Expanding the definition of unmet medical need: Findings from a multistakeholder workshop

Abstract

BACKGROUND:

Unmet medical need (UMN) is frequently referenced in health policy and health care decision-making, including the Centers for Medicare & Medicaid Services’ (CMS) Drug Price Negotiation Program (DPNP). However, the concept of UMN is often defined inconsistently and has a narrow focus, primarily centered on clinical effectiveness and the availability of treatment. This approach may overlook the wider needs of patients, caregivers, and society as a whole. A broader conceptualization of UMN has been proposed, yet there remains a lack of an actionable implementation framework.

OBJECTIVE:

To examine how stakeholders interpret and distinguish among UMN elements, prioritize these elements based on perceived importance, and elicit perspectives on the applicability of a prioritized UMN framework in a decision-making context.

METHODS:

A multistakeholder workshop was conducted in June 2025 involving 17 participants representing researchers, industry, payers, policymakers, and a patient organization. Participants evaluated 38 UMN elements grouped into 7 domains derived from a prior structured literature review. Using an interactive polling platform, participants assessed the conceptual distinctness of elements within domains, ranked domains by importance, and completed a pairwise weighting exercise grounded in the Simple Multi-Attribute Rating Technique. Quantitative voting results were aggregated, and qualitative feedback from plenary discussions was recorded and thematically synthesized.

RESULTS:

Participants confirmed the conceptual distinctness of the proposed domains and elements. In the weighting exercise, traditional clinical effectiveness received the highest importance (24%), followed by availability of other treatments (19%) and impacts on patient, caregiver, or family quality of life (16%), although differences across domains were modest and should be interpreted with caution, given the exploratory sample. Economic burden on patients and families (13%), societal perspective elements (11%), and economic burden on society (10%) received moderate weights, whereas treatment administration attributes received the lowest weight (7%). Participants agreed that the broader element set captures dimensions of UMN not addressed by the current framework under CMS’s DPNP. Key qualitative themes included perceived overlap within elements of the quality-of-life domain, the need for clearer conceptual boundaries between UMN and comparative effectiveness, and feasibility concerns around implementation.

CONCLUSIONS:

UMN is inherently multidimensional and extends beyond traditional clinical measures. These exploratory findings suggest that structured prioritization of UMN domains may support more transparent and policy-relevant frameworks for evaluating unmet need in the United States, including within CMS’s DPNP.

Plain language summary

Unmet medical need refers to situations where current treatments fail to meet patients’ needs. Policy definitions often focus too narrowly on clinical outcomes, neglecting what matters most to patients and society. To address this, we held a multistakeholder workshop to evaluate 38 concepts addressing unmet needs, organized into 7 domains. Participants affirmed these domains’ relevance and ranked clinical effectiveness and treatment availability as top priorities, while also acknowledging the importance of quality-of-life and economic factors. Understanding which dimensions of unmet need matter most has direct implications for patients and society, as a broader and more structured definition can help direct health care innovation and policy attention toward the areas where patients and caregivers experience the greatest burden.

Implications for managed care pharmacy

This study demonstrates that unmet medical need is multidimensional and extends well beyond traditional clinical criteria. By integrating qualitative stakeholder feedback with a structured quantitative prioritization exercise, we establish an initial, exploratory framework that decision-makers can build upon. Managed care pharmacists and payers should consider incorporating broader patient-centered, economic, and societal dimensions into frameworks for evaluating unmet need, thereby supporting more transparent coverage decisions and steering innovation toward areas of greatest need and value.

Unmet medical need (UMN) is a key concept for guiding the focus of health care innovation and policy attention. It is embedded in regulatory incentives, such as the US Food and Drug Administration’s (FDA) expedited approval pathways,¹ and has also been referenced in recent CMS programs, including the Drug Price Negotiation Program (DPNP),^2,3 where UMN is a statutory factor in the negotiation process. Despite its importance, UMN is rarely defined in a clear, consistent, and actionable way.^4,5 Existing definitions are often narrow, centered on clinical effectiveness and the availability of treatment alternatives, which may not fully reflect what matters to patients, caregivers, or society. It is also important to distinguish UMN from related but distinct constructs such as comparative effectiveness or broader value assessment.^6,7 This study focuses specifically on the dimensions of need that a disease and its available treatments leave unaddressed, rather than the relative performance of one therapy vs another.

A recent analysis highlighted a broader conceptualization of UMN, identifying 40 elements grouped into 7 domains that extend beyond traditional clinical measures.⁶ These include impacts on quality of life for patients, caregivers, and families; economic burden at the household and societal levels; societal considerations such as rarity, equity, and spillovers; and treatment administration barriers influencing real-world use. This broader framing suggests that a narrow definition risks overlooking meaningful dimensions of UMN and may limit the usefulness of UMN in decision-making contexts.

However, broadening UMN creates its own challenge: a long list of UMN elements is not actionable unless the elements are prioritized. When many elements are considered simultaneously, there is a risk of treating them as equally important, making it difficult for assessors and innovators to understand which aspects of UMN matter most and how well a therapy addresses them. In addition, UMN elements may be interpreted differently across various stakeholder groups, including patients, caregivers, clinicians, payers, and researchers, raising questions about how these elements should be understood, how distinct they are, and how they should be weighed in practice. These interpretive challenges underscore the need for a structured approach to prioritization.

A further challenge is the lack of clarity on how a prioritized UMN framework could be incorporated into existing decision-making structures. The CMS’s DPNP provides an example of where a prioritized understanding of UMN may enhance decision-making. CMS currently relies largely on FDA guidance, which centers on traditional clinical outcomes and does not distinguish the relative importance of broader patient, caregiver, or societal elements.^1,6 This is particularly limiting given that CMS has stated it will consider “outcomes such as changes to productivity, independence, and quality of life” and “considerations related to patient experience” elsewhere in its program guidance,⁶ dimensions largely absent from the FDA’s UMN definition. For UMN to inform the DPNP in a consistent and meaningful way, clarity on what UMN dimensions to include and, importantly, how they should be prioritized could help CMS focus its assessments, improve transparency, and signal priorities for evidence generation. Such prioritization may also offer insights for other assessment frameworks that consider UMN.

Building on the element framework established by DiStefano et al,⁶ which identified 40 UMN elements but did not address their relative importance, the aim of this study is to address these gaps by (1) examining how stakeholders interpret and distinguish among elements that may define UMN, (2) prioritizing these elements based on their perceived importance, and (3) eliciting stakeholder perspectives on the prioritization results and on the usefulness and potential applicability of a prioritized UMN framework in decision-making contexts. These elements were originally identified through a structured review of the literature and extend beyond the conventional criteria referenced by CMS via FDA guidance, while remaining focused on aspects that medicines can directly influence. The findings aim to support the development of more comprehensive and transparent assessment frameworks in the United States, including potential applications within the CMS’s DPNP.

Methods

DEVELOPMENT OF UMN ELEMENTS AND DOMAINS

The elements and domains of UMN used in this study were based on a literature review by DiStefano et al,⁶ which identified 40 distinct elements organized into 7 domains reflecting patient, caregiver, and societal perspectives.

For the workshop, the list was refined for clarity, removing 2 overlapping elements from the “Treatment Administration” domain: overall complexity of treatment and convenience of administration. They were judged to be largely captured by other elements already included in the domain, such as length of treatment regimen, medication frequency, and route of administration. Their removal was not considered to alter the conceptual completeness of the domain, as the remaining 9 elements collectively cover the key dimensions of treatment administration relevant to patient decision-making. This refinement was intended to maintain coverage of the original framework while improving clarity and response feasibility for participants. The rationale for the overall domain and element structure is described in detail in DiStefano et al,⁶ and the conceptual distinctness of the retained elements was assessed empirically as part of the workshop exercises reported in the section “Stakeholder Assessment of Distinctness and Scope of UMN Elements.” The final set contained 38 elements grouped into 7 domains: (1) Traditional clinical effectiveness; (2) Impacts on patient, caregiver, and family quality of life; (3) Economic burden on patient, caregiver, and family; (4) Economic burden on society; (5) Societal priorities; (6) Attributes of treatment administration; and (7) Availability of other treatments.

Definitions provided to participants distinguished between elements and domains. Elements were defined as specific concepts representing needs that could be addressed through health care interventions and should be considered in decision-making (eg, psychological/emotional health, activities of daily living). Domains were broader categories including multiple UMN elements associated with a common overarching concept (eg, quality of life, clinical effectiveness).

WORKSHOP DESIGN AND PARTICIPANTS

The multistakeholder prioritization workshop was conducted in person in June 2025 and lasted approximately 5 hours. Participants were purposively selected to ensure representation of key decision-making perspectives relevant to UMN assessment in the United States, including policy experts, payers, industry representatives, researchers, and patient organization members. No formal expertise thresholds were applied. Instead, invitations targeted individuals with recognized professional experience in health policy, payer decision-making, pharmaceutical industry, or health outcomes research. For the research group specifically, we invited experts in bioethics and humanities, clinical research, research on payer and insurance decision-making, value assessment research, and patient preferences research. Invitations were issued through professional networks and relevant organizations to represent a range of perspectives. All participants received the same prereading materials (Supplementary Materials ^{(438.2KB, pdf)} , available in online article) summarizing the 7 UMN domains and their 38 elements. The session was moderated by an experienced health-policy facilitator, who presented questions neutrally and avoided expressing opinions on substantive issues. To manage cognitive burden, the workshop was structured to present one domain at a time. Before each voting exercise, the facilitator led a brief discussion of the domain and its constituent elements to ensure a shared understanding of each concept. Group results were displayed after each exercise to support reflection but were not discussed by the facilitator prior to subsequent voting rounds to minimize anchoring effects. Quantitative data were collected via Mentimeter (a polling platform), and qualitative observations through researcher notes. This study involved professional stakeholders providing input on policy-relevant concepts and did not include personal or patient-level data. It was therefore classified as non–human subjects research and exempt from formal institutional review board (COMIRB #25-1021) oversight.

VOTING PROCESS

The ranking and weighting process followed the preference-elicitation approach, which systematically gathers and quantifies stakeholder judgments about the relative importance of competing criteria, as described by McQueen et al⁷ for broader contextual value elements, adapted here to the UMN framework. The structured, group-based elicitation was implemented through Mentimeter, an interactive polling platform that enables anonymous ranking and weighting. This process is grounded in the Simple Multi-Attribute Rating Technique,^8,9 a method for multicriteria decision analysis.

Participants completed a sequence of interactive exercises, as summarized in Figure 1. The session began with concept differentiation, where each element of unmet need was evaluated using two 1 to 5 Likert scales. The first scale assessed how distinct each element was from others in the same category, whereas the second scale determined whether the element addressed aspects not encompassed by the FDA’s definition of unmet need and, consequently, CMS’s definition as well. The first scale informed whether the domain structure was sufficiently granular to proceed with prioritization, whereas the second established the added value of the broader framework relative to current CMS practice. Following this, participants ranked 7 domains based on their perceived importance in representing UMN. This ranking was performed in iterative rounds: participants initially selected their 3 highest-priority domains, which were then removed before the exercise was repeated. The final rank order was determined by the sequence in which domains were eliminated across rounds, with domains selected in the first round ranked highest. This iterative approach was chosen over a single full ranking to reduce cognitive burden and encourage more deliberate prioritization. Next, participants conducted a pairwise comparison of adjacent domains to establish relative trade-off weights. Weights were assigned individually and anonymously through Mentimeter, and the final weights reflect the average across all participants. Only adjacent domains were compared, as the prior ranking exercise provided an ordered structure that allowed relative importance to be captured through sequential comparisons without requiring all possible pairwise combinations. Using a continuous scale from 0% (indicating equal importance) to +50% (indicating the higher-ranked domain was considered 1.5 times as important as the adjacent lower-ranked domain), participants indicated how much more important one domain was compared with the next. The average difference score for each adjacent pair was then used to derive relative weights using the following formula: each domain’s weight = (average difference score/sum of all average difference scores) × 100, normalizing weights to sum to 100%. Domains were compared sequentially in the order established by the prior ranking exercise, resulting in 6 adjacent pairwise comparisons. The average difference score for each pair was then used to derive relative weights, which were subsequently normalized to sum to 100%. Individual responses were averaged across participants prior to normalization. Disagreement across participants is reflected in the variability of the resulting weights, reported in the section “Prioritization and Weighting of UMN Domains.”

FIGURE 1.

Workshop Voting Exercises

CMS = Centers for Medicare & Medicaid Services; FDA = US Food and Drug Administration; UMN = unmet medical need.

Group results were displayed after each exercise to support reflection but not influence subsequent voting. This approach, combining ranking and weighting, allows transparent aggregation of stakeholder preferences and provides a quantitative hierarchy that reflects the collective prioritization of UMN elements and domains.

QUALITATIVE FEEDBACK COLLECTION

Discussions were not audio-recorded. Two researchers independently recorded participant comments during plenary discussions and the breakout synthesis phase. Notes were compared and reconciled immediately after the session to ensure accuracy. A thematic synthesis was performed using a primarily inductive approach, grouping feedback under emergent themes (eg, conceptual clarity, applicability for payers, perceived missing domains). No formal coding framework or software was used, and themes emerged organically from the recorded notes. Differences in interpretation between the 2 researchers were resolved through discussion until agreement was reached. No formal inter-rater reliability statistic was calculated, which we acknowledge as a limitation of the qualitative approach.

Results

A total of 17 participants contributed responses in the multistakeholder workshop, including researchers (n = 9), industry representatives (n = 4), payers (n = 2), a patient organization representative, and a policymaker, ensuring that perspectives from key sectors involved in health care decision-making were represented. Participants reviewed and validated 7 overarching domains of UMN encompassing 38 distinct elements (Table 1).

TABLE 1.

Overview of Domains and Associated Elements Evaluated in the Workshop

Domain	Number of elements	Elements
Traditional clinical effectiveness	5	Safety/adverse events; time to event; impact on key clinical/biological parameters; symptoms; durability of treatment effect
Impacts on patient, caregiver, or family quality of life	10	Physical health; psychological/emotional health; cognitive functioning; social relationships; activities of daily living; aesthetic/cosmetic; independence; reproduction; career; education
Economic burden on patient, caregiver, or family	2	Impact on income; financial burden of disease (medical and nonmedical costs)
Economic burden on society	3	Impacts on productivity; cost to payers; nonfinancial resource demands
Societal perspective elements	6	Insurance value; fear of contagion; scientific spillover; disease rarity; disease severity; impact of disease or treatment on disadvantaged or vulnerable target populations
Elements of treatment administration	9	Length of treatment regimen; medication frequency; route of administration; needle/tablet size; polypharmacy need; medicine storage requirements; drug-drug interactions; uncertainty of treatment effectiveness and safety; interaction with religion/culture
Availability of other treatments	3	No treatment available; lack of ideal (eg, curative) treatment; number of treatments available

STAKEHOLDER ASSESSMENT OF DISTINCTNESS AND SCOPE OF UMN ELEMENTS

The subsequent analysis assessed how distinct or overlapping participants perceived these individual elements to be within their respective domains (Figure 2).

FIGURE 2.

Stakeholder Ratings of Uniqueness of Elements Within UMN Domains

Participants were asked: “To what extent does each unmet need element within the domain describe a different concept of need?” Ratings ranged from 1 = not at all (overlapping) to 5 = very much (unique). During the workshop discussion, participants interpreted ratings of 3 or higher as indicating that the elements were sufficiently distinct to proceed with the prioritization exercise. Boxplots display the median, IQR, and overall distribution of responses; orange points represent mean ratings. Exact summary statistics (N, median, and IQR) for each domain are reported in the text.

UMN = unmet medical need.

Overall, most domains received moderate to high ratings (≥3), which participants interpreted during discussion as sufficient distinctness to proceed with prioritization, indicating general agreement that their constituent elements represented distinct and nonoverlapping aspects of UMN. Summary statistics by domain were as follows: Traditional clinical effectiveness (N = 17, median = 4.0, IQR = 1.0); Impacts on patient, caregiver, or family quality of life (N = 17, median = 3.0, IQR = 0.0); Economic burden on patient, caregiver, or family (N = 17, median = 5.0, IQR = 1.0); Economic burden on society (N = 16, median = 4.0, IQR = 1.0); Societal perspective elements (N = 15, median = 5.0, IQR = 0.5); Elements of treatment administration (N = 16, median = 4.5, IQR = 1.5); and Availability of other treatments (N = 16, median = 4.0, IQR = 1.5). The quality-of-life domain received the lowest median rating (3), and responses were tightly clustered around this value (resulting in a collapsed boxplot because the IQR was zero), indicating that participants were largely neutral regarding the distinctness of its elements and perceived some conceptual overlap among them. In contrast, societal perspective elements received the highest ratings with relatively low dispersion, suggesting greater agreement that these elements represent distinct dimensions of UMN. Treatment administration showed the greatest variability in responses, indicating less consensus among participants regarding the conceptual distinctness of its elements. Overall, participants generally supported all domains as conceptually distinct, with minor variation in perceived clarity across domains.

Narrative Feedback on the Distinctness of UMN Domains and Elements

Participants discussed both relationships across domains and potential overlap among elements within individual domains. At the domain level, stakeholders generally agreed that the 7 domains captured distinct dimensions of UMN. However, some participants noted conceptual relationships across domains, such as links between economic burden and societal productivity or between career impacts and social functioning. Most discussion focused on potential correlations among elements within specific domains, particularly within the quality-of-life domain, where concepts such as independence and activities of daily living were viewed as closely related even if conceptually distinguishable.

Clarifications were offered on several element definitions. “Impact on wealth” was considered encompassed within total income; concerns that the term “financial burden” might overemphasize financial toxicity led to suggestions to interpret it more broadly as the cost to the family unit; and “societal perspective” was clarified as referring to non–health care sector impacts. Insurance value was described as capturing both peace of mind from avoiding financial catastrophe and peace of mind from reducing disease-related risk. Participants also questioned why uncertainty appears under treatment administration given that CMS-eligible drugs have long market histories; in response, others noted that individual-level variation may still lead patients to avoid a medicine because of perceived uncertainty. The concept of a “lack of an ideal (eg, curative) treatment” was raised as relating to a value-of-hope perspective, helping clarify how participants distinguished long-tail benefits from other clinical elements.

Finally, several scope-related points were raised. Although some participants proposed adding health system functioning as a domain of UMN, including elements such as workforce shortages or infrastructure constraints, this was deemed conceptually relevant but outside the scope of what medicines directly impact. Questions arose about whether the economic burden aligns with UMN. The group affirmed its appropriateness, noting that the literature broadly defines medical need to include both financial and nonfinancial impacts. Across these discussions, participants agreed that although many elements are distinctly measurable, they may still be correlated. Correlation alone does not imply conceptual duplication, and no elements were merged during the workshop. However, participants acknowledged that future refinement could benefit from clearer operational definitions or empirical methods such as factor analysis to guide decisions about when lumping or splitting is warranted.

These insights informed the subsequent prioritization exercise and helped contextualize how participants interpreted the relative importance of UMN elements.

PRIORITIZATION AND WEIGHTING OF UMN DOMAINS

Participants completed a weighting exercise in which they distributed relative importance across the 7 UMN domains; weights were normalized to sum to 100% (Figure 3). Traditional clinical effectiveness received the highest weight (24.1%, SD = 4.2%), followed by availability of other treatments (19.5%, SD = 3.2%) and impacts on patient, caregiver, or family quality of life (15.8%, SD = 1.2%). Economic burden on patients, caregivers, and families (13.0%, SD = 1.5%), societal perspective elements (11.0%, SD = 1.9%), and economic burden on society (9.5%, SD = 2.2%) received intermediate weights. Elements of treatment administration received a relatively lower weight (7.1%, SD = 2.4%).

FIGURE 3.

Stakeholder Prioritization of UMN Domains

SMART = Simple Multi-Attribute Rating Technique; UMN = unmet medical need.

Narrative Feedback on Prioritization

Participants viewed the overall ranking pattern as broadly consistent with how unmet need is commonly judged. Traditional clinical effectiveness and the availability of other treatments were seen as appropriately receiving the highest weights, given that unmet need is often assessed relative to the effectiveness of existing options. Several participants emphasized that the substantial weighting assigned to quality-of-life impacts reflected their importance as a core dimension of unmet need. Economic burden on patients and families was also regarded as a meaningful contributor.

In contrast, participants noted that the lower weighting of societal-level elements may reflect the composition of the group, which included relatively few individuals involved in population-level decision-making. Some also observed that comparing domains that represent fundamentally different types of unmet need requires individual judgment and may lead different stakeholders to emphasize different aspects. Overall, participants considered the rankings plausible and aligned with the domains they view as most central to unmet need.

STAKEHOLDER ASSESSMENT OF THE NEED FOR BROADER UMN DESCRIPTION AND ITS POLICY IMPLICATIONS

Respondents also rated the extent to which the identified elements reflected aspects of unmet need not currently addressed by the FDA framework (and, by extension, CMS). The mean rating was 4.5 (SD = 0.8), indicating broad agreement that the broader set of elements encompasses dimensions of unmet need beyond those presently considered in CMS evaluations.

Narrative Feedback on Policy Implications

Participants generally expressed the view that the FDA-derived CMS definition of UMN is narrow and does not encompass several dimensions they viewed as important. They highlighted the importance of a more granular and comprehensive definition of quality of life, extending beyond daily living activities, which is the primary focus of the FDA description, to include psychological/emotional health, physical functioning, independence, and cognitive impacts.

Participants also discussed how a broader UMN framework could clarify expectations for evidence generation. By specifying which dimensions matter for assessing unmet need, the framework could help indicate what types of outcomes should be measured, including those not routinely captured in existing data sources. At the same time, participants recognized practical constraints: several domains, especially those related to quality of life, caregiver impacts, and economic burden, would require more consistent collection of patient-reported outcomes.

Feasibility considerations were also raised. Participants observed that CMS typically favors simple, categorical approaches (eg, high, medium, or low unmet need) and that any broader UMN framework would need to be communicated in a way that avoids misinterpretation or being perceived as a prescriptive weighting scheme.

Participants further raised conceptual questions about the definition and boundaries of UMN. Some questioned whether UMN should be treated as a dichotomous or continuous construct and noted potential overlap between UMN elements and other components of CMS’s DPNP, such as comparative effectiveness. Others noted that UMN is dynamic rather than static, observing that levels of unmet need may shift as new therapies enter the market and that current policy processes rarely differentiate between UMN at launch and UMN several years later.

Finally, several participants suggested that future empirical work would be valuable for refining how UMN elements are grouped or differentiated. Methods such as multicollinearity analysis or clustering using patient-level data were mentioned as potential approaches for understanding relationships among elements and informing decisions about grouping or distinguishing related concepts.

Overall, participants viewed a broader UMN framework as potentially useful for improving transparency and structuring deliberation, while emphasizing the need for clarity, feasibility, and alignment with existing decision-making processes.

Discussion

This study provides insights into how UMN can be conceptualized and prioritized in ways that move beyond the narrow FDA-derived description currently used by CMS. The prioritization exercise showed that stakeholders assigned the greatest weight to traditional clinical effectiveness (24%), followed by treatment availability (19%) and quality-of-life impacts (16%), indicating that although clinical outcomes remain central, broader patient-centered and societal dimensions meaningfully contribute to the perception of UMN. Prior research has shown that UMN is inconsistently defined across regulatory and policy settings and is often operationalized in ways that do not support transparent decision-making.^5,10–12 Although several articles have proposed broader conceptualizations, including quality of life, carer burden, economic burden, or the rarity of the disease,^4,6,10,11 none have clarified their relative importance or offered an approach suitable for policy use. By applying a structured, multistep prioritization process across a diverse group of experts, this study offers an initial foundation for broadening and structuring UMN and illustrates a potentially feasible approach that CMS could refine for its own purposes and that other decision-makers could more broadly adapt.

A key finding from this work is that UMN appears to be multidimensional and cannot be adequately characterized by clinical outcomes and functional limitations alone. Stakeholders assigned substantial importance to domains not reflected in FDA’s framing of UMN¹ from perspectives of patients and society, highlighting gaps in the current CMS approach. This supports earlier analyses showing that the FDA definition is incomplete for comprehensive assessment and that CMS should develop its own UMN framework aligned with its statutory responsibilities.⁶ A CMS-specific approach would also clarify what types of evidence are relevant and signal to manufacturers and researchers where additional data collection may be needed.

The study also highlights several conceptual considerations that will be important as broader UMN frameworks evolve. First, the boundaries between UMN and other components of the DPNP framework, such as comparative clinical effectiveness and patient experience, are not clearly defined in current policy language. Some UMN dimensions relevant to patients are already addressed elsewhere in the DPNP guidance, whereas broader societal considerations related to UMN are not explicitly captured. Ensuring that UMN incorporates societal perspectives may therefore address these gaps in the framework while clarifying how different dimensions of value are reflected in CMS decision-making. Clarifying this distinction is crucial to prevent overlap with other aspects of value assessment. Second, unmet need is inherently dynamic, changing as new therapies enter the market. However, existing policy processes rarely differentiate between the need at launch and residual unmet need over time. Third, adopting a broader UMN framework requires clarification of perspective to reflect. For example, in the context of DPNP, CMS should clarify whose perspective is being prioritized—Medicare beneficiaries, taxpayers, or society as a whole—because different groups might assign varying weights to the same domains.

UMN is often cited as a reason for regulatory flexibility and policy incentives. However, if defining it too narrowly, it may inadvertently steer innovation away from areas that truly represent patient and societal priorities. Adopting a broader and more structured definition of UMN can better align innovation incentives with the most pressing needs of patients and society as a whole. Feasibility considerations also inform the interpretation of a broader UMN framework. Several domains viewed as important rely on information not routinely collected in CMS processes, particularly patient-reported outcomes.¹³ Systematic collection of patient-reported outcomes, caregiver burden measures, and societal impact data would be needed to support evidence-based assessment across all 7 domains, and current clinical development programs do not routinely capture many of these.

A related but distinct feasibility consideration concerns decision-making processes. CMS generally favors administratively simple, categorical approaches. Recent work has shown that categorical structures can provide a practical way to incorporate broader concepts without requiring complex scoring systems.¹⁴ To operationalize a multidimensional UMN framework in a reproducible and transparent manner, CMS or payers could adopt structured deliberative processes, such as the prioritization approach demonstrated here, that make explicit which domains are considered, how they are weighted, and how evidence is evaluated against each dimension, reducing reliance on implicit judgment. In this context, a prioritized UMN structure offers a productive path forward: it broadens the concept while remaining interpretable and suitable for deliberation without requiring complex formulas or quantitative scoring systems. The framework presented here should therefore be viewed as a prototype demonstrating the usefulness of prioritization, not a complete policy tool. It should also be noted that although expanding UMN domains may raise evidentiary expectations for manufacturers regarding outcomes not routinely collected in clinical development programs, a prioritized framework simultaneously reduces manufacturer uncertainty by clearly signaling which dimensions matter most, potentially improving the relevance and efficiency of evidence generation compared with the current state where manufacturers lack clear guidance on what evidence will be considered. Broader stakeholder engagement, including payers and policymakers, would be needed to assess whether and how such a framework could be integrated into existing decision-making processes without adding administrative burden.

Methodologically, the results suggest opportunities for further refinement of domain structure through the use of analytic tools like clustering or factor analysis. These tools can help determine when to combine or differentiate elements. Additionally, conducting a larger and more representative group elicitation process could enhance reproducibility and clarify how to incorporate various perspectives. Considering these points, the framework presented here should be seen as an initial prototype. It emphasizes the importance of both broadening and prioritizing the understanding of UMN, rather than being viewed as a final structure ready for direct implementation.

This study demonstrates that a broader understanding of UMN is both important and feasible. It provides a framework for CMS and other decision-makers to create a UMN approach that aligns with stakeholder priorities while facilitating discussion. By identifying key dimensions and demonstrating a practical approach to structuring them, this work lays the groundwork for a more transparent and relevant treatment of UMN in the United States. Future work could expand this framework through larger multistakeholder elicitation exercises and by integrating findings from patient-focused workshops to better understand how expert and patient priorities align or diverge.

LIMITATIONS

This study has several limitations. First, although multiple stakeholder groups participated, researchers were overrepresented, and individual-level prioritization results could not be linked to stakeholder categories, limiting the ability to assess systematic differences across perspectives. In particular, the overrepresentation of researchers may have influenced the domain weights, with societal-level elements potentially receiving lower weights than they would in a group with greater representation of population-level decision-makers. The exercise also involved a relatively small group of experts, which is appropriate for exploratory work but may not reflect the views of a broader or more nationally representative population. Patients were not included in this workshop because the objective was to explore how experts involved in policy, payer decision-making, industry, and research interpret and prioritize UMN elements. Patient perspectives were examined separately in a dedicated workshop and will be reported elsewhere. Second, the workshop relied on a predefined set of UMN elements derived from a structured literature review. Although this ensured consistency, it may have constrained participants from introducing additional concepts beyond the initial framework. Third, group results were displayed after each exercise to support reflection. Although this was intentional to encourage deliberation, it may have introduced anchoring or groupthink effects that could have influenced subsequent responses. Despite efforts to manage cognitive burden, fatigue effects from evaluating 38 elements in a single session may have influenced responses, particularly in later exercises. Fourth, no formal sensitivity analyses or alternative weighting approaches were conducted, which limits the ability to assess the robustness of the derived domain weights. Fifth, narrative feedback was obtained through facilitated discussion rather than structured interviews or formal qualitative coding. Although themes were synthesized systematically, some nuances may not have been fully captured. Although domains used in this research were developed based on a literature review, more recent research has echoed to multidimensionality of UMN definition and distinction across stakeholders.¹⁵ Finally, the workshop took place during the early implementation of the CMS DPNP, and stakeholder views may evolve as CMS issues additional guidance, potentially affecting the applicability of these findings over time. Overall, these limitations reflect the exploratory nature of the study and highlight opportunities for further empirical work, broader stakeholder engagement, and methodological refinement.

Conclusions

This multistakeholder study suggests that UMN extends beyond traditional clinical measures and that structured prioritization can provide a clearer understanding of which domains matter most for evaluation. The approach demonstrated here offers one possible illustration of how CMS and other decision-makers might begin to develop their own UMN frameworks rather than relying solely on FDA’s regulatory definitions. Although the prototype requires further empirical refinement and broader engagement, it provides a foundation for creating more transparent, structured, and policy-relevant assessments of UMN in the United States.

Disclosures

Drs Campbell and Zheng are employed by the National Pharmaceutical Council. Drs McQueen, DiStefano, and Zemplenyi are employed by the University of Colorado, which received funding from the National Pharmaceutical Council to support the work described in this article.