Optimizing Patient-Reported Outcomes for HTA Success: Lessons from German Early Benefit Assessments

Abstract

Introduction

Use and relevance of patient-reported outcomes (PROs) in health technology assessment (HTA) has increased, reflecting a shift toward patient-centered evaluation. Germany’s HTA benefit assessment process, including initial assessment by the Institute for Quality and Efficiency in Health Care (IQWiG) and final resolution by the Federal Joint Committee (G-BA), has driven this shift by prioritizing patient-relevant endpoints such as PROs. However, guidance on optimally designing and positioning PROs in clinical trials to support HTA remains limited. Using Germany as a reference, this study reviewed HTA reports to identify methodological expectations and guidance for PRO evidence generation.

Methods

A targeted review of German benefit assessments from January 2019 to August 2024 was conducted. Assessments including PROs in oncology, cardiovascular, immunology/dermatology, and neurology were reviewed. Quantitative and qualitative analyses examined HTA feedback, critiques, and recurring themes related to PRO design, conduct, and analysis.

Results

A total of 249 benefit assessments (283 datapoints) were reviewed. Common PRO critiques reflected nonadherence to IQWiG General Methods such as response (i.e., compliance) rates < 70%, failure to apply the 15% scale-span responder definition, open-label design, and statistical analysis issues. A framework linking methodological critiques to PRO rejection and bias showed that the frequency of a critique did not necessarily correspond to rejection of PRO data. PRO rejection was most strongly associated with inadequate instrument validation, low or imbalanced response rates, differing observation periods across treatment arms, and incomplete follow-up.

Conclusions

Early optimization of PRO selection, trial design, and analyses may enhance PRO acceptance in German HTA and may strengthen access in other markets. Nonacceptance of PRO results was driven primarily by methodological and operational design misalignment rather than by lack of patient relevance. While several factors can be addressed through early PRO strategy development, operational challenges, particularly related to data completeness, remain difficult to mitigate.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12325-026-03724-y.

Key Summary Points

Why carry out this study?

Patient-reported outcomes (PROs) evidence is increasingly considered in German health technology assessment (HTA) but critiques regarding PRO data can affect its acceptance in benefit assessments.

This study identified common PRO criticisms in German HTAs and evaluated how these concerns can be addressed through improved PRO design and analysis.

The findings inform best practice recommendations for generating more robust PRO evidence to support future German HTA submissions.

What has been learned from the study?

Common concerns included inadequate PRO instrument validation, response rate issues, missing data, and differences in assessment time points and follow-up duration between treatment arms, which reduced confidence in PRO results.

Many PRO-related concerns can be proactively addressed through thoughtful trial design and appropriate statistical analysis methods.

The findings provide practical guidance to optimize PRO evidence generation, improve alignment with HTA expectations, and enhance acceptance in benefit assessments.

Introduction

Health technology assessment (HTA) plays a central role in informing reimbursement, pricing, and payer coverage decisions by systematically evaluating the clinical benefit, safety, and value of new health technologies. Across Europe, HTA processes have evolved toward greater harmonization and transparency, most recently through the implementation of Joint Clinical Assessments (JCAs) under the European Union HTA Regulation (EU HTAR), aiming to align outcome expectations and methodological standards across Member States [1]. Within this evolving framework, the identification and evaluation of patient-relevant outcomes have become increasingly important.

Among patient-relevant outcomes, patient-reported outcomes (PROs) are defined as reports coming directly from patients about their health condition, without interpretation by clinicians or others [2]. PRO measures capture symptoms, functional status, tolerability, and health-related quality of life, thereby reflecting dimensions of benefit that may not be fully represented by clinical or surrogate endpoints. Regulatory authorities, including the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA), have formally recognized the value of PROs in supporting labelling claims [2, 3]. Beyond regulatory settings, PROs are increasingly incorporated into HTA to assess added benefit from the patient perspective and to inform value-based decision-making [4, 5].

Germany occupies a particularly influential position within the European HTA landscape. Introduced in 2011, the German HTA process, the Early Benefit Assessment, established under the Act on the Reform of the Market for Medicinal Products (AMNOG), defined the framework for evaluating the added benefit of new medicinal products in Germany. Benefit assessments are conducted by the Federal Joint Committee (G-BA), informed by scientific evaluations from the Institute for Quality and Efficiency in Health Care (IQWiG), and form the basis for subsequent price negotiations with the umbrella organization of statutory health insurance funds (GKV-SV) [6]. Because Germany is regarded as a significant market in Europe, HTA approaches applied there are often considered by other countries when shaping their own market access decisions.

IQWiG’s General Methods paper provides guidance for the analysis and interpretation of PRO data within benefit assessments, including recommendations on thresholds for clinically relevant differences, responder definitions, statistical analysis, and handling of missing data [7]. However, while analytical standards are well defined, limited formal guidance exists regarding the optimal inclusion, positioning, and operationalization of PROs within clinical trial design. The German HTA system uses the bias principle to evaluate the reliability of clinical data, deciding if study results are high-quality or potentially skewed by design or execution limitations. This assessment can influence the certainty of conclusion and magnitude for a drug’s added benefit, depending on the level of potential bias [7].

Empirical evidence suggests that even data generated from the use of well-validated PRO measures do not consistently influence IQWiG’s benefit ratings, particularly in oncology, where methodological limitations and implementation issues often restrict their contribution to added benefit determinations [8]. Broader European analyses similarly demonstrate variability in how PRO evidence is interpreted and weighed across HTA bodies [4, 5]. At the same time, payer perspectives indicate that well-designed and methodologically robust PRO evidence can meaningfully inform healthcare decision-making [9]. International initiatives aimed at strengthening the use and comparability of patient-reported measures further highlight the growing policy relevance of PRO data in health system performance assessment [10].

Despite this increasing emphasis, limited systematic research [4, 5] has examined how PRO design characteristics influence acceptance within HTAs. Given Germany’s position within the European HTA landscape and focus on patient relevance, the objective of this study was to systematically review publicly available German HTA benefit assessments to identify methodological factors influencing the acceptance of PRO evidence. Specifically, we aimed to (1) identify common methodological critiques associated with PRO endpoints, (2) distinguish between critiques leading to rejection of PRO data versus those increasing perceived risk of bias, and (3) derive practical considerations for optimizing PRO design and analysis planning in future clinical trials. By translating recurring HTA feedback into actionable insights, this study seeks to strengthen the impact of PRO evidence in German HTA and potentially across other markets.

Methods

Ethical Approval

This study utilized only publicly available data and did not involve human subjects or patient-level information. Therefore, ethics committee approval was not required.

Data Selection

A targeted review of publicly available benefit assessment documents conducted under the German AMNOG framework in the therapeutic areas of oncology, cardiovascular, immunology/dermatology, and neurology was performed. These therapeutic areas were selected on the basis of horizon scanning, which identified these areas as having a high volume of HTA decisions and sufficient use of PRO evidence to enable a meaningful evaluation of methodological critiques related to PROs. Although other therapeutic areas (i.e., diabetes) had a high volume of decisions, limited inclusion of PRO data precluded inclusion and reduced opportunity to evaluate HTA feedback on PRO evidence.

Screening, extraction, and analysis of benefit assessments were performed by a single, independent reviewer. Data was retrieved from official German HTA agency websites [11, 12] and the NAVLIN HTA database [13], including published dossiers, IQWiG benefit assessments, and final G-BA resolutions and justifications. Documents published in German or English were included.

Eligibility criteria were defined using the Population, Intervention, Comparator, Outcomes, and Study Design (PICOS) framework (Table 1) tailored to HTA document review to ensure transparent and reproducible case selection.

Table 1.

PICOS framework

Population (P)	Patient populations from the Act on the Reform of the Market for Medicinal Products (AMNOG) dossier submissions were included across oncology, cardiovascular, immunology/dermatology, and neurology
Intervention (I)	Medicinal products undergoing AMNOG assessment that included patient-reported outcome (PRO) evidence in the submitted dossier
Comparator (C)	Not independently defined, as comparator selection was determined within each AMNOG assessment
Outcomes (O)	Submitted PRO results and their evaluation within the AMNOG assessment, including their influence on benefit determination
Study design/setting (S)	German Early Benefit Assessments conducted and published between January 2019 and August 2024

Data Extraction

For each identified benefit assessment, data was extracted according to Table S1 in the appendix of the electronic supplementary material, including information on the medicinal product, PRO characteristics, PRO schedule of activities, PRO endpoint positioning within the trial, PRO outcomes, and the G-BA resolution.

Analytical Approach

Extracted data were aggregated for analysis. Consistent with the review-based design, all analyses were descriptive or qualitative. Categorical variables were reported as frequencies and percentages. Qualitative data was categorized on the basis of themes.

Descriptive characteristics of benefit assessments were summarized using frequency counts and percentages, including number of benefit assessments identified and included in the review, number of benefit assessments by therapeutic area, and trends in PRO implementation.

To analyze the frequency and impact of German HTA methodological critiques, qualitative G-BA/IQWiG feedback on PRO evidence was thematically categorized and anchored to IQWiG General Methods [7] (Table 2). Once categorized, methodological critiques were linked to PRO acceptability outcomes within the German HTA process: not further assessed (rejected), acceptable, acceptable with bias, or acceptable with new or alternative analyses.

Table 2.

PRO critique categories and definitions anchored to IQWiG guidance

Critique category	Applied critique definition	Relevant Institute for Quality and Efficiency in Health Care (IQWiG) general methods [7] section	Methodological anchoring
Study design
Single-arm design	Trial design with no comparator arm	3.3.2 Studies for the benefit assessment of pharmaceuticals	Benefit assessment of pharmaceuticals requires comparative trial designs to allow causal attribution of effects on patient-relevant endpoints. Single-arm studies do not provide sufficient certainty as a rule for patient-reported outcome (PRO)-based benefit claims
Open-label design	Nonblinded trial design	3.2.4 Patient-reported outcomes 10.1.4 Risk of bias	Because PROs are subjective, open-label studies have limited validity. In such studies, the size of the observed effect is a key factor when assessing the risk of bias and the resulting certainty of conclusions
Other study design issues	Inappropriate comparators, nonrandomized controls with poor comparability, or mid-study protocol amendments (e.g., endpoint changes, sample size changes)	3.2.1. Consequences of incomplete data availability 3.3.2 Studies for the benefit assessment of pharmaceuticals 10.3.13 Types of bias	Inappropriate comparator selection or reliance on nonrandomized control groups with poor comparability generally prevents answering the comparative research question underlying the benefit assessment. Mid-study protocol amendments increase the risk of systematic bias and reduce the certainty and interpretability of study results
Inadequate PRO instrument validation	PRO instrument(s) lack adequate content validity or external psychometric validation, are used outside their validated context of use, are shortened or custom, not disease-specific	3.1.1 Definition of patient-relevant endpoints 3.2.4 Patient-reported outcomes 10.3.3 Assessment of clinical relevance	To make clinically meaningful conclusions in benefit assessments, only validated and established PRO instruments are accepted
PRO methodology and implementation
Different observations across treatment arms	Difference in longitudinal observation periods or misaligned PRO assessment schedules across treatment arms	10.3.12 Handling of variable observation periods 10.3.13 Types of bias	Comparable observation periods are required across treatment arms. Otherwise, interpretability is limited as a result of systematically shortened data collection and potential detection bias. Assessment time points must be comparable between treatment arms to derive a comparative conclusion
PRO observation period shorter than other endpoints	PRO data not collected during follow-up or after disease progression (i.e., shorter duration than other endpoints)	3.2.3 Study duration 3.1.4 Endpoint-specific assessment 10.3.12 Handling of variable observation periods	To support the overall benefit conclusion, patient-reported outcomes, such as other patient-relevant endpoints, should be assessed over the entire evaluation period
PRO collection schedule issue	Changing frequency of PRO collection during study or adding/removing PROs from schedule of activities	3.2.1 Consequences of incomplete data 10.3.11 Handling of incomplete data 10.3.13 Types of bias	To avoid bias, a study needs to be analyzed as planned. Selective changes can lead to outcomes and reporting bias
Low response rates/missing data	PRO compliance is < 70% of the study participants to be included in the analysis	3.1.4 A) Reliability of study results 10.3.11 Handling of incomplete data	Missing data can lead to systematic bias if it cannot be considered completely at random. Thus, the validity of the data is questioned if the proportion of missing values exceeds the predefined tolerance threshold
Different response rates across treatment arms	The difference in compliance rates between arms exceeds 15 percentage points	10.3.11 Handling of incomplete data 10.3.13 Types of bias	Differential missingness indicates that data are not missing completely at random, leading to systematic bias. Thus, the validity of the data is questioned if the difference in missing data between study arms exceeds 15 percentage points
Statistical analysis
Statistical analysis issues	Inappropriate or poorly justified statistical methods or assumptions, mishandling of missing data, inadequate post-hoc analyses, using incorrect analysis set	10.3.1–10.3.14 Special biostatistical aspects	Statistical methods must be appropriate for the endpoint and underlying data type; otherwise, no robust conclusion can be drawn
Time-to-deterioration analysis flawed	Unclear or unsuitable definition of time-to-deterioration, mishandling of intercurrent events, or misaligned observation periods	3.1.4 Endpoint-specific assessment 10.3.12 Handling of variable observation periods	Time-to-event analyses must adequately account for intercurrent events and censoring. However, when observation periods differ substantially between treatment arms, “permanent changes” (improvements or deteriorations) cannot be meaningfully interpreted
Inadequate responder/minimally important difference (MID)a threshold	Responder criteria/MID not validated or pre-specified, or 15% scale-span responder definition not implemented	10.3.3 Assessment of clinical relevance	Responder analyses require clinically justified MID/cutoffs; otherwise, the magnitude of the effect cannot be assessed

^aMinimally important difference (MID) refers to the smallest change in a PRO that is perceived as meaningful by patients or considered clinically relevant by clinicians

With this information (categorization of critiques mapped to outcomes), a framework was developed in which methodological critiques were classified on the basis of qualitative G-BA/IQWiG assessment and author consensus as having low, medium, or high impact on (1) the certainty of PRO results (overall risk of bias), and (2) likelihood of PRO rejection in the German benefit assessment.

The critiques that resulted in the PRO evidence not being further assessed (i.e., rejected) were considered to have a high impact on rejection. Critiques that were deemed acceptable were considered to have a low impact on rejection, while those that were acceptable only with new or alternative analyses were considered to have a medium impact on rejection. A similar approach was applied when assessing the impact of methodological critiques on the risk of bias, with classifications (low, medium, or high) based on IQWiG or G-BA assessments where explicitly reported.

Illustrative Case Selection

On the basis of the analyses, two cases were selected to be illustrative of areas in which the IQWiG General Methods may allow for interpretive variation and thus benefit from further in-depth review. Case selection was not intended to be representative of the most frequently occurring critiques.

Results

Descriptive Characteristics

Overall, 654 benefit assessments published between January 2019 and August 2024 were screened for PRO inclusion, of which 249 (283 data points) met the inclusion criteria and were included in the final analysis (Fig. 1). The higher number of data points reflects that some benefit assessments included multiple PICOs, with each PICO analyzed separately. Most benefit assessments were found in oncology (76%), followed by immunology and dermatology (13%), neurology (8%), and cardiovascular diseases (3%). PROs were primarily evaluated as secondary endpoints (70%), followed by exploratory endpoints (27%) and primary endpoints (3%) in the evidence included in the benefit assessments.

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram

Overall Trends in PRO Instruments

Temporal trends in PRO instrument selection were evaluated (Fig. 2). Among assessments with PROs, the majority used a combination of generic and disease-specific instruments, accounting for 32 (64%) in 2019, 46 (88%) in 2020, 44 (70%) in 2021, 29 (74%) in 2022, 37 (77%) in 2023, and 19 (61%) January through August 2024. Use of only generic PROs was less frequent (16 [32%], 4 [8%], 13 [21%], 9 [23%], 8 [17%], and 10 [32%], respectively), while exclusive use of disease-specific PROs remained minimal across all years (2–6 assessments annually; ≤ 10%).

Fig. 2.

Temporal trends in PRO instrument submission

Frequency and Impact of Methodological Critiques

A frequency analysis was conducted to identify the most common PRO-related methodological critiques raised in the German HTA process (Fig. 3). The majority (89.4%) of benefit assessments had multiple critiques related to PRO data. The most frequently cited issue was low response (i.e., compliance) rates or missing data (35%), followed by use of inadequate responder, or minimally important difference (MID), thresholds (20%). Open-label study design was also frequently criticized (17%). Endpoint position was not found to be a PRO-related critique (0%). Approximately 10% of benefit assessments had no PRO critiques.

Fig. 3.

Frequency of PRO methodological critiques in German benefit assessments.

Note: Percentages may sum to more than 100% because some assessments included multiple critiques

The frequency of a critique did not necessarily correspond to rejection of PRO data. The impact of methodological critiques on benefit assessment outcomes was described using the predefined framework (Fig. 4). As previously noted, critiques that resulted in PROs not being further assessed were classified as leading to rejection, whereas critiques deemed acceptable with caution were considered to primarily affect the overall risk of bias, leading to a downgrading in the certainty of conclusions. Aside from single arm design (which impacts interpretability across all endpoints, not just PROs), the critiques most likely leading specifically to PRO rejection included inadequate PRO instrument validation, low response rates and missing data, > 15% difference in response rates between treatment arms, different PRO observations across treatment arms, and not submitting PRO data for the entire follow-up period. While open-label study design was frequently criticized and associated with a high perception of data bias, it did not typically result in rejection of PRO evidence in isolation.

Fig. 4.

Impact of methodological characteristics on perceived bias and rejection of PRO data

Illustrative Case Studies

Illustrative examples from German benefit assessments provide context for observed critiques, particularly in areas where IQWiG General Methods may allow for interpretive variation, such as aligning PRO assessment periods across treatment arms and longitudinally.

Case Study 1

Different PRO observations across treatment arms, such as different data collection schedules, pose a problem for data comparability. However, in some cases in which treatment schedules for the intervention and comparator greatly differ, total overlap may not be feasible within the trial design. In these instances, additional analyses or methodological adjustments have the potential to mitigate concerns and support PRO acceptance. For example, in the case of selinexor for multiple myeloma [14], differing data collection schedules between study arms resulted in an unequal number of assessment time points. By week 21, the intervention arm had four data points compared with seven in the control arm. This imbalance may have introduced bias, as time to deterioration in PRO scores, an event-driven endpoint, was observed earlier in the control arm. Given more frequent assessments in the control arm, it is uncertain whether deterioration occurred earlier or was detected sooner as a result of more frequent assessments.

The G-BA noted that analyses with aligned data collection time points across arms would have been necessary for reliable assessment of additional benefit and proposed harmonization strategies during consultations (Fig. 5).

Fig. 5.

Case study 1 visualization of PRO data collection.

= weeks = PRO assessments Omitted datapoints

Case Study 2

Differences across treatment arms in the longitudinal observation periods can induce methodological concerns within benefit assessments, an example being the time-to-deterioration analysis of the DESTINY-Breast03 study evaluating trastuzumab deruxtecan in HER2-positive breast cancer [15]. Treatment duration in the intervention arm was more than twice as long as the comparator arm, resulting in unequal PRO observation periods. In the time-to-deterioration analysis, the study sponsor defined “confirmed” deterioration as worsening of the PRO score at ≥ 2 consecutive study visits or worsening at the last assessment. As a result of the last assessment occurring sooner in the comparator arm, a “confirmed” deterioration was more likely in the comparator arm versus the intervention arm.

IQWiG/G-BA ultimately concluded that the operationalization of confirmed deterioration and inconsistent observation periods rendered the analyses unsuitable for assessing additional benefit. Instead, only time to first deterioration was used for the benefit assessment, illustrating the risk of linking the PRO follow-up-time to a clinical outcome event (such as progression as often done within oncology trials) (Fig. 6).

Fig. 6.

Case study 2 visualization of PRO data collection.

*Time to confirmed deterioration defined as PRO worsening at ≥ 2 consecutive visits OR worsening at the last available assessment. = PRO assessments = end of treatment

Discussion

This study provides a systematic evaluation of how PROs are interpreted within the German HTA process and identifies methodological characteristics that influence their acceptability.

PROs are valuable data to capture dimensions of treatment benefit that are not fully reflected by traditional clinical endpoints, including symptom burden, functional status, and health related quality of life. Within the AMNOG framework, PROs are generally recognized as patient relevant endpoints across morbidity and health related quality of life domains. However, their practical contribution to added benefit ratings depends not only on statistical significance but also on methodological credibility and interpretability within comparative assessments. Drawing on these insights collected during the review; study sponsors can be guided to optimize their PRO strategies by addressing key design elements and aligning their approaches with established HTA expectations.

Key Findings

Several important findings emerged from the analysis.

The combined use of disease-specific and generic PRO instruments has increased, while reliance on generic questionnaires alone has declined, likely because disease-specific measures capture outcomes most relevant to patients with a specific condition and are thus preferred by regulators and HTA bodies focused on clinical effectiveness.

Recurring methodological critiques were identified, most notably low response rates, insufficient validation of responder thresholds or MIDs, and lack of appropriate instrument validation. In particular, low response rates represent a recurring, industry-wide challenge [16] driven by administrative (e.g., technological failures), site-level (e.g., time constraints or missed administration), patient-level (e.g., illness severity or missed visits), disease-related (disease progression, as in late-stage oncology), and treatment-related (e.g., toxicity) factors, and mitigation efforts can be burdensome with limited effectiveness. Oncology assessments face additional challenges, as shortened PRO observation periods, often driven by prioritization of other endpoints or limited feasibility of maintaining long-term data collection, can limit the interpretability of deterioration analyses.

Importantly, while certain methodological concerns resulted in outright exclusion of PRO evidence, others increased the perceived risk of bias without fully precluding consideration. A key observation is that the frequency of critique does not necessarily correspond to its impact on acceptance. For example, an open-label study design was frequently cited as a limitation given the subjective nature of PROs, yet within the German HTA process an open-label design does not by default lead to full rejection of the PRO data but rather decreases the certainty of the evidence. In contrast, response rates below 70% or differences in response rates exceeding 15% between the treatment arms are perceived as a strong indication of underlying systematic bias (data not missing at random) leading to the PRO data not being further assessed. Likewise, the absence of adequate PRO instrument validation calls into question the reliability of the collected data in relation to the research question and thus leads to the complete rejection of the data. This distinction highlights the importance of differentiating between issues that elevate uncertainty and those that fundamentally compromise interpretability.

Moreover, several critiques categorized as having medium impact reflect remediable design or analytical limitations. Misaligned assessment schedules across treatment arms or linking PRO follow-up to patient-individual events, inconsistent operationalization of time-to-event analyses, and insufficient sensitivity analyses represent areas where methodological refinement could potentially have preserved PRO acceptability. These findings indicate that exclusion of PRO evidence is often design driven rather than a reflection of limited patient relevance.

Implications Beyond Existing IQWiG Guidance

The results are consistent with prior research showing that PROs do not uniformly translate into higher benefit ratings despite formal recognition as patient-relevant outcomes [17]. The findings are also aligned with established IQWiG methodological guidance [7] regarding clinically meaningful responder definitions, response rate thresholds, handling of missing data, and appropriate time to event analyses. However, existing methodological guidance primarily focuses on measurement validity and analytical standards; less detail is placed on operational clinical design features that determine HTA interpretability in comparative settings and address potential bias risks.

By systematically translating recurring HTA critiques into structured methodological considerations, the present findings move beyond established guidance and highlight practical design elements that enhance the utility of PRO instruments in reimbursement decision-making. Strengthening these aspects during clinical development can improve the likelihood that PRO data meaningfully contributes to added benefit determinations and supports more patient-centered healthcare evaluation.

Recommendations to Maximize HTA Success

• Plan Early and Ongoing Engagement with HTA Bodies Early interaction with HTA authorities, including advisory meetings, may help align PRO strategy with assessment expectations. Clear understanding of potential differences between regulatory and HTA research questions can support appropriate PRO selection and study design decisions.

• Demonstrate Robust Validation of PRO Instruments Avoid reliance solely on PRO validation evidence generated within the clinical trial context. Instruments supported by external validation evidence, including independent peer-reviewed publications, longitudinal validation studies, and prior HTA evaluations, are more likely to be accepted, as are instruments demonstrating strong psychometric properties across relevant patient subgroups.

• Plan PRO Data Collection to Meet HTA Expectations Key elements include aligning PRO data collection schedules across treatment arms, prespecifying assessment timepoints and stopping rules based on milestones that minimize bias potential, maintaining consistent PRO schedules for all participants, and avoiding mid-trial modifications to data collection procedures.

• Robust Statistical Methods Robust statistical planning is essential to mitigate methodological concerns. This includes addressing differences in treatment exposure, observation periods, and follow-up duration; and conducting sensitivity analyses to evaluate robustness in the presence of missing data or variable visit windows.

• Proactively Mitigate Missing Data Potential Pre-define acceptable compliance thresholds and monitor compliance during conduct of clinical trial. Plan for PRO collection in special circumstances (e.g., participant cannot attend office visit or has visual impairment). Train study team and sites on importance of compliance.

• Submit a Comprehensive Evidence Package Complete and methodologically sound data submission is critical for PRO acceptance. This includes ensuring an adequate duration of PRO data collection, monitoring and reporting changes in response rates over time, identifying and addressing patterns of missing data, avoiding truncated datasets, and submitting complete longitudinal PRO trajectories to support reliable assessment of treatment effects and patient-perceived benefit.

Limitations

This study has limitations that should be considered when interpreting the findings. Foremost, the results are primarily applicable to oncology, as most of the benefit assessments identified through the literature search were conducted in this therapeutic area. The limited representation of other therapeutic areas precluded meaningful subgroup analyses, and future research should assess whether the observed patterns are consistent across a broader range of disease areas. Nevertheless, the proposed recommendations to maximize HTA success can serve as a general framework to enhance the strength and impact of PRO evidence.

The review of benefit assessments was conducted by a single reviewer, which may have introduced subjectivity in data extraction and interpretation. In addition, the study did not evaluate temporal trends in methodological critiques or evidentiary expectations, representing an opportunity for future research as HTA frameworks continue to evolve. The analysis focused on critiques of PRO evidence and did not assess the appropriateness or selection of the PRO measures themselves; future studies could explore how PRO instrument selection influences HTA outcomes. Finally, the mapping of identified critiques to potential bias and the likelihood of evidence rejection was based on expert author consensus and may therefore be subject to interpretation.

Conclusion

Earlier integration of methodological best practices into clinical development and continued refinement of PRO design frameworks can improve the practical utility of PRO instruments in comparative benefit assessments.

Exclusion of PROs in the benefit assessment often reflects suboptimal study design, limited validation, or analytical shortcomings rather than a lack of patient relevance. Careful trial planning, including validated instruments, aligned assessment schedules, consistent data collection across arms, and robust statistical considerations, increases the likelihood that PRO evidence informs benefit decisions. However, certain challenges, such as maintaining high response rates over extended follow-up periods, may persist despite well-designed studies. Strengthening operational and analytical rigor enhances the impact of PRO data while acknowledging that not all limitations can be fully mitigated through study design alone.

As European HTA processes evolve under the EU HTAR, greater alignment in expectations for PRO operationalization may further strengthen their role in decision-making. Future research should assess cross-country patterns in PRO evaluation and examine whether methodological harmonization improves consistency in reimbursement and resource allocation decisions.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

All authors made substantial contributions to the study conception and design, data acquisition, analysis, or interpretation of the data. Shivani Shah drafted the manuscript, and all other authors revised it critically for important intellectual content. All authors approved the final version of the manuscript and are accountable for all aspects of work. Kristi A. Bertzos: Conceptualization, methodology, interpretation of data, writing, review and editing. Nicole Alt: Conceptualization, methodology, interpretation of data, writing, review and editing. Peter C. Coyle: Conceptualization, methodology, interpretation of data, writing, review and editing. Katherine S. Gries: Conceptualization, methodology, interpretation of data, writing, review and editing, supervision. Anke K. Kummer: Conceptualization, methodology, interpretation of data, writing, review and editing. Anna Lieb: Conceptualization, methodology, interpretation of data, writing, review and editing. Sophie Thiemann: Conceptualization, methodology, interpretation of data, writing, review and editing. Daniel Wirth: Conceptualization, methodology, interpretation of data, writing, review and editing. Shivani K. Shah: Formal analysis, interpretation of data, writing, original draft, writing, review and editing. Markus J. Frosien: Conceptualization, methodology, interpretation of data, writing, review and editing, supervision.

Funding

This research was funded by Johnson & Johnson. The sponsor provided funding for the study, manuscript development, and the Rapid Service Fee/Open Access publication fee. Johnson & Johnson also participated in the study design, interpretation of the findings, review of the manuscript, and the decision to submit the manuscript for publication

Data Availability

Johnson & Johnson’s data sharing policy is available at https://www.janssen.com/clinical-trials/transparency. As noted on this site, requests for study data access can be submitted through the Yale Open Data Access (YODA) Project site at http://yoda.yale.edu.

Declarations

Conflict of Interest

Kristi Bertzos, Nicole Alt, Peter Coyle, Katharine Gries, Anke Kummer, Anna Lieb, Sophie Thiemann, Daniel Wirth, and Markus Frosien are employees of Johnson & Johnson and may hold stock in the company. Shivani Shah is an employee of Eversana, which received funds from Johnson & Johnson to conduct the research described in this manuscript.

Ethical Approval

This study utilized only publicly available data and did not involve human subjects or patient-level information. Therefore, ethics committee approval was not required.