Systematic review of discrete choice experiments in multiple sclerosis

By Dr Edward Webb, Academic Unit of Health Economics (University of Leeds)

As part of research on how people make choices about treatments for Multiple Sclerosis led by Dr Ana Manzano, we have published a  systematic review of discrete choice experiments.

Multiple Sclerosis (MS) is a degenerative disease of the central nervous system, which can cause many varied symptoms such as problems with walking, balance, cognition and vision. While there is no cure, disease modifying treatments (DMTs) can help reduce attacks of symptoms and slow the gradual accumulation of disability in a type of MS called Relapsing Remitting MS. Many different DMTs exist, each with different risks and benefits. The CRIMSON  study is funded by the UK MS Society and is looking at how people with MS trade-off these risks and benefits when making decisions about which DMT to take, or whether to not take one at all.

I am a health economist, and my work focuses on how patients and health professionals make decisions. One way of doing this is through a survey technique called a discrete choice experiment (DCE). In a DCE survey, participants make a series of hypothetical choices based on 5-6 attributes. The levels of these attributes are varied in each question. For example, if you were doing a DCE about pizza, participants would make a series of choices between which hypothetical pizza they would choose. The pizzas would be characterised in terms of attributes such as size, toppings and cost. Statistical analysis of the responses would then reveal things such as which topping is most preferred, and what trade-offs people make between size and cost.

DCE is a good method to examine the preferences of people with MS for DMTs, as they can be characterised by attributes such as the reduction in the number of clinical attacks, side effects, or mode of administration. In the CRIMSON project, before designing our own DCE, we conducted a systematic review of DCE and similar studies on DMTs. Rather than focusing on their findings, we looked at how the studies were carried out.

We found that many studies replied on previous literature and clinicians’ views to develop the attributes in their survey. In the development process they did not tend to get input from the sort of patients who might answer the survey, for example through focus groups. This kind of qualitative work with the sort of patients who will answer your survey is vital for most DCEs. It ensures that researchers are capturing the most important part of decision making situations, that questions make sense to participants, and that attributes mean what researchers think they mean.

Choices about treatments for MS, like a lot of illnesses, involve weighing up risks and benefits. The exact benefits for an individual patient are unknown, and some treatments carry the risk side effects. However, most DCE studies did not explore these issues fully, nor did they consider how to best communicate risks to participants.

We also identified some neglected areas. For example, very few studies looked at choices in relation to reproduction. But this is an important as most people are first diagnosed with MS in their 20s and DMTs can have an impact on reproduction for both men and women. Patients are advised to stop taking many DMTs during conception, pregnancy and breastfeeding.

At the moment, the project team is designing our own DCE, and we’re carrying through many of the things we’ve learned from the systematic review. We did focus groups and interviews with almost 70 people with MS in order to develop our survey. We took advice from the literature about how to best communicate risk, and are illustrating probabilities visually in our survey using ideographs. We are also including questions in the survey which look at attributes of DMTs related to reproduction.

There are a lot more interesting and exciting results to come from the CRIMSON project!

Follow @ana__manzano  on twitter for updates.

Image: Photomicrograph of a demyelinating MS-Lesion

Author: Marvin 101

Licence: Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License