The work, published as a comment in the journal Cost Effectiveness and Resource Allocation, argues that the way outcomes are selected in these evaluations can quietly distort how much value a treatment appears to deliver, and that no single outcome measure currently performs well across all the criteria that matter for decision-making.
The team, led by Annick Meertens with colleagues including Lieven Annemans, Floor Verstraete, Isabelle Hoorens and Nick Verhaeghe, points out that healthcare systems worldwide face increasing resource constraints, making robust health economic evaluations essential for evidence-based decisions in skin cancer care. Their central concern is that selecting incomplete or inappropriate outcomes may under- or overestimate benefits for patients, misrepresent the trade-offs between costs and outcomes, and ultimately lead to suboptimal allocation of scarce healthcare resources.
The quality-adjusted life year, or QALY, remains the dominant outcome measure in health economic evaluations and is widely recommended in international methodological guidelines. QALYs combine length of life and quality of life using preference-based utility values on a scale from 0, representing death, to 1, representing perfect health. In many health technology assessment settings, cost-effectiveness is judged by the incremental cost per QALY gained relative to a willingness-to-pay threshold, which means considerable weight rests on how outcomes are measured and incorporated into the analysis.
Yet QALYs have been widely debated. The authors note that QALY-based evaluations do not always align with societal preferences about how scarce healthcare resources should be distributed. The value assigned to a QALY may vary depending on those preferences. Traditionally, priority goes to interventions that generate the greatest health gains per unit of investment regardless of context, while other approaches, such as severity weighting, assign greater value to health gains in more severely ill patients.
Beyond these conceptual debates, practical problems arise when QALYs derived from generic instruments are applied in disease-specific contexts. Their narrow focus on health may not capture all relevant benefits of an intervention, particularly in skin cancer, where generic instruments often lack sufficient sensitivity to capture meaningful changes or reach the minimal clinically important difference. Skin cancer interventions generate effects across multiple domains, including survival, recurrence, cosmetic outcomes and care processes, which the authors argue require a broader conceptualization of benefits.
To structure these effects, the researchers categorized outcome measures used in skin cancer health economic evaluations into two main domains: quality of life outcomes and clinical effectiveness outcomes, while acknowledging that some concepts may overlap depending on how they are measured. Quality of life can be assessed with generic instruments such as the EQ-5D, the SF-6D or the 15D, and preference-based methods such as the time trade-off are commonly used to derive value sets for these instruments. Generic instruments are suitable for comparability between diseases and are well interpretable for health technology assessment, but they may lack sensitivity for outcomes particularly relevant in skin cancer, such as the appearance-related effects of treatment and scarring, fear of recurrence or subsequent skin cancers given the considerable risk of additional lesions, and the burden of adapting sun-related behavior.
This limitation applies especially to keratinocyte carcinomas, meaning basal cell carcinoma and squamous cell carcinoma, where mortality is minimal and utility values often show ceiling effects. Some evaluations therefore employ disease-specific patient-reported outcome measures, such as the Functional Assessment of Cancer Therapy – Melanoma questionnaire, the Basal and Squamous Cell Carcinoma Quality of Life instrument, or the Skin Cancer Index, to better reflect skin cancer-related quality of life. These measures can be mapped to utility values where a validated algorithm exists, but mapping introduces additional uncertainty, and the validity of disease-specific instruments is often insufficiently established. According to the authors, only one mapping algorithm currently exists, for the FACT-M questionnaire, and such endpoints remain dependent on the choice of questionnaire, since each instrument addresses different domains of disease burden.
Clinical outcomes form the second domain and can be split into hard and intermediate endpoints. Hard endpoints such as life years gained, mortality and survival are important considerations in melanoma, particularly in advanced disease, but are less informative in keratinocyte carcinoma given its low mortality rate. Intermediate outcomes capture earlier effects within diagnostic or treatment pathways, including stage at diagnosis, time to diagnosis, diagnostic accuracy, surgical margins, recurrence rates and treatment complications. Although clinically relevant, intermediate outcomes do not necessarily establish effects on health-related quality of life or survival, so they require additional modelling to be linked to downstream clinical consequences, resource use and costs, which introduces structural assumptions and associated uncertainty. For example, recurrence can be readily represented as a health state or event within a model, but estimating its effect on QALYs requires evidence on subsequent treatment, quality of life, resource use and survival.
The decision matrix itself was developed as a conceptual framework informed by an iterative search of the methodological and applied literature, including PubMed searches on outcome measurement, patient-reported outcomes and health economic evaluations in skin cancer, plus guidance from NICE, ISPOR task forces and the Belgian Health Care Knowledge Center. The authors stress that the search was intended to inform conceptual development rather than provide a systematic or exhaustive synthesis. Assessment criteria were organized into four domains: measurement properties, covering sensitivity, validity, reliability and alignment with intervention mechanisms; model integration, covering compatibility with standard model structures and economic translatability; decision relevance, covering patient-relevant impact, comparability across studies and interpretability for reimbursement decisions; and feasibility, covering data availability, measurement burden and transferability across countries and healthcare systems. Qualitative ratings were assigned through discussion and consensus within the multidisciplinary author team, informed by literature and their expertise, rather than through a quantitative scoring approach, because the matrix is intended to make strengths, limitations and trade-offs explicit rather than to calculate an aggregate ranking.
Applying the matrix highlights two key trade-offs. The first lies between sensitivity and comparability: generic quality of life instruments perform well in comparability and model integration, making them suitable for health technology assessment, but may miss specific aspects of skin cancer burden such as cosmetic outcomes, treatment-related anxiety or fear of recurrence, whereas disease-specific instruments are more sensitive to these effects but lack direct comparability across diseases and interventions. The second trade-off arises between clinical relevance and economic translatability: hard endpoints such as survival are recognized in guidelines and directly incorporable into evaluations, but their relevance varies across skin cancer types, while intermediate outcomes such as diagnostic accuracy align closely with intervention mechanisms yet require additional modelling assumptions when the aim is to estimate downstream life years or QALYs. Intermediate outcomes can still be used directly in evaluations reporting cost per case detected or recurrence avoided. The authors emphasize that the matrix is not intended to eliminate trade-offs, that performance may vary by measure, intervention, population and context, and that the framework has not yet been tested in actual evaluations. They also acknowledge the absence of a formal consensus process and direct patient or stakeholder involvement as an important limitation, and propose refinement through a Delphi panel including patients, researchers, clinicians and policy makers. Although developed in the skin cancer context, the underlying principles may be relevant to other disease areas with multidimensional intervention effects, such as atopic dermatitis and mental health disorders, though applicability beyond skin cancer requires further evaluation.
Subject of Research: Outcome measure selection for health economic evaluations in skin cancer care
Article Title: Towards meaningful outcome measures for health economic evaluations in skin cancer
Article References: Meertens, A., Annemans, L., Verstraete, F., Hoorens, I., & Verhaeghe, N. (2026). Towards meaningful outcome measures for health economic evaluations in skin cancer. Cost Effectiveness and Resource Allocation, 24(1), Article 130. https://doi.org/10.1186/s12962-026-00855-5
Image Credits: AI Generated
DOI: 10.1186/s12962-026-00855-5
Keywords: skin cancer, health economic evaluation, QALY, outcome measures, health technology assessment, quality of life, decision matrix, keratinocyte carcinoma, melanoma, patient-reported outcomes, cost-effectiveness, resource allocation
News Source: Nathaniel Bowman. (October 29, 2026). New Decision Matrix Aims to Fix Outcome Selection in Skin Cancer Economics. Scienmag.
