Addressing Uncertainty in Health Economic Analysis for Insurers

Evaluation, Implications, and Ways to Reduce Uncertainty

Uncertainty is inherent in Health Economic Analysis (HEA). This uncertainty can come from a variety of sources, including data quality and availability, model structure and assumptions, and the general lack of knowledge about the future course of a disease or treatment. 

While there is no way to eliminate uncertainty in HEA completely, there are ways to reduce it. In this blog post, we will discuss some common sources of uncertainty in HEA and how they can be addressed.

What is uncertainty in health economic analysis?

Uncertainty in health economic analysis refers to the lack of complete information or knowledge about the costs, outcomes, and probabilities associated with healthcare interventions. There are two main types of uncertainties that can arise in health economic analysis:

1. Structural uncertainty. It relates to decisions about the model structure and assumptions.
2. Parameter uncertainty. It concerns the precision of the inputs used in the model.

Parameter uncertainty can be further categorized into one-way sensitivity analysis, multiway sensitivity analysis, and probabilistic sensitivity analysis. It is important to understand uncertainty in health economic analysis as it allows decision-makers to make informed decisions based on the limitations and potential biases of the analysis.

Types of uncertainty in health economic analysis

1. Parameters: The values of the inputs of the model

Parameters are numerical values that are used in health economic analysis to represent various aspects of a disease, treatment, or intervention. Examples of parameters include the probability of a disease event, the cost of treatment, and the quality of life associated with a health state.

Parameters are significant in modeling because they influence the final results of the analysis. However, there are different types of uncertainty related to parameters that can affect the accuracy and reliability of the model outputs:

1. Inter-individual variability
2. Measurement error
3. Parameter estimation uncertainty
4. Model structure uncertainty

2. Model uncertainty: Different values of the model

This uncertainty can stem from two sources: uncertainty relating to decisions about the structure of the model and uncertainty concerning the precision of the inputs used in the model. Structural uncertainty can be explored by examining how alternative scenarios affect the results, while parameter uncertainty can be characterized using sensitivity analysis.

3. Input prices: Whether the costs of inputs are accurate

Uncertainty is a common challenge in health economic analysis related to input prices. This can be due to estimating uncertainty and policy uncertainty. Estimating uncertainty occurs when there is limited research evidence available or when a policy affects prices through several interconnected channels.

4. Qualitative information: Whether the qualitative information is included in the analysis

Qualitative information plays a crucial role in health economic analysis, but its inclusion can introduce various types of uncertainty. One such uncertainty is the subjectivity of data collection and interpretation. Qualitative data often rely on individual experiences and perspectives, making it difficult to standardize and generalize results.

Another uncertainty is the potential for bias, as qualitative research may be influenced by the researcher’s own beliefs and values. Additionally, the lack of standardization in qualitative research can lead to inconsistencies in data collection and analysis.

5. Methodology: Whether the analysis complies with the standards

Methodological uncertainty refers to the uncertainty surrounding the choice of methodology used to conduct the analysis. For example, the use of different discount rates may lead to different cost-effectiveness results.

6. Data sets: Whether the datasets are of sufficient quality

Insufficient quality of datasets can lead to various types of uncertainty in health economic analysis. These uncertainties include methodological, data, and parameter uncertainties. Methodological uncertainty arises from the use of inappropriate methods, while data uncertainty arises from the use of incomplete or unreliable data.

Related: Data Mining in HEOR: Types, Steps, Uses, Big Data, and AI

How can we identify and assess the implications of uncertainty in health economic analyses?

Step 1: Identify the source of uncertainty in the health economic analysis

As insurers make decisions on which healthcare interventions to cover, it is essential to have a clear understanding of the economic impact of these interventions:

1. Examine the structure of the model.
2. Characterize parameter uncertainty.
3. Perform multiway sensitivity analysis.
4. Use probabilistic sensitivity analysis (PSA).

Step 2: Select appropriate estimation methods to address the underlying uncertainty

To select appropriate estimation methods, the following steps should be followed:

1. Consider the type of uncertainty, whether it is structural or parameter uncertainty, and use appropriate methods to address it.
2. Identify appropriate statistical models that can capture the uncertainty and variability in the data.
3. Assess the validity of assumptions made in the model and adjust them if necessary.
4. Use sensitivity analyses to evaluate the robustness of the results and identify the parameters that have the most significant impact on the outcomes.

Step 3: Implement an analysis strategy that incorporates the selected estimation method

It is crucial to implement an effective analysis strategy that considers the selected estimation method in order to identify and assess the implications of uncertainty in health economic analyses.

Step 4: Execute the analysis under real-world conditions

Uncertainty is a major concern in health economic analyses since it can impact the validity and reliability of the results. Therefore, it is crucial to identify and assess the implications of uncertainty while executing the analysis under real-world conditions.

Related: Common Data Sources in HEOR

Step 5: Review, evaluate, and assess different analyses and models

When conducting health economic analyses, it is important to review and evaluate different analyses and models to ensure that the results are reliable and informative:

1. Consider the model assumptions
2. Conduct sensitivity analysis
3. Incorporate expert opinion
4. Document limitations

Step 6: Conduct post-analysis reviews to assess the accuracy of the results

Post-analysis reviews play a critical role in health economic analyses as they help to identify and assess the implications of uncertainty. The accuracy of results in health economic analyses can be affected by several factors, including the methodology used and the quality of data available. Therefore, conducting post-analysis reviews is essential to ensure that the results are accurate and reliable.

Conclusion

Addressing uncertainty in health economic analysis is vital for insurers to make informed decisions about coverage, pricing, and resource allocation. By understanding the various sources of uncertainty, such as data limitations, model assumptions, parameter estimation, and generalizability, insurers can implement strategies to mitigate uncertainty effectively.

Sensitivity analyses, probabilistic modeling, incorporation of real-world data, external validation, and expert consultation are valuable approaches. By embracing these strategies, insurers can enhance the robustness and reliability of their economic analyses, leading to more informed and confident decision-making in the ever-evolving landscape of healthcare.