Valuing health states in practice

There are several different techniques available for valuing health states. Here we’re going to look at one called the time trade-off method. This is an exercise where we ask a person to consider a particular health state that we want them to value, and we present them with a choice between two alternatives– choice A and choice B. Let’s see an example. Here is a health state taken from the EQ-5D, five level version. Choice B involves living in this health state for a certain number of years– let’s say 10 years. Or you could choose choice A, which is to live in full health for a lesser number of years– let’s say 6 years. What would you choose?

To find out the value an individual places on this health state, we vary the time lived in the full health state until the respondent is indifferent. That is, they can’t decide between the two choices. We can then calculate the value of the health state as X divided by T. So for our example, if we were indifferent, or couldn’t choose between choice A and choice B, the value for this health state will be 6 divided by 10, which is 0.6.

Researchers can value a whole number of different health states from a descriptive system– however, as there are usually so many health states within a descriptive system, researchers typically value a selection of them, and undertake modelling to predict values for all the health states. The valuation tasks, such as the time trade-off we have just seen, could be done by post, over the internet in an online survey, or in a one to one interview setting. The advantage of a one to one interview setting is that the researcher is present to help explain the task, and to help with any queries, as these can be quite tricky tasks to do.

One way in which we can make it easier for our respondent is to use a prop board. This allows the respondent to see the choices displayed visually. Have a look at the following clip of one of our researchers here at the University of Sheffield carrying out a time trade-off interview on a health state from the EQ-5D-5L. As you can see, these choices can be quite challenging to think about, especially when considering a health state you may not have experienced yourself. We have another health state here. Could you please read this one out loud? ‘Unable to walk about’, ‘moderate problems washing or dressing yourself’, ‘slight problems doing your usual activities’, ‘extreme pain or discomfort’, ‘slightly anxious or depressed’. OK.

So would you prefer to die immediately in life A, or live for 10 years in the health state you just read out, or do you think that they’re the same? 10 years in life B. OK.

So would you prefer to live for 5 years in full health, life A, or 10 years in the health state, life B, or are they the same? I’m going to ask a question, and you’re probably going to say that it’s up to me to decide, but ‘extreme pain or discomfort’ – can you take any medication for the pain? We know this is a difficult one for people to imagine, but we try and say that there will be no relief or treatment for the pain. Well, that does at least make it black and white. So that does make it easier for me to decide now. Because I think extreme pain or discomfort would just be horrific.

So 5 years of life A. OK, so would you prefer four years of full health, or 10 years of the health state? Or are they the same? Four years of life A. Four years of life A. So would you prefer three years of full health, or 10 years of the health state? Or are they the same?

Extreme pain is– that’s nasty. For 10 whole years, as well. Yeah.

Still life A. Still life A. OK, so would you prefer two years in life A, two years in full health, or 10 years in the health state? Or are they about the same?

And you’re going to die at the end of both of them, aren’t you? Yes. Two years followed by death, or 10 years followed by death. And you’re in extreme pain or discomfort.

Life A. Still life A. Yeah. I’m coming right down on this one. So would you prefer one year in life A, and then followed by death, or 10 years in the health state, followed by death, or are they about the same? This is tough. Because what you’re doing is putting a value on people’s life who are living with pain, and perhaps you do get more used to it. But it feels, at the moment, as though I’m kind of condemning somebody to death because they’re in pain. Yeah, this is all about personal preferences as well, towards different types of health states.

And also, we know that it’s hard to imagine living in that health state for 10 years with extreme pain without imaging people adapting to it as well. So these are all parts of the process of valuing these health states, and your personal preferences with different health dimensions as well. Right, so that’s true, actually. So people do live with extreme pain, don’t they? And they perhaps learn to manage it, and learn techniques, —- ah, there’s no treatments, though, are there? There’s not meant to be any treatment. We’re meant to imagine that this is without relief or treatment, but we understand that it’s difficult to imagine these health states without imagining people adapting to them.

So, for example, adapting to having any of these problems would be something that people hypothetically imagine as part of this task. Although we try to make it just this health state for 10 years, without any relief or treatment. OK. I’m going to stick to one year, then, I think, at this point. Life– they’re about the same. OK. So one year in full health is about the same as 10 years with being ‘unable to walk about’, having ‘moderate problems washing or dressing yourself’, having ‘slight problems doing your usual activities’, having ‘extreme pain or discomfort’, and being ‘slightly anxious or depressed’. Yes. OK. Thank you.