Browse
Themes
The catalogue grouped into 10 thematic families. Each notes how its methods differ.
Economic Evaluation and Costing Methods
9Methods focused on estimating healthcare costs, cost-effectiveness, and economic impact of health technologies.
How they differ: These methods differ in whether they provide cost data (e.g., EU HCSCD, PECUNIA RUC), standardize cost measurement (e.g., PECUNIA templates, Guidance for cost definition), estimate future costs (e.g., PAID 4.0), model individual-level cost-effectiveness (Baseline Risk Score), assess broader fiscal impact (Algorithm), measure resource use (PECUNIA RUM), or evaluate less effective but cheaper interventions (d-CEIs). Choose based on need: use data sources for costs, templates and guidelines for standardization, models for personalized predictions, and toolboxes like d-CEIs for specific policy decisions.
Explore 9 methods →Real-World Data and Evidence Generation
10Approaches and tools for generating evidence from real-world health data, including observational studies and data quality assessment.
How they differ: These methods differ in their purpose: some (like ClinFlow, Trajectories, TreatmentPatterns) focus on exploratory analysis and visualization of real-world data, while others (like CohortMethod, Target trial emulation) aim to estimate causal treatment effects. Choose based on whether the goal is descriptive insight, comparative effectiveness, or trial-like study design.
Explore 10 methods →Patient-Centered Outcomes and Measurement
7Methods that incorporate patient-reported outcomes, quality of life, and patient preferences into health technology assessment.
How they differ: These methods differ in whether they provide value sets for standardizing quality-of-life measurements (e.g., EQ-5D variants, supra-national sets), offer tools to select appropriate PROMs (e.g., PROM-select app, PECUNIA compendium), or give population-specific norms; choose based on need—use value sets for economic evaluation, selection tools for study design, and population norms for benchmarking.
Explore 7 methods →HTA Frameworks and Guidance
0Structured frameworks, checklists, and guidelines to support consistent and transparent health technology assessment processes.
Explore 0 methods →Artificial Intelligence and Predictive Modelling
5AI/ML methods and tools for predicting treatment effects, patient outcomes, and personalizing healthcare decisions.
How they differ: The methods differ in their purpose and technical approach: AI/ML for Treatment Effect Prediction and the PatientLevelPrediction/DeepPatientLevelPrediction R packages focus on predicting individual patient outcomes from observational data, with the latter two providing standardized software frameworks, while DeepPatientLevelPrediction specifically uses deep learning; Explainable AI (XAI) emphasizes transparency in predictions, making models interpretable for clinical trust; the Registry identification tool uses AI to select appropriate patient registries rather than predict outcomes. Choose based on need: use the R packages for model development in OMOP-standardized data, XAI when interpretability is critical, and the registry tool for study design and data sourcing.
Explore 5 methods →Core Outcome Sets for Clinical Trials
5Standardized sets of outcomes developed for specific diseases to ensure consistency in clinical research and practice.
How they differ: These core outcome sets differ primarily in their disease-specific focus and the patient populations involved in their development; the choice between them depends on the clinical context and condition being studied, with each tailored to capture outcomes most relevant to patients and providers in that specific disease area.
Explore 5 methods →Managed Entry Agreements and Reimbursement Tools
8Methods and templates supporting outcomes-based agreements, reimbursement decisions, and post-market monitoring of treatments.
How they differ: The methods differ in their focus: some provide practical tools like checklists, templates, and forms for implementing OBMEAs, while others offer guidance, recommendations, or observational data on reimbursement models; choice depends on whether the user needs procedural support, implementation examples, or policy-level direction for OBMEAs in rare diseases.
Explore 8 methods →Statistical Methods for Treatment Comparison
5Advanced statistical techniques for comparing treatments across diverse populations and study designs.
How they differ: Cross-NMA/NMR and the crossnma R package focus on integrating randomized and non-randomized studies for broader evidence synthesis, while RiskStratifiedEstimation, Two-stage, and Three-stage network meta-regression methods specifically model heterogeneous treatment effects across patient risk levels; choose based on whether the goal is general synthesis (Cross-NMA/NMR) or risk-stratified, individual-level predictions (RiskStratifiedEstimation or two/three-stage regression methods).
Explore 5 methods →Data Standards and Interoperability
6Tools and taxonomies ensuring health data is standardized, high-quality, and usable across systems and countries.
How they differ: The methods differ in their focus: data quality tools like DataQuality Dashboard and its R package assess completeness and consistency of standardized health data, while The Book of OHDSI provides broader guidance on using standardized data and tools for research. Choose based on need—specific data validation (DQD) versus end-to-end research methodology (OHDSI).
Explore 6 methods →Modelling and Simulation Platforms
4Interactive tools and software platforms for building, validating, and running health economic and clinical simulation models.
How they differ: Decision Curve Analysis focuses on comparing multiple treatment strategies to guide clinical decisions, while DICE and SMART are model development tools—DICE provides a structured simulation platform, whereas SMART helps balance model complexity and transparency; the guidelines support best practices in modelling for personalized medicine. Choose Decision Curve Analysis for treatment choice, DICE for building simulations, SMART for model design decisions, and the guidelines for methodological rigor.
Explore 4 methods →