European Medicines Agency encourages regulatory integration of patient experience data

The European Medicines Agency (EMA) has published a reflection paper establishing principles for the systematic incorporation of patient experience data (PED) throughout medicine development and regulatory assessment. The initiative represents a significant step towards embedding the patient voice across all stages of a medicine's lifecycle from early research through to post-marketing surveillance.

Context and purpose

The EMA paper seeks to address a fundamental gap in current medicines development practice. While the EMA has long acknowledged that PED can contribute meaningfully to regulatory assessment, such data have not been systematically integrated into all aspects of medicine development programmes, marketing authorisation applications or post-marketing safety monitoring.

The agency recognises that patients may value different aspects of their disease and available treatments than developers typically consider, including particular outcome measures such as quality of life, specific populations or disease stages requiring study and acceptable risk tolerability.

This policy development follows a 2022 multistakeholder workshop on PED, which established consensus on the importance of including this data at all stages of medicines development and regulatory decision making. Although some EU guidelines exist, these are described as either fragmented or outdated, creating uncertainties for developers.

The paper's stated purpose is to encourage systematic consideration of PED in medicine development programmes and regulatory submissions. It describes general principles for using this data across various phases and identifies the types and main sources of data available. 

Integration with EU frameworks

The EMA situates PED within the existing architecture of EU medicines regulation – which is currently subject to comprehensive reform under the EU Pharmaceutical Package – rather than introducing wholly new requirements. The EMA's approach defines PED as information that directly reflects the experience of a patient or carer, collected without input or interpretation by healthcare professionals, third parties or artificial intelligence (AI)-based devices.

PED can be collected using quantitative methods such as surveys exploring clinical outcomes or health-related quality of life instruments, qualitative methods including interviews and focus groups, or mixed approaches integrating both methodologies.

Developers are encouraged to engage early with regulators through scientific advice to discuss optimal approaches for generating and collecting PED as development plans become concrete.

Joint scientific advice with health technology assessment (HTA) bodies is also encouraged, as the preferred route to ensure alignment with downstream decision makers and inclusion of PED pertinent for post-launch evidence generation, relative effectiveness assessments and economic evaluations.

The paper also highlights the qualification procedure for novel methodologies as a route for assessing and endorsing innovative methods for collecting and using PED. Once the Committee for Medicinal Products for Human Use issues a qualification opinion confirming that a proposed method is suitable for use in a defined regulatory evidence generation context, the opinion is published and available for multiple stakeholders to use. Examples of methods applicable to data qualifying for scientific advice and qualification procedures include clinical outcome assessments, patient preference studies, symptom scales and spontaneously generated online PED.

Digital health and AI 

The growing role of digital technologies in capturing PED is acknowledged in the paper, though this receives relatively brief treatment within the broader framework. The document recognises digital and AI-powered methods as examples of PED collection approaches that can undergo scientific advice and qualification procedures, an approach that intersects with emerging EU AI Act requirements for high-risk AI systems in healthcare. Specifically, electronic patient-reported outcomes and computerised adaptive test questionnaires are identified as relevant digital methodologies.

The paper also references spontaneously generated online PED, noting that posts written by individuals on social media platforms represent one example of such data in unstructured form. There is acknowledged increasing interest among biomedical researchers in developing methods to analyse large volumes of unstructured data and generate knowledge from these sources, including through AI-powered analytical tools which the EMA has recently addressed in separate guidance.

More broadly, the reflection paper discusses patient-generated digital data as one source of PED, though detailed guidance on validation requirements for digital health technologies and AI devices capturing such data remains outside the paper's scope.

For developers of digital health technologies and AI-driven devices intended to capture PED for medicine regulatory purposes, the reflection paper signals that such tools must meet the same quality and validation standards as conventional methods, with early regulatory dialogue, scientific advice or qualification procedures providing appropriate mechanisms for discussing digital methodologies.

Types and sources of patient data

The reflection paper distinguishes PED based on whether they report health outcomes or express patient preferences on treatment trade-offs and whether data are quantitative or qualitative in nature. Three principal types are identified: patient-reported outcomes, patient preference studies and data obtained through patient engagement activities.

Patient-reported outcomes are health outcomes directly reporting the patient's experience of their health status without amendment or interpretation by clinicians or other parties. These are normally collected through patient-reported outcome measures such as questionnaires and surveys evaluating the impact of a health condition, treatment or intervention at single time points or over time.

Patient preference studies, though not yet fully established nor systematically integrated in drug development, can complement evidence from pivotal clinical trials. These studies include any qualitative or quantitative assessment of the relative desirability or acceptability to patients of aspects differing among alternative health interventions. They can help characterise medical need, select endpoints, estimate meaningful effect size and identify subgroups with different preferences.

Data obtained through patient engagement activities represent a distinct category defined as interactions with patients to gather their experience with disease and preferences regarding treatments and outcomes. Various methodologies can be used to seek patient input, including surveys, interviews, written consultations, stakeholder meetings, workshops, scientific advice procedures, scientific advisory groups, committee consultations and public hearings. 

Regarding sources, PED are most commonly collected in clinical trials, particularly phase III studies, to support decision making in regulatory settings, HTA and reimbursement decisions and clinical care. Notably, real-world data represent a significant source, with PED collection in real-world clinical care settings providing added value in informing patients' healthcare needs and preferences and increasing knowledge on benefit-risk profiles. To support regulatory assessment, data collected outside clinical trials must meet quality standards equivalent to trial-based PED. This can occur through primary data collection following pre-specified research protocols in non-interventional studies, or through extraction from existing healthcare data sources such as patient registries.

Fitness for purpose

Concrete requirements to ensure PED is considered fit for regulatory purpose are set out in detail by the EMA, which makes explicit reference to the following criteria:

• Data quality. PED must meet the criteria laid out in the EU Data Quality Framework and be appropriately validated and analysed using fit-for-purpose methods.
• Representativeness. PED must reflect the experience of the target patient group and be responsive to population characteristics like age, gender and vulnerability.
• Study design. Collection of PED should correspond to the intended use, stage of development and clinical context and may require randomised, non-interventional, retrospective or real-world evidence (RWE) designs.
• Data collection methods and tools. Tools must be psychometrically validated and suit the language, health literacy and cultural context of the target population.
• Participant burden. PED collection should be balanced against the potential for overburdening participants, risking data quality and trial retention.
• Training and capacity. Sponsors and researchers should ensure personnel are adequately trained in PED methodology and interpretation.
• Transparency. Clear disclosure of the use of PED in submissions and assessment processes is essential, both for confidence in regulatory decision making and for stimulating further patient engagement.

Osborne Clarke comment

The EMA’s reflection paper on patient experience data represents an evolution rather than a departure in EU regulatory attitudes to patient experience data. The agency’s ambition to "encourage systematic consideration of PED" as an evidence source acknowledges the inherent value in understanding the patient’s journey with both disease and treatment.

For developers, pharmaceutical operators and digital innovators, the key message is to move early: robust, scientifically sound PED collection – thoughtfully aligned with clinical development plans and regulatory strategies – is likely to be viewed most favourably by both regulators and downstream HTA assessors. The paper also suggests that joint scientific advice, parallel consultation and EMA qualification procedures for novel tech not only reduce risk but may generate regulatory confidence in innovative methodologies, especially for digital health and RWE approaches.

The public consultation on the reflection paper is set to run until 31 January 2026, giving stakeholders have a valuable window to engage with regulators, highlight practical and methodological challenges and establish where additional guidance is needed. This consultation aligns with the broader EU life sciences strategy aimed at strengthening innovation and competitiveness in pharmaceutical research. The anticipated international and EU-level guidance on patient preference studies and digital PED further underscores the direction of travel.

The paper signals that thorough and patient-centred PED integration stands to become an important competitive advantage for market access in the EU. Effective early engagement, methodological rigour, multidisciplinary partnership and focus on patient perspectives will likely shape successful medicines development in the coming years.