**Probabilistic Sensitivity Analysis**

Probabilistic Sensitivity Analysis (PSA) is a technique used in economic modeling to assess the robustness of model outputs in relation to the uncertainty of input parameters. PSA quantifies the level of confidence in the results by considering the variability and uncertainty inherent in the model inputs, which can originate from clinical trials, observational studies, or expert opinion.

**Key features of PSA include:**

**– Uncertainty in Input Parameters:** Recognizing that each input parameter value has an associated uncertainty, PSA uses distributions rather than single point estimates. These distributions are often derived from empirical data or expert consensus.

**– Representation of Parameters:** In PSA, input parameters are represented by probability distributions that reflect the range of possible values. For instance, a normal distribution might represent measures of effect such as hazard ratios or relative risk reductions, while a Weibull distribution could represent survival curves.

**– Random Sampling:** The PSA process involves random sampling from the defined distributions for each input parameter. This set of sampled values is then used to ‘run’ the model, generating outputs such as costs and health outcomes.

**– Iteration:** This process is repeated many times (typically between 1,000 and 10,000 iterations), producing a distribution of results. These results can then be graphed on a cost-effectiveness plane and analyzed to understand the range and likelihood of different outcomes.

**– Key Outputs:**

**– Distribution of Outputs:** The resulting distribution of outputs shows the variability in cost and health outcomes due to the uncertainty in input parameters.

**– Cost-Effectiveness Acceptability Curve (CEAC):** A CEAC can be generated to illustrate the probability that an intervention is cost-effective at various willingness-to-pay thresholds. This curve shows the proportion of iterations where the intervention meets or exceeds a given cost-effectiveness threshold.

PSA is critical in economic evaluations as it provides a comprehensive view of the uncertainty and helps decision-makers understand the confidence they can have in the model’s recommendations. It supports more informed and transparent decision-making in healthcare by demonstrating the potential range of outcomes and their associated probabilities.

Click HEOR TERMINOLOGY