Uncertainty is always present when assessing the introduction of a new medicine
From the perspective of decision-making, it is important that a health economic evaluation conducted on a pharmacotherapy also clearly brings up the uncertainties related to the assessment.
When assessing the expected benefits and costs of a new medicine about to enter the market, there is always uncertainty involved in the economic assessment. That is why it is important that the health economic evaluation conducted on a pharmacotherapy clearly also describes the key sources of uncertainty.
“A good and high-quality health economic evaluation brings up the uncertainties related to the structure and parameters of the health economic model and to the methods of assessment and how these impact the results of the assessment”, stated Pharmacoeconomist Piia Rannanheimo from the Finnish Medicines Agency Fimea.
The uncertainty related to the health economic model is a very central issue from the perspective of a decision-maker.
“Companies could always pay more attention precisely on the reporting of uncertainty, which is of course not always easy.”
The focus of Fimea’s assessment of pharmacotherapies (HTA) is on the assessment of new hospital-only medicinal products. If they so wish, a company may provide a cost-effectiveness analysis for the use of Fimea to form part of the knowledge base of the assessment. The purpose is to produce an assessment on the average costs and health impacts of a new therapy in terms of its treatment alternatives. In addition, it should be assessed whether the cost-effectiveness of the therapy differs in different patient groups.
As part of the cost-effectiveness analysis, companies also provide Fimea with an IT application, usually an Excel file, on the basis of which the model of the health economic evaluation has been constructed. It helps Fimea’s experts assess themselves the quality of the work.
“We want to review, assess and test by ourselves what happens if one changes assumptions in the model. We also examine from a macro perspective the mathematic accuracy and technical functionality, but above all, we want to identify the issues that have the larges impact on the result, such as incremental cost-effectiveness rate.”
The more personalised treatment and smaller patient populations, the thinner the clinical evidence of a new medicinal product often is. That means that also the preparation of a statement related to the reimbursement or inclusion in the service supply of a product involves more decision-making uncertainty.
Assessment of the evidence of a product is not, according to Rannanheimo, more difficult in a small or large patient population, since the evidence is always assessed on the basis of available data. In contrast, decision-making is more challenging, because there are many different perspectives at stake.
According to Rannanheimo, decision-making can be challenging, for example, with products for patient groups having a major therapeutic need and no good treatment alternatives available.
“This is the situation when the therapy is indicated for a serious disease with a poor prognosis and promising and hopeful results have been reported on the effectiveness of a new therapy. At the same time, the follow-up periods of the studies are still short, and data has not been gathered on the duration of response or longer-term treatment results. However, the cost of the treatment and the price of the products area also often high in relation to the expected benefit of the therapy based on the modelling. This kind of a combination is difficult from the perspective of decision-making and assessment of a reasonable price.”
Significance of local data
According to Rannanheimo’s experience, domestic data is always used in health economic reviews if it is possible. She says that a health economic model should describe, for example, the background risk, age and progression of a disease in a local patient population, if geographical variation has been demonstrated in these.
Data on the utility and quality of life, use of resources and price information are often also derived from local data.
“The use of local data depends on which data is available and what is the assumption of the variation between different countries.”
More specific social and healthcare data would support the decision-maker
In the introduction phase of a medicinal product, the knowledge basis available to a decision-maker is, according to Rannanheimo, always preliminary and involves uncertainty. She hopes that there was more social and healthcare data, which could help describe the current treatment praxis, the patient population being treated and the use of resources.
“The personalised therapies become, I think the smaller the amount of data on a certain specific population.”
Cancer therapy and cancer medicines are a good example of this sort of a situation. There is currently available data to support decision-making, for example, on the patients with a certain type of breast cancer and their treatment.
“When one needs more specific data on patients that have already received certain treatment for breast cancer and are being treated, for example, with third-line therapy, that data is difficult to find in the social and healthcare registry. However, new medicinal products are targeted towards these types of specifically limited patient populations.”
Rannanheimo hopes that the development of quality registers could contribute to complementing the knowledge basis needed in the controlled introduction of new medicinal products. At the same time, she wonders how detailed data can be included in those.
“Finland has a strong principle of one-time registry entry, and the same data serves many different purposes. The needs in the assessment of pharmacotherapies compete with many other purposes of social and healthcare data and it may not be realistic to expect that quality registers would include sufficiently specific data for decision-making related to the controlled introduction of new medicinal products.”