The most popular Quality of Life (QoL) measure used in the UK is EQ-5D – a framework used by NICE, the body in charge of funding decisions for the NHS. EQ-5D is a standardized instrument developed by the EuroQol Group as a measure of health-related quality of life that can be used for a wide range of health conditions and treatments.

EQ-5D takes the form of a simple questionnaire which comprises five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. The user completes the questionnaire, and the result gives us their ‘Health State’, a five-digit number which represents their health at that moment. 

Given we have five dimensions and five responses, we have a total of 55 = 3125 unique health states. Consider the health state 23154, this represents:

  •       I have slight problems in walking about
  •       I have moderate problems washing or drying myself
  •       I have no problems doing my usual activities
  •       I have extreme pain or discomfort
  •       I am severely anxious or depressed

Now we have our health state domain we need to apply some utility preferences over the states. In short, we need a system to compare health states. In order to understand whether moving from 12345 to 54321 represents a QoL improvement, we require an evaluation of how society comparatively values the five health dimensions.

As we want to use these preferences for public health decisions, we need to find aggregate population preferences, and this requires us to use a representative sample estimator. It is important to note here that preferences across societies are not homogeneous, Germans have a different set of preferences to the British, hence you are restricted to your country’s data set when transforming the health state into a quality of life index value. So our mapping function takes a health state as an input and outputs an indexed value (between 0 and 1, with 1 representing perfect health or health state 11111) which represents society’s aggregated preference for that particular health state.

In order to establish societies preferences for certain health states, that is to identify the mapping function, a team of researchers runs numerous tests with a sample group, and these results are extrapolated to form a population model. The Time Trade Off game is the most common method and is relatively simplistic in its design.

We ask our sample group to make a decision between two theoretical outcomes, A or B, an example is given below. Option A is X years in perfect health, and option B is 10 years of a particular health state, in this case 23154. 

With a value of X =10, it is assured that the preference will be for choice A. The question is repeated with continuously decreasing values of X, until eventually we reach a point of inflexion and the preference changes to B. We can deduce from this the existence of a value of X where the participant is indifferent between both A and B. The table below shows some mock data, where we can see that at X=5 the participant is indifferent. From this we can make the conclusion that this participant values, at this moment, living for 5 years in perfect health the same as living for 10 years in the health state 23154 and we make the claim that they must then have identical QALY scores. 

This exercise would be repeated across the whole sample group and with a range of different health states. Models are then used to ‘fill in the gaps’ until a complete mapping from every health state to an index value is identified. Now we have this mapping, we are able to take a user’s EQ-5D responses, generate a health state, transform this into an indexed health value and then use this to calculate QALY’s uplift for a healthcare evaluation. These values can then be used for economic analysis and can inform evidence based decision making. Below we will explore how these scores are applied to real life situations.

Consider the development of a new cancer treatment (Treatment N), and a decision has to be made as to whether the treatment should be funded as part of a public health system. There are two factors we must consider, it’s cost, and it’s incremental QALY benefit when compared to the existing treatment (Treatment E).

This cost/QALY (the metric ICER – Incremental Cost Effectiveness ratio – is often used with generally the same meaning) is what will be considered when making the decision to fund the treatment or not. The National Institute for Health and Care Excellence (NICE), the body in charge of these decisions in the UK, have a general rule that any intervention must provide a cost/QALY of between £20,000 – £30,000. However, this is not a hard limit and a higher cost/QALY is often accepted in instances of palliative care (around £50,000). A third threshold of £100,000 exists for ultra-orphan drugs, in order to incentivise the development of treatments for rare conditions. 

The NICE guidelines on CBPM (cannabis-based products for medicinal use) provides a key insight into how important these metrics are, as they state (for chronic pain): 

“Using THC:CBD spray, which is the cheapest CBPM with a publicly available price, the model produced an incremental cost-effectiveness ratio (ICER) of over £150,000/QALY gained over the standard of care, a value far higher than the commonly accepted decision threshold of £20,000-£30,000/QALY gained”

Although to proponents of CBPM’s this may seem damning – now that we understand these figures we can unpack them and in fact find areas of positivity. The first obvious point is that by the existence of a cost/QALY score, it is unequivocal that the use of CBPM results in an increase in QALY’s and hence quality of life. For NICE to acknowledge that cannabis is an effective medicine – considering until November 2018 cannabis was a Schedule 1 substance with no medicinal value – is a huge step in the right direction. The guidelines state that although cannabis is an effective medicine, it is not cost effective when using a pharmaceutical THC:CBD spray. From our understanding of QALY mechanics we can conclude that the CBPM’s would have to be around 6 times less expensive, or around 8 times more effective (or some combination of both) to bring the cost/QALY into the threshold accepted by NICE.

This leaves us with two clear objectives, firstly, broaden the evidence base for CBPM’s, identifying the groups of patients that stand to benefit the most from CBPM’s, and secondly, provide a market structure where CBPM’s can be provided at an affordable and cost effective price.