Incorporating the preferences of patients with rare diseases in health technology assessment

Health technology assessments (HTAs) are a crucial part of the journey to bring newly developed pharmaceuticals and medical devices to patients. HTAs include evidence from clinical studies, economic evaluations, and stakeholder perspectives that inform decisions on how healthcare resources are allocated. The aim is to help policymakers and healthcare providers make informed choices about the adoption and use of new health technologies to improve health outcomes and optimise the allocation of scarce resources.

In HTAs, a greater focus has been placed recently on patient preference information because it can offer an understanding of what matters to patients and reduce uncertainty about clinical benefits. Patient preference research involves generating data directly from patients about how they value various healthcare options. This is typically done by exploring how patients make trade-offs between hypothetical treatment choices based on the attributes of the treatment, such as effectiveness, side effects, convenience, and cost, to name a few. Including patient preference information in HTA can enhance patient-centred care to ensure that healthcare decisions reflect patients’ needs and desires.

Patient preference information is particularly salient in rare diseases, which can be defined as conditions that occur in fewer than 1 in 2000 people, with ultra-rare diseases affecting fewer than 1 in 50,000 people. Although individually uncommon, rare diseases collectively represent a large burden to society. Including the preferences of patients with rare disease is particularly important because their experiences and the challenges associated with their conditions tend to be unique and are often insufficiently captured in clinical studies. Patients with rare diseases may also value trade-offs between potential treatment benefits and their side effects or risks differently than patients with more common conditions; therefore, capturing their preferences can help decision-makers better balance the benefits and costs of potential treatments.

Several approaches can be employed to capture patient preference data. Qualitative methods can obtain rich data on what matters to patients. Quantitative research can be used to explore how much these core aspects matter to patients. However, there are challenges associated with generating quantitative preference evidence for rare diseases. Patient populations with a rare disease are, by their nature, relatively small, and small sample sizes can lead to underpowered studies that do not produce robust evidence. Additionally, there is a lack of fundamental knowledge about individual rare diseases, including their epidemiology and natural history. Furthermore, there is considerable heterogeneity within disease areas, such as variations in symptoms and disease characteristics.

The current gold standard approach to quantitative preference elicitation is a discrete choice experiment (DCE). DCEs can provide valuable information about the relative importance of different attributes related to a new treatment or diagnostic tool, risk-benefit trade-offs, willingness to pay, or even the relative expected uptake of different treatments. This approach is accepted by many regulatory and HTA bodies globally; however, DCEs are not designed to provide robust results with very small data samples. In the case of rare diseases, the DCE approach may not be the most appropriate choice, and an alternative might be recommended.

A number of approaches can be employed to generate quantitative patient preference evidence in small populations with rare diseases. Some examples of these include the following:

• Swing weighting: an approach that involves patients ranking, rating, and scoring improvements in different attributes
• Best-worst scaling: a ranking exercise to identify the extremes of patient preferences to explore what is most important to them
• Q methodology: a mixed quantitative and qualitative method that involves a ranking exercise to allow the identification of a range of viewpoint typologies among a sample of patients

Each of these alternative approaches can be used with small samples that make them preferable candidates for eliciting patient preference data in rare conditions. To learn more about these approaches, Putnam are offering a micro-tutorial on this topic at ISPOR EU in Barcelona. If you are interested in attending, please reach out to:

Sarah Hill
Director, HEOR

sarah.hill@putassoc.com