Concept Architecture
Concept
Theoretically, Sensitivity Analysis is a systematic analytical framework used to evaluate how uncertainty in model inputs, assumptions or structural choices influences model outputs. It provides a means of assessing the robustness of conclusions by examining the extent to which changes in uncertain components alter estimated costs, health outcomes and cost-effectiveness. In health economics, sensitivity analysis is an essential component of decision modelling because it quantifies the impact of uncertainty on reimbursement and policy decisions.
Mathematically, sensitivity analysis involves recalculating model outputs after varying uncertain parameters, groups of parameters or model structures according to predefined methods. These methods include deterministic approaches, such as one-way and multi-way sensitivity analyses, and stochastic approaches, such as probabilistic sensitivity analysis. The resulting changes in model outputs provide quantitative measures of model robustness and decision uncertainty.
In practice, sensitivity analysis is performed by varying parameter values, probability distributions, structural assumptions or methodological choices using evidence-based ranges or alternative scenarios. Results are presented using tornado diagrams, threshold analyses, cost-effectiveness acceptability curves, scenario comparisons or global sensitivity measures to identify influential sources of uncertainty and evaluate the reliability of model conclusions.
Purpose
Used to evaluate the robustness of health economic model results, quantify decision uncertainty, identify influential assumptions and support transparent evidence-based decision-making.
Mathematical Formulae
Primary Formula
There is no universally recognised canonical mathematical formula.
Supporting Formulae
Incremental Cost-Effectiveness Ratio:
ICER = ?C / ?E
Net Monetary Benefit:
NMB = ?E ? C
Incremental Net Monetary Benefit:
INMB = ??E ? ?C
Percentage change in outcome:
% Change = ((Outcome??w ? Outcome????) / Outcome????) ? 100%
Related Mathematical Methods
- Deterministic Sensitivity Analysis
- One-Way Sensitivity Analysis
- Multi-Way Sensitivity Analysis
- Probabilistic Sensitivity Analysis
- Global Sensitivity Analysis
- Scenario Analysis
- Threshold Analysis
- Tornado Diagram
Example
A base-case analysis estimates an ICER of �26,500 per QALY. One-way sensitivity analysis demonstrates that increasing treatment cost raises the ICER to �31,200 per QALY, while probabilistic sensitivity analysis estimates an 81% probability of cost-effectiveness at a willingness-to-pay threshold of �30,000 per QALY. Together, these analyses demonstrate the robustness of the model and identify the principal sources of decision uncertainty.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| Data Table | =TABLE(ParameterInputCell,) | Perform deterministic sensitivity analyses by recalculating outcomes across parameter values. |
| IF | =IF(B2<=30000,"Cost-effective","Not cost-effective") | Evaluate model conclusions at the willingness-to-pay threshold. |
| RAND | =RAND() | Generate random values for probabilistic sensitivity analysis. |
| NORM.INV | =NORM.INV(RAND(),Mean,SD) | Sample uncertain continuous model parameters during Monte Carlo simulation. |
| SUMPRODUCT | =SUMPRODUCT(Weights,Results) | Calculate weighted model outcomes where appropriate. |
VBA (Optional)
VBA can automate deterministic and probabilistic sensitivity analyses, generate tornado diagrams and summarise uncertainty across multiple health economic scenarios.
Sources
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press.
- Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes. Oxford University Press.
- NICE. NICE Health Technology Evaluations: The Manual.
- Briggs AH, Weinstein MC, Fenwick EAL, Karnon J, Sculpher MJ, Paltiel AD. Model parameter estimation and uncertainty analysis: a report of the ISPOR-SMDM Modeling Good Research Practices Task Force. Medical Decision Making. 2012;32(5):722?732.
Related Concepts (3)
Library
Publications
2
Foundations of Cost-Effectiveness Analysis for Health and Medical Practices — Weinstein & Stason, Vol. 296, No. 13 ed., 1977 (New England Journal of Medicine)
The founding paper of health cost-effectiveness analysis, establishing the cost-per-outcome ratio as an index for setting priorities, the use of quality-adjusted life expectancy, discounting of future costs and benefits, and sensitivity analysis — the intellectual origin of the modern CEA/QALY framework.
Journal ArticleView source →OMB Circular A-4: Regulatory Analysis — Office of Management and Budget, Reinstated 2025 ed., 2003 (Executive Office of the President of the United States)
Federal guidance on benefit-cost analysis, baseline selection, valuation, discounting, uncertainty and comparison of regulatory alternatives.
Frequently Asked Questions (6)
What is sensitivity analysis?
A general term for any technique assessing how a model's results change in response to variation in its inputs or structural assumptions.
Source: Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press; 2006. doi:10.1093/oso/9780198526629.001.0001.
Why is sensitivity analysis needed even for a well-estimated model?
Every model rests on inputs and assumptions that are uncertain to some degree, so a single result, however carefully derived, could change if those inputs were different. Sensitivity analysis probes this by seeing how the result responds when inputs and assumptions are varied, revealing whether the conclusion is robust or hinges on particular values. Even a well-estimated model needs it, because a decision-maker must know how much confidence the result deserves. It tests the durability of the conclusion. Briggs and colleagues (2006) describe its role.
Source: Briggs et al. 2006
What are the main types of sensitivity analysis?
The main types of sensitivity analysis include deterministic sensitivity analysis, varying inputs to fixed alternative values, one at a time in one-way analysis or together in multi-way analysis; probabilistic sensitivity analysis, representing inputs by distributions and sampling them to characterise combined parameter uncertainty; scenario analysis, examining coherent alternative sets of structural and methodological assumptions; and threshold analysis, finding the input values at which conclusions change. Each addresses different aspects of uncertainty, and they are used together to give a full picture of how robust a model's results are.
Source: Briggs, Claxton & Sculpher 2006
Why is sensitivity analysis important?
Sensitivity analysis is important because a model's results depend on uncertain inputs and assumptions, so presenting a single result without exploring their effect would give a false sense of precision and could mislead decisions. By showing how results change as inputs and assumptions vary, sensitivity analysis reveals which uncertainties matter, whether conclusions are robust, and where further research would help. It thus supports honest, informed decision making under uncertainty and is expected by guidance bodies, making it a core part of any credible economic evaluation.
Source: Drummond et al. 2015
How does sensitivity analysis address different sources of uncertainty?
Sensitivity analysis addresses different sources of uncertainty with different methods: parameter uncertainty, imperfect knowledge of input values, is handled by deterministic variation and, more fully, by probabilistic analysis sampling input distributions; structural and methodological uncertainty, concerning model form and analytical choices, is handled by scenario analysis of coherent alternatives; and heterogeneity, genuine differences between subgroups, is examined by subgroup analysis. Matching the method to the source ensures each kind of uncertainty is explored appropriately, so together these techniques give a comprehensive assessment of a model's robustness.
Source: Briggs, Claxton & Sculpher 2006
What are the limitations of sensitivity analysis?
Sensitivity analysis depends on the ranges, distributions, and scenarios chosen, so it explores only the variations considered and can miss uncertainties not represented, and different methods address different sources, so no single technique captures all uncertainty. Deterministic methods miss interactions, probabilistic analysis misses structural uncertainty, and scenario analysis attaches no probabilities. These limitations mean a combination of methods is used, with the assumptions behind each made explicit, so that the overall assessment of uncertainty is as complete and honest as the evidence and methods allow.
Source: Briggs, Claxton & Sculpher 2006
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 29 Oct 2025
Content version: 1.0.0
Canonical Identity
- Term code
- HE-EM-UA-064
Stable URI · Machine-readable · Resolvable · CC BY 4.0