Univariate/One Way Sensitivity Analysis

Univariate or one-way sensitivity analysis allows a reviewer to assess the impact of changes in a single input parameter on the output results of an economic evaluation, most often those based on a model. This analysis evaluates the robustness of the result of variations in that parameter.

Key points about univariate/one-way sensitivity analysis include:

– Single Parameter Variation: The parameter of interest is variable between plausible extremes, ideally justified by a review of available evidence. Only one parameter is changed at a time, simplifying the analysis.

– Assessing Robustness: By varying one parameter, the analysis helps determine how sensitive the model’s results are to changes in that particular parameter. This helps identify which parameters have the most significant impact on the outcome.

– Simplest Form of Sensitivity Analysis: This method is straightforward since it does not account for correlations between different parameters.

Tornado Diagrams: Results of univariate sensitivity analyses for multiple parameters are often summarized using tornado diagrams. These diagrams visually display the range of output values resulting from variations in each parameter, with the most influential parameters at the top.


– If the cost of a new drug is varied from $50 to $150 per dose while other parameters remain constant, the impact on the incremental cost-effectiveness ratio (ICER) can be observed. If the ICER changes significantly, it indicates that the model is sensitive to changes in the drug cost.


– Easy to perform and interpret.

– Identifies key drivers of model outcomes.

– Provides clear insights into the uncertainty associated with individual parameters.


– Does not account for correlations between parameters.

– May not capture the combined effect of varying multiple parameters simultaneously.

Univariate sensitivity analysis is a fundamental tool in economic evaluations, helping to highlight the parameters that most influence the results and guiding further data collection and research efforts.