Concept Architecture
Concept
Theoretically, One-Way Sensitivity Analysis is a deterministic sensitivity analysis method in which one model parameter is varied across a plausible range while all other parameters remain fixed at their base-case values. The method evaluates how uncertainty in a single input influences model outputs and identifies parameters to which results are most sensitive. In health economics, it is a standard approach for assessing parameter uncertainty within decision-analytic models.
Mathematically, one-way sensitivity analysis examines the relationship between a model output and a single varying parameter while holding all remaining parameters constant. The model is repeatedly evaluated over the specified range of the selected parameter, producing a deterministic response function. The magnitude of the resulting change quantifies the sensitivity of the outcome to that parameter.
In practice, one-way sensitivity analysis is routinely performed for costs, utilities, probabilities, treatment effects and discount rates in cost-effectiveness models. Results are commonly presented using tornado diagrams or threshold analyses to identify influential parameters and evaluate the robustness of reimbursement or policy decisions.
Purpose
Used to evaluate the influence of uncertainty in a single model parameter on economic outcomes, thereby identifying influential inputs and assessing the robustness of health economic conclusions.
Mathematical Formulae
Primary Formula
y = f(x?)
where:
- x? = parameter varied across its plausible range
- all other parameters remain fixed at their base-case values
Supporting Formulae
Sensitivity change:
?y = y(x?,max) ? y(x?,min)
Percentage change:
%?y = [(y? ? y?) / y?] ? 100
Incremental Net Monetary Benefit:
INMB = ? ? ?E ? ?C
Related Mathematical Methods
- Deterministic Sensitivity Analysis
- Tornado Analysis
- Threshold Analysis
- Incremental Cost-Effectiveness Analysis
- Decision Analysis
- Scenario Analysis
Example
A cost-effectiveness model estimates an ICER of �24,000 per QALY.
The treatment cost is varied from �8,000 to �12,000 while all other parameters remain unchanged.
| Treatment Cost | ICER |
|---|---|
| �8,000 | �20,500/QALY |
| �10,000 | �24,000/QALY |
| �12,000 | �28,100/QALY |
The analysis demonstrates that the treatment cost is a major driver of cost-effectiveness and may warrant further evidence collection.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| Data Table | One-variable Data Table | Evaluate model outputs across a parameter range |
| IF | =IF(ICER<=30000,"Cost-Effective","Not Cost-Effective") | Assess decision changes as the parameter varies |
| INDEX | =INDEX(OutputRange,MATCH(ParameterValue,ParameterRange,0)) | Retrieve model results for each parameter value |
| MATCH | =MATCH(InputValue,ParameterRange,1) | Locate parameter values during sensitivity analysis |
| MIN / MAX | =MAX(ResultRange)-MIN(ResultRange) | Quantify the impact of parameter variation |
VBA (Optional)
A VBA procedure can automatically vary a selected model parameter across predefined values, recalculate the model and generate tornado diagram input tables.
Sources
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
- Drummond MF, et al. Methods for the Economic Evaluation of Health Care Programmes.
- ISPOR-SMDM Modeling Good Research Practices Task Force Reports.
- NICE. Health Technology Evaluation Manual.
Related Concepts (2)
Library
Tools & Resources
1
SAVI — Sheffield Accelerated Value of Information — Mark Strong, Jeremy Oakley & Penny Breeze (University of Sheffield), Web application ed., 2024 (University of Sheffield)
A free, open-access web calculator that computes value-of-information measures (EVPI, partial EVPI/EVPPI and EVSI) directly from a model’s probabilistic sensitivity analysis output — no need to re-run the model. Also reports payer strategy-specific and uncertainty burden.
Web Tool (R Shiny)View source →
Frequently Asked Questions (6)
What is one-way sensitivity?
A sensitivity analysis in which a single input parameter is varied across a range while all others are held at their base case values.
Source: Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press; 2006. doi:10.1093/oso/9780198526629.001.0001.
What does one-way sensitivity hold constant?
One-way sensitivity varies a single input across its plausible range while every other input is held at its base-case value, so the change in the result can be attributed entirely to that one input. Repeating this for each input in turn shows which ones move the result most. Its simplicity is also its weakness, since holding all others fixed means it cannot reveal how inputs behave in combination or reflect their joint uncertainty. It isolates one input at a time. Briggs and colleagues (2006) describe it.
Source: Briggs et al. 2006
How is one-way sensitivity performed?
One-way sensitivity is performed by selecting a parameter, changing it across a plausible range, such as from a low to a high value, while keeping all other inputs at their base-case values, and recording how the result changes. This is repeated for each parameter of interest. The resulting ranges of the output are often displayed together in a tornado diagram, which orders the parameters by how much each moves the result, highlighting the inputs to which the result is most sensitive individually.
Source: Briggs, Claxton & Sculpher 2006
What are the advantages of one-way sensitivity?
One-way sensitivity is simple to perform and interpret, clearly showing the effect of each parameter individually, which makes it useful for identifying the inputs that most influence the result and for communicating those drivers transparently. Because only one input changes at a time, the cause of any change in the result is unambiguous. It requires only the parameter ranges, not full distributions. These advantages make one-way sensitivity a common first step in exploring how a result depends on its individual inputs.
Source: Drummond et al. 2015
What are the limitations of one-way sensitivity?
One-way sensitivity varies only one parameter at a time, so it cannot capture the combined effect of several inputs changing together or their interactions, potentially understating uncertainty when parameters act jointly. Its results depend on the chosen ranges, and holding other inputs fixed at base case ignores their uncertainty. It gives no probability for the result. These limitations mean one-way sensitivity is complemented by multi-way analysis for combined effects and by probabilistic sensitivity analysis, which varies all inputs together according to their distributions.
Source: Briggs, Claxton & Sculpher 2006
How does one-way sensitivity relate to a tornado diagram?
One-way sensitivity relates to a tornado diagram in that the diagram displays the results of many one-way analyses together: for each parameter, a bar shows the range over which the result moves as that parameter is varied between its low and high values, and the bars are ordered from widest to narrowest, giving a tornado shape. The widest bars mark the parameters to which the result is most sensitive individually. The tornado diagram thus summarises one-way sensitivity across all inputs in a single, ordered picture.
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
- Persistent URI
- https://healtheconomics.wiki/concept/one-way-sensitivity
- Term code
- HE-EM-UA-047
Stable URI · Machine-readable · Resolvable · CC BY 4.0