Metandien> research costs → invest in further studies; else, proceed with current evidence. |
| **6. Select Appropriate Modeling Approach** | – Scenario analysis if variability limited.
– Probabilistic sensitivity analysis if uncertainty high.
– Monte Carlo simulation for risk distribution.
– Multi-attribute utility theory for complex trade-offs. | Use the model that best captures uncertainty while remaining transparent and reproducible. |
| **7. Validate, Document, & Communicate** | – Cross‑check results against known benchmarks.
– Provide clear assumptions and limitations.
– Tailor presentation to stakeholders (policy makers vs clinicians). | Ensures credibility and facilitates informed decision making. |

---

### How to Apply This in Practice

1. **Gather Data:** Collect the most recent, high‑quality evidence on effectiveness, costs, safety, patient preferences, and any relevant contextual factors.

2. **Characterize Uncertainty:** Use sensitivity analysis (deterministic or probabilistic), scenario building, or Bayesian updating as appropriate for the problem at hand.

3. **Choose an Approach:**
- If you need a quick estimate and data are limited → simple cost‑effectiveness ratio or decision tree.
- If there is enough data to model long‑term outcomes and multiple uncertainties → a Markov model or microsimulation with probabilistic sensitivity analysis.
- For complex interventions or when patient heterogeneity matters → individual‑level simulation.

4. **Interpret Results:** Present the key findings (e.g., incremental cost per QALY, probability of cost‑effectiveness) along with uncertainty and assumptions. Discuss robustness via scenario analyses.

5. **Communicate Clearly:** Use visual aids such as CEACs or tornado diagrams to convey uncertainty to decision makers who may not be familiar with modeling jargon.

By following these structured steps, you can determine the appropriate model for your specific healthcare cost‑effectiveness problem, ensuring that the chosen approach accurately reflects the clinical and economic realities of the intervention under study.