that improve recruitment and retention, which has direct implications for trial timelines and costs.

What ICH E22 Requires of Sponsors

The guideline lays out a clear methodological framework for PPS, and sponsors should understand its core requirements.

A pre-specified protocol and analysis plan are non-negotiable. As with clinical studies, ICH E22 expects that the research question, study design, sample definition, attribute selection, and statistical analysis plan are documented before data collection begins. The guideline even encourages pre-registration of protocols to enhance credibility and transparency. Sponsors who treat PPS as an informal qualitative exercise — generating data and then deciding how to analyze and present it — will not produce evidence that survives regulatory scrutiny.

Qualitative work must precede quantitative work. The guideline is structured around a progression: qualitative research to identify and refine the attributes that matter to patients, followed by quantitative research to measure the relative importance of those attributes. Skipping the qualitative phase, or conducting it superficially, undermines the validity of the quantitative instrument. Regulators will ask how attributes were selected and whether patients were meaningfully involved in that process.

The study sample must match the intended use population. A mismatch between the PPS sample and the patient population of the regulatory submission limits the generalizability of the findings. The guideline is direct on this point: if a different population was studied, the sponsor must justify its relevance. Demographic diversity, disease stage, and prior treatment experience are all characteristics that can affect preferences and must be considered in sampling design.

Data quality checks must be built in from the start. The guideline identifies a specific set of quality concerns unique to preference surveys — attribute non-attendance, illogical responses, fraudulent responses, and response inconsistencies — and expects sponsors to address them through pre-specified instrument design and analysis approaches. These are not afterthoughts. They must be woven into the survey design and reflected in the analysis plan.

Submission placement is defined. PPS reports belong in CTD Module 5.3.5.4 as other clinical study reports, and are referenced in Module 2, including in the Product Development Rationale and, when the data support benefit-risk conclusions, in the Benefits and Risks section. Sponsors who generate PPS data but do not know where or how to integrate it into a submission are leaving value on the table.

The Multidisciplinary Requirement

One aspect of ICH E22 that deserves particular attention is its emphasis on multidisciplinary expertise. The guideline states clearly that PPS design, conduct, analysis, and submission should be undertaken by a cross-functional team with both PPS methodology expertise and clinical expertise.

This matters because patient preference research sits at an intersection that few individual disciplines fully occupy. Health economists bring knowledge of discrete choice methods and preference elicitation theory. Biostatisticians bring expertise in pre-specified analysis, sample size determination, and sensitivity analysis. Clinicians bring understanding of the disease context and the clinical meaningfulness of attributes and levels. Regulatory specialists understand how evidence needs to be packaged and positioned within a submission.

Getting this right requires genuine collaboration among all of these functions — not a handoff from one group to another. For most sponsors, particularly smaller medical device companies, this means either building that expertise internally or engaging partners who already have it.

A Signal Worth Taking Seriously

ICH E22 is a draft guideline, currently under public consultation. It will be revised before finalization. But the direction it represents is not in question. Regulators across the United States, European Union, Japan, and other ICH regions have already endorsed its core principles. The expectation that sponsors will generate structured, rigorous evidence about patient preferences — and submit that evidence as part of their regulatory dossiers — is becoming a standard part of the development landscape, not an optional enhancement.

For sponsors who have not yet built patient preference capabilities into their development programs, the question is not whether to start. It is how quickly they can build the infrastructure — methodological, operational, and regulatory — to do it well.

Biomedical Statistical Consulting LLC (BSC®) has supported FDA medical device and drug submissions since 1986, with deep expertise in clinical trial design, benefit-risk assessment, and regulatory biostatistics. For questions about integrating patient preference studies into your development program, contact us at biomedstat.com.