Health Economic Modeling
Expert-defined terms from the Professional Certificate in Health Economics and Market Access course at London School of Planning and Management. Free to read, free to share, paired with a professional course.
Absolute Risk Reduction (ARR) #
Absolute Risk Reduction (ARR)
Explanation #
The difference in event rates between a control group and an intervention group, expressed as a proportion. Example: If 10 % of patients on standard care experience a heart attack versus 6 % on a new drug, ARR = 4 %. Practical application: Used to quantify the clinical benefit of a therapy in cost‑effectiveness models. Challenge: Small absolute differences can lead to large uncertainty when sample sizes are limited.
Accelerated Approval Pathway #
Accelerated Approval Pathway
Explanation #
A regulatory mechanism that permits earlier market entry based on surrogate endpoints that are reasonably likely to predict clinical benefit. Example: Oncology drugs approved after showing tumor shrinkage rather than overall survival. Practical application: Models often incorporate conditional reimbursement and post‑marketing evidence requirements. Challenge: Estimating long‑term value is difficult when definitive outcomes are unavailable.
Adverse Event (AE) Costing #
Adverse Event (AE) Costing
Explanation #
Assigning monetary values to the management of side effects, including treatment, monitoring, and hospitalisation. Example: Grade 3 neutropenia may require inpatient care costing $5,000 per episode. Practical application: Integrated into incremental cost‑effectiveness ratios (ICERs) to reflect safety profiles. Challenge: Data on AE incidence and resource use may be sparse or heterogeneous across trials.
Agency for Healthcare Research and Quality (AHRQ) #
Agency for Healthcare Research and Quality (AHRQ)
Explanation #
A U.S. Federal agency that produces tools such as the Quality‑Adjusted Life‑Year (QALY) and disease‑specific cost databases. Practical application: Modelers use AHRQ cost estimates for inpatient stays and procedures. Challenge: Aligning U.S. Cost data with international pricing structures can be problematic.
Allocation Rule #
Allocation Rule
Explanation #
A predefined principle that determines how limited resources are distributed among competing health interventions. Example: “First‑come, first‑served” or “maximise QALYs.”
Practical application #
Embedded in budget impact models to simulate policy decisions. Challenge: Ethical considerations and stakeholder preferences may conflict with efficiency goals.
Alternative Discount Rate #
Alternative Discount Rate
Explanation #
The rate used to convert future costs and outcomes to present values, often varied in sensitivity analyses. Typical values: 3 % For costs, 1.5 % For health outcomes. Practical application: Influences the magnitude of long‑term benefits in chronic disease models. Challenge: Selecting an appropriate rate that reflects societal time preferences and inflation.
Amortisation of Capital Costs #
Amortisation of Capital Costs
Explanation #
Spreading the expense of capital equipment (e.G., MRI scanner) over its useful life. Example: A $2 million scanner with a 10‑year lifespan yields $200 000 annual cost. Practical application: Included in health‑system perspective models to capture infrastructure investments. Challenge: Determining appropriate lifespan and residual value for medical technology.
Analytic Horizon #
Analytic Horizon
Explanation #
The total period over which costs and outcomes are projected in a model. Common horizons: Lifetime, 5‑year, or 10‑year. Practical application: Determines whether long‑term benefits such as survival gains are captured. Challenge: Longer horizons increase uncertainty and require extrapolation beyond trial data.
Application Programming Interface (API) #
Application Programming Interface (API)
Explanation #
A set of protocols that allow software applications to exchange data, often used to pull real‑world evidence from electronic health records. Practical application: Automates data extraction for model inputs. Challenge: Ensuring data privacy, standardisation, and compatibility across systems.
Assumption Testing #
Assumption Testing
Explanation #
Systematically varying model assumptions to assess their impact on results. Example: Changing disease progression rates from fast to slow. Practical application: Provides robustness checks for decision‑makers. Challenge: Identifying which assumptions are most influential without over‑complicating the analysis.
Budget Impact Analysis (BIA) #
Budget Impact Analysis (BIA)
Explanation #
Estimates the financial consequences of adopting a new intervention within a specific budget context. Example: Projected increase in pharmacy spend after introducing a biologic. Practical application: Required by many HTA bodies to inform reimbursement decisions. Challenge: Requires detailed uptake estimates and may be sensitive to price negotiations.
Cost‑Benefit Analysis (CBA) #
Cost‑Benefit Analysis (CBA)
Explanation #
Compares the monetary value of benefits to costs, producing a net benefit figure. Example: Valuing a QALY at $150 000 and subtracting intervention costs. Practical application: Facilitates comparison across sectors (e.G., Health vs. Transportation). Challenge: Assigning a societal willingness‑to‑pay per QALY can be contentious.
Cost‑Effectiveness Threshold #
Cost‑Effectiveness Threshold
Explanation #
The maximum amount a payer is prepared to spend for one additional unit of health benefit (e.G., $50 000 Per QALY). Practical application: Determines whether an intervention is deemed “cost‑effective.”
Challenge #
Thresholds vary across jurisdictions and may not reflect budget constraints.
Cost‑Effectiveness Plane #
Cost‑Effectiveness Plane
Explanation #
A graphical representation dividing outcomes into four quadrants based on cost and effectiveness differences. Quadrants: More effective‑more costly, more effective‑less costly, less effective‑more costly, less effective‑less costly. Practical application: Visualises uncertainty via scatter plots from probabilistic sensitivity analysis. Challenge: Interpretation can be ambiguous when points cross multiple quadrants.
Cost‑Effectiveness Ratio (CER) #
Cost‑Effectiveness Ratio (CER)
Explanation #
The ratio of total costs to total health outcomes for a single intervention (e.G., $30 000 Per QALY). Distinguished from the incremental ratio, which compares two alternatives. Practical application: Provides a baseline efficiency measure. Challenge: May be misleading without a comparator.
Cost‑Utility Analysis (CUA) #
Cost‑Utility Analysis (CUA)
Explanation #
A form of economic evaluation that incorporates quality of life into the measurement of benefits. Example: Calculating cost per QALY gained for a new vaccine. Practical application: Preferred by many HTA agencies for its ability to compare across disease areas. Challenge: Requires robust utility data, often derived from generic instruments like EQ‑5D.
Cost‑of‑Illness (COI) Study #
Cost‑of‑Illness (COI) Study
Explanation #
Quantifies the economic burden of a disease, including medical expenditures and productivity losses. Example: Annual COI for diabetes in the United States exceeds $300 billion. Practical application: Supplies baseline cost inputs for model calibration. Challenge: Capturing intangible costs such as pain and suffering.
Cost‑Sharing #
Cost‑Sharing
Explanation #
The portion of health expenses paid by patients rather than insurers. Practical application: Influences adherence rates in models of chronic therapies. Challenge: Varying cost‑sharing structures across plans complicate generalisation.
Credible Interval #
Credible Interval
Explanation #
In Bayesian analysis, the range within which a parameter lies with a specified probability (e.G., 95 %). Practical application: Communicates uncertainty around model parameters. Challenge: Requires prior distributions that may be subjective.
Decision Analytic Model #
Decision Analytic Model
Explanation #
A structured representation of clinical pathways used to estimate costs and outcomes under alternative strategies. Practical application: Core tool in health‑economic evaluations. Challenge: Balancing model complexity with data availability.
Discount Rate #
Discount Rate
Explanation #
The percentage used to convert future costs and health effects to present‑day values. Standard rates: 3 % For costs, 1.5 % For health outcomes. Practical application: Affects long‑term cost‑effectiveness results, especially for chronic diseases. Challenge: Debate exists over appropriate rates for different jurisdictions.
Discrete Event Simulation (DES) #
Discrete Event Simulation (DES)
Explanation #
A modelling technique that tracks individual entities as they experience events over time, allowing for complex interactions and resource constraints. Practical application: Captures queueing effects in hospital settings. Challenge: Requires detailed data on event timing and often high computational demand.
Deterministic Sensitivity Analysis (DSA) #
Deterministic Sensitivity Analysis (DSA)
Explanation #
Varies one parameter at a time while holding others constant to assess impact on outcomes. Practical application: Identifies key drivers of model results. Challenge: May underestimate joint uncertainty when parameters are correlated.
Drug Price Index (DPI) #
Drug Price Index (DPI)
Explanation #
A measure that tracks changes in pharmaceutical prices over time, often adjusted for inflation. Practical application: Used to update model inputs for future cost projections. Challenge: Variability across therapeutic classes and market‑specific pricing agreements.
Dynamic Transmission Model #
Dynamic Transmission Model
Explanation #
Simulates the spread of an infectious agent through populations, incorporating time‑varying contact patterns and immunity. Practical application: Evaluates vaccination strategies and herd‑immunity effects. Challenge: Requires detailed epidemiological data and assumptions about mixing patterns.
Epidemiological Parameter #
Epidemiological Parameter
Explanation #
Quantitative measures describing disease frequency or progression used as inputs in health‑economic models. Example: Annual incidence of myocardial infarction in a 65‑year‑old cohort. Practical application: Drives the number of events and associated costs. Challenge: Sources may differ in case definitions and diagnostic criteria.
EQ‑5D #
EQ‑5D
Explanation #
A standardized instrument that generates a five‑dimension health state profile, convertible to a utility value for QALY calculations. Practical application: Frequently used to collect patient‑reported outcomes in clinical trials. Challenge: Ceiling effects and cultural differences may affect comparability.
Extrapolation #
Extrapolation
Explanation #
Extending observed trial data beyond the follow‑up period to estimate long‑term outcomes. Common methods: Weibull, exponential, log‑logistic distributions. Practical application: Generates lifetime survival curves for cost‑effectiveness analysis. Challenge: Choice of distribution can markedly influence results; external validation is essential.
External Validation #
External Validation
Explanation #
Comparing model predictions with independent data sources not used in model development. Practical application: Increases confidence in the model’s applicability to real‑world settings. Challenge: Availability of high‑quality external datasets may be limited.
Factorial Design #
Factorial Design
Explanation #
A study design that evaluates multiple interventions simultaneously, allowing assessment of combined effects. Example: 2 × 2 Design testing drug A, drug B, both, or neither. Practical application: Provides interaction parameters for models when therapies may be used together. Challenge: Increases sample size requirements and analytical complexity.
Frequentist Approach #
Frequentist Approach
Explanation #
A statistical paradigm that interprets probability as the long‑run frequency of events. Practical application: Commonly used for hypothesis testing in clinical trials that feed model inputs. Challenge: Does not incorporate prior information, which may be valuable in sparse data contexts.
Future Discounting #
Future Discounting
Explanation #
The process of applying a discount factor to future costs and outcomes to reflect societal preferences for present consumption. Practical application: Standard practice in health‑economic evaluations. Challenge: Choosing a discount rate that balances inter‑generational equity and fiscal reality.
Generic Utility Measure #
Generic Utility Measure
Explanation #
Instruments designed to assess health‑related quality of life across a wide range of conditions, enabling cross‑disease comparisons. Practical application: Allows aggregation of QALYs in multi‑indication models. Challenge: May lack sensitivity for disease‑specific symptoms.
Health Technology Assessment (HTA) #
Health Technology Assessment (HTA)
Explanation #
A systematic process that evaluates the clinical and economic value of health technologies to inform policy. Examples: NICE (UK), CADTH (Canada), IQWiG (Germany). Practical application: Determines market access and pricing negotiations. Challenge: Varying methodological requirements across agencies.
Health‑Adjusted Life Year (HALY) #
Health‑Adjusted Life Year (HALY)
Explanation #
A generic metric that combines quantity and quality of life, encompassing both QALYs (positive health) and DALYs (burden of disease). Practical application: Facilitates comparison of interventions with differing health impacts. Challenge: Converting between QALY and DALY frameworks may require assumptions about disability weights.
Incremental Cost‑Effectiveness Ratio (ICER) #
Incremental Cost‑Effectiveness Ratio (ICER)
Explanation #
The ratio of the difference in costs to the difference in effectiveness between two alternatives (ΔCost/ΔEffect). Practical application: Core output for decision‑makers evaluating new therapies. Challenge: Interpretation becomes ambiguous when the comparator is dominated (more costly and less effective).
Input Parameter Uncertainty #
Input Parameter Uncertainty
Explanation #
Uncertainty arising from variability in model inputs such as transition probabilities, utilities, or costs. Practical application: Modeled using distributions (e.G., Beta for probabilities). Challenge: Requires specification of appropriate distribution types and parameters.
Intention‑to‑Treat (ITT) Analysis #
Intention‑to‑Treat (ITT) Analysis
Explanation #
An analytical approach that includes all randomised participants in the groups to which they were assigned, regardless of adherence. Practical application: Provides conservative effectiveness estimates for model inputs. Challenge: May dilute treatment effect if crossover or dropout rates are high.
Item‑Response Theory (IRT) #
Item‑Response Theory (IRT)
Explanation #
A statistical framework that models the probability of a particular response to a questionnaire item based on underlying latent traits. Practical application: Improves the precision of utility measurement from PROMs. Challenge: Requires large sample sizes and complex software.
Joint Modeling #
Joint Modeling
Explanation #
Simultaneously analyses repeated measures (e.G., Biomarkers) and time‑to‑event data to capture their interdependence. Practical application: Enhances prediction of disease progression in cost‑effectiveness models. Challenge: Computationally intensive and demands sophisticated statistical expertise.
Lifetime Horizon #
Lifetime Horizon
Explanation #
Extending the model’s projection until the cohort’s death, ensuring capture of all relevant costs and benefits. Practical application: Standard for chronic diseases where benefits accrue over many years. Challenge: Requires extrapolation beyond observed data, increasing uncertainty.
Markov Model #
Markov Model
Explanation #
A stochastic model that represents disease progression through a finite set of health states, with transition probabilities applied each cycle. Practical application: Widely used for chronic conditions such as diabetes or cardiovascular disease. Challenge: The “memoryless” property may oversimplify histories that affect future risk.
Markov Chain Monte Carlo (MCMC) #
Markov Chain Monte Carlo (MCMC)
Explanation #
An algorithm that generates samples from a probability distribution by constructing a Markov chain, enabling Bayesian parameter estimation. Practical application: Used to derive posterior distributions for uncertain model inputs. Challenge: Requires convergence diagnostics and can be computationally demanding.
Micro‑Costing #
Micro‑Costing
Explanation #
Detailed quantification of each resource used in patient care, assigning unit costs to generate total cost per patient. Practical application: Provides high‑resolution cost data for specific interventions. Challenge: Time‑consuming and may be limited by data availability.
Microsimulation #
Microsimulation
Explanation #
Simulates the life course of individual patients, allowing heterogeneity in risk factors and treatment pathways. Practical application: Captures patient‑level variability and complex treatment histories. Challenge: Requires extensive data on individual characteristics and can be computationally intensive.
Monte Carlo Simulation #
Monte Carlo Simulation
Explanation #
A technique that repeatedly samples from probability distributions of inputs to generate a distribution of outcomes. Practical application: Quantifies overall model uncertainty and produces cost‑effectiveness acceptability curves. Challenge: Number of iterations must be sufficient to achieve stable results.
Net Monetary Benefit (NMB) #
Net Monetary Benefit (NMB)
Explanation #
A reformulation of the ICER that expresses value in monetary terms: (ΔEffect × WTP) − ΔCost. Practical application: Simplifies probabilistic analysis and decision rules. Challenge: Dependent on the chosen WTP threshold, which may be uncertain.
Net Health Benefit (NHB) #
Net Health Benefit (NHB)
Explanation #
The difference in health outcomes (e.G., QALYs) after adjusting for costs using a willingness‑to‑pay value. Formula: ΔEffect − (ΔCost/WTP). Practical application: Allows comparison across interventions with different cost structures. Challenge: Requires a consensus WTP value.
Network Meta‑Analysis (NMA) #
Network Meta‑Analysis (NMA)
Explanation #
A statistical method that combines direct and indirect evidence across a network of interventions to estimate relative effects. Practical application: Supplies comparative efficacy inputs when head‑to‑head trials are absent. Challenge: Assumes transitivity and consistency, which may be violated.
Non‑Parametric Bootstrap #
Non‑Parametric Bootstrap
Explanation #
A resampling technique that draws repeated samples with replacement from the observed data to estimate the sampling distribution of a statistic. Practical application: Generates empirical confidence intervals for cost or effectiveness estimates. Challenge: Requires sufficient original sample size to produce reliable resamples.
Observational Data #
Observational Data
Explanation #
Data collected outside of randomised controlled trials, often from electronic health records, claims databases, or disease registries. Practical application: Informs model inputs such as resource utilisation, adherence, and long‑term outcomes. Challenge: Susceptible to confounding and selection bias.
Outcome Measure #
Outcome Measure
Explanation #
The health endpoint used to assess the benefit of an intervention, ranging from clinical events to utility‑adjusted life years. Practical application: Determines the numerator in cost‑effectiveness ratios. Challenge: Selecting measures that capture all relevant aspects of patient benefit.
Patient‑Reported Outcome Measure (PROM) #
Patient‑Reported Outcome Measure (PROM)
Explanation #
Instruments that capture patients’ perspectives on their health status, symptoms, and functional abilities. Examples: EQ‑5D, SF‑36, disease‑specific questionnaires. Practical application: Generates utility values for QALY calculations. Challenge: Missing data and cultural differences can affect validity.
Pharmacoeconomic Evaluation #
Pharmacoeconomic Evaluation
Explanation #
The systematic assessment of the value of pharmaceutical products, incorporating both costs and outcomes. Practical application: Supports reimbursement and pricing decisions. Challenge: Rapidly changing market dynamics and confidential discounts complicate analyses.
Placebo Effect #
Placebo Effect
Explanation #
Improvement in patient outcomes attributable to expectations rather than the active intervention. Practical application: Must be accounted for when translating trial efficacy to real‑world effectiveness. Challenge: Quantifying the magnitude of the placebo effect for model inputs is difficult.
Probabilistic Sensitivity Analysis (PSA) #
Probabilistic Sensitivity Analysis (PSA)
Explanation #
Simultaneously varies all uncertain parameters according to predefined probability distributions to assess overall model uncertainty. Practical application: Produces cost‑effectiveness acceptability curves and probability of being cost‑effective. Challenge: Requires specification of appropriate distributions and correlation structures.
Quality‑Adjusted Life Year (QALY) #
Quality‑Adjusted Life Year (QALY)
Explanation #
A measure that combines length of life with health‑related quality of life, where 1 QALY equals one year in perfect health. Practical application: Standard metric for cost‑utility analyses. Challenge: Utility measurement methods and cultural valuation of health states can vary.
Real‑World Evidence (RWE) #
Real‑World Evidence (RWE)
Explanation #
Clinical evidence regarding the usage and potential benefits or risks of a medical product derived from analysis of real‑world data. Practical application: Supplements trial data for long‑term effectiveness and safety inputs. Challenge: Data quality, standardisation, and privacy concerns.
Reference Case #
Reference Case
Explanation #
A predefined set of methodological assumptions (e.G., Perspective, discount rates, time horizon) used to ensure comparability across evaluations. Practical application: Aligns model outputs with agency expectations. Challenge: May limit flexibility to address specific decision contexts.
Resource Utilisation #
Resource Utilisation
Explanation #
The quantity and type of health‑care services consumed (e.G., Hospital stays, physician visits, medication doses). Practical application: Drives cost calculations in economic models. Challenge: Capturing variation across settings and patient pathways.
Scenario Analysis #
Scenario Analysis
Explanation #
Evaluates model outcomes under alternative sets of assumptions (e.G., Best‑case, worst‑case). Practical application: Explores the impact of structural choices such as alternative comparators or policy changes. Challenge: Selecting plausible scenarios without over‑complicating the analysis.
Sensitivity Analysis #
Sensitivity Analysis
Explanation #
A broad term for any systematic variation of model inputs to assess robustness of results. Practical application: Identifies parameters that most influence cost‑effectiveness. Challenge: Balancing thoroughness with interpretability.
Simulation Horizon #
Simulation Horizon
Explanation #
The period over which a simulation runs, often synonymous with analytic horizon but may refer specifically to the computational runtime. Practical application: Determines data storage and processing requirements. Challenge: Longer horizons increase computational load and uncertainty.
Societal Perspective #
Societal Perspective
Explanation #
An analytic viewpoint that includes all costs and benefits regardless of who incurs them, encompassing productivity losses, informal care, and taxes. Practical application: Provides a comprehensive assessment of an intervention’s economic impact. Challenge: Data on indirect costs are often scarce and methodologically contentious.
Standardised Mortality Ratio (SMR) #
Standardised Mortality Ratio (SMR)
Explanation #
The ratio of observed deaths in a study population to the number expected based on a reference population. Practical application: Adjusts baseline mortality in models for specific sub‑populations. Challenge: Requires accurate baseline mortality data and appropriate standardisation.
Structural Uncertainty #
Structural Uncertainty
Explanation #
Uncertainty arising from the choice of model type, health states, or the way processes are represented. Practical application: Explored through alternative model structures or scenario analyses. Challenge: Hard to quantify formally; often addressed qualitatively.
Survival Analysis #
Survival Analysis
Explanation #
Statistical methods for analysing the time until an event occurs, often using Kaplan‑Meier curves or Cox proportional hazards models. Practical application: Provides transition probabilities for Markov models. Challenge: Censoring and competing risks must be appropriately handled.
Time‑Dependent Transition Probabilities #
Time‑Dependent Transition Probabilities
Explanation #
Transition probabilities that vary over time, reflecting changing risk as patients age or disease progresses. Practical application: Increases model realism for chronic diseases. Challenge: Requires detailed longitudinal data to estimate.
Utility #
Utility
Explanation #
A numeric representation of the desirability of a health state, anchored at 0 (dead) and 1 (full health). Practical application: Multiplied by time spent in a state to calculate QALYs. Challenge: Preference elicitation methods (e.G., Standard gamble, time trade‑off) can produce divergent values.
Value‑Based Pricing #
Value‑Based Pricing
Explanation #
Setting the price of a health technology based on its estimated health benefit and willingness‑to‑pay threshold. Practical application: Aligns manufacturer price with the value delivered to the health system. Challenge: Requires robust, transparent cost‑effectiveness data and may be resisted by payers.
Variance Reduction Techniques #
Variance Reduction Techniques
Explanation #
Methods used to improve the efficiency of stochastic simulations, reducing the number of iterations needed for stable estimates. Practical application: Speeds up probabilistic sensitivity analyses. Challenge: Implementation may be complex and requires careful validation.
Virtual Twin Modeling #
Virtual Twin Modeling
Explanation #
Creating a simulated counterpart for each real patient to estimate outcomes under alternative treatment strategies. Practical application: Enables personalised cost‑effectiveness estimates. Challenge: Demands high‑dimensional data and sophisticated modelling platforms.
Willingness‑to‑Pay (WTP) Threshold #
Willingness‑to‑Pay (WTP) Threshold
Explanation #
The maximum amount a society or payer is prepared to spend for one additional unit of health gain (e.G., $50 000 Per QALY). Practical application: Determines whether an ICER is considered acceptable. Challenge: Thresholds vary by country, disease severity, and budget constraints.
World Health Organization (WHO) Cost‑Effectiveness Benchmarks #
World Health Organization (WHO) Cost‑Effectiveness Benchmarks
Explanation #
Recommendations that interventions costing less than three times a country's gross domestic product per capita per DALY averted are cost‑effective. Practical application: Provides a reference for low‑ and middle‑income settings. Challenge: Criticised for being too permissive and not reflecting actual willingness to pay.
Zero‑Cost Intervention #
Zero‑Cost Intervention
Explanation #
A therapeutic option that incurs no additional costs compared with standard care, often due to existing infrastructure or generic status. Practical application: May still require evaluation of health outcomes to assess overall value. Challenge: Hidden costs such as training or monitoring may be overlooked.