This article is best viewed in a PDF Format.
|
|
Practitioners do not have the time to read the wealth of information on new health technologies and interventions - including pharmaceuticals, surgical procedures, screening programmes and nutritional supplements - and so previously they often had little option but to adopt new technologies without evidence of their clinical effectiveness. There were also issues of lack of uptake of new technologies that had been demonstrated to be effective. NICE was established with a view to enhancing the NHS quality of patient care by providing clear and robust advice to practitioners.1
Overall, NICE has three main functions:
- Appraisal of new and existing health technologies
- Development of clinical guidelines
- Promotion of clinical audit and confidential enquiries.2
As well as clinical appraisal of new technologies, NICE also sets out to ensure value for money for the NHS. Its stated aim is to prioritise, rather than ration, and it is primarily concerned with the appropriate allocation of available resources. To this end, economic evaluations of health technologies are carried out alongside clinical evaluations.
Economic evaluation can be conducted from various perspectives, and although the identification and measurement of various costs is similar across the board, the methods by which consequences of health technologies are measured and valued can vary. A cost-effectiveness analysis would involve comparing health technologies in terms of cost per unit clinical outcome, for example unit change in systemic blood pressure, or mortality.3 A cost-benefit analysis, however, would measure both costs and consequences of the health technology in monetary terms, and compares 'doing something' with 'doing nothing' rather than comparing two methods of dealing with the same health issue.4 The cost-benefit analysis also attempts to value the wider benefits of a health technology, in other words, those that do not directly relate to the technology itself. For example, the reduction in time that a family member spends supporting a patient following cataract surgery. In this way, the cost-benefit analysis is based in welfare theory, and has a societal perspective.
The NICE perspective is different again, and involves cost-utility analysis. This form of analysis has a narrower perspective than cost-benefit analysis and involves use of a 'common outcome currency' to describe benefits and permit comparison between health technologies. The unit of 'currency' is the quality adjusted life year (QALY).5
The concept of the QALY was first introduced in 1968,6 and it can simultaneously capture the gain from reduced morbidity (quality gains), reduced mortality (quantity gains), and combine these into a single measure.
In Figure 1, an individual not receiving the health technology experiences deterioration in health-related quality of life that follows the lower path, and dies at time Death 1. However, if in receipt of the health technology, an individual would experience a slower deterioration, would live longer, and would die at time Death 2. The area between the two curves is the QALY gained by the health technology. The gain in QALYs can be divided into two sections section A is the QALY gained due to quality improvement and section B is the QALY gained due to quantity improvement.
To put the QALY concept into practice, it is necessary to apply quality weights to the health states in question. The quality weights are the scale for the vertical axis in Figure 1. Every possible health state is given a weight between 0 and 1, where 0 is equivalent to death, and 1 is equivalent to perfect health. The QALY weights for health states are based on preferences for health states so that more preferred health states are given greater weight and will be favoured in the analysis.9 So, for example, having a cold would be given greater weight than having pneumonia. The advantage of using 1 to represent perfect health is that the resulting QALY is then measured in units of 'perfect health years', such that one year in perfect health equals 1 QALY, and half a year in perfect health equals 0.5 QALY, and so on.
To compare one health technology with another, or with standard care, it is necessary to compare the additional costs of that technology with the additional benefits that it delivers. This is an incremental approach to economic analysis and involves calculation of the incremental cost effectiveness ratio (ICER), by taking the difference in cost between the health technology in question and the best alternative, and dividing that figure by the difference in outcome (eg QALYs):
ICER = (Cost A - Cost B) / (QALY A - QALY A)
Where A = health technology under investigation and B = the comparator, or best alternative.
The ICER is also known as the 'cost per QALY'. Interpretation of the ICER may be aided by use of the cost-effectiveness plane (Figure 2).
In Figure 2, the vertical axis represents the difference in cost between the health technology and its comparator, and the horizontal axis represents the difference in cost. The comparator could be an alternative technology, or the current 'standard care'. The diagram is used by plotting the difference in effect against the difference in cost. If this point lies within quadrant 2 or 4, then the decision is clear. For example, in quadrant 2, the health technology is both cheaper and more effective than the comparator, so it is said to dominate the comparator and will be accepted by the decision maker. In quadrant 4, the reverse is true. In quadrants 1 and 3, the choice depends upon the maximum cost-effectiveness ratio that is considered to be acceptable. In Figure 2 the dashed red line represents a hypothetical threshold ICER value. It has been suggested that the maximum threshold set by NICE is between £20,000 and £30,000,11 although NICE has not confirmed this.
Cost-effectiveness of the Age-Related Eye Disease Study formulation
United States perspective
A cost-utility analysis of the Age-Related Eye Disease Study (AREDS) formulation that contained antioxidants plus zinc12 has been carried out in a US healthcare setting in 2007.13 Table 1 shows a breakdown of the AREDS formulation.
AREDS participants were randomised to receive either (1) placebo, (2) antioxidants (vitamins C and E and beta-carotene), (3) zinc (and copper), or (4) a combination of antioxidants plus zinc (and copper). The combination of antioxidants plus zinc significantly retarded or prevented the development of reduced visual acuity and neovascular or atrophic AMD in people in AREDS categories 3 and 4 at baseline. Table 2 show AREDS categories.
The cost of prescribing the AREDS formulation for AMD diagnosed in the course of routine eye care was compared with the cost of no nutritional supplementation for patients older than 50 years. This was achieved using a model that simulated patients at age 50 years and followed them until death or age 100 years. The model tracked each patient's costs, incidence and subsequent progression of AMD, visual impairment caused by AMD, and QALYs. Visual impairment and blindness were based on the better-seeing eye, accounting for losses in visual acuity and contrast sensitivity. Upon initiation of the model, the numbers of people with the various stages of AMD were based on previously published prevalence data,14 and probabilities for patient progression were based on AREDS data.15 Patient disease states were based on the presence or absence of large drusen (>125 µm) or retinal pigment epithelium (RPE) abnormalities in one or both eyes. Patients were categorised into five states (Table 3).
Loss of visual acuity was programmed to occur only from advanced AMD. All patients diagnosed with AMD were assumed to have received medical treatment recommended by the American Academy of Ophthalmology's preferred practice patterns.16 Costs were assigned based on the Centres for Medicare & Medicaid Services fees schedules, and the model also tracked the use and costs of nursing home services.
Patient QALYs were calculated annually by multiplying the value of 1 minus the visual impairment-related health utility decrement (if any) by a background utility of 0.87 (the average health utility of people aged 50 years). Normal use of ophthalmic services was taken into account, as well as costs of vision-related medical care, vision-related nursing home placements, and total costs of the intervention. A discount rate of 3.5 per cent to both costs and QALYs was applied, which is the same as is applied in the UK. A discount rate is necessary to reflect the preference society has for delaying costs to a future date and enjoying benefits now.
Including the cost of the nutritional supplement, treatment costs per person in the sample increased from $583 to $721, and total costs, compared with no nutritional supplementation, increased by $88 per person. However, nursing home costs decreased from $266 to $217 per person. The ICER was $21,287 (£10,717) per QALY gained. Relative to no nutritional supplementation, the percentage of people who ever developed GA decreased from 10.1 per cent to 8.1 per cent, and the percentage who ever developed CNV in either eye decreased from 16.0 per cent to 13.0 per cent. The percentage of patients who ever developed visual impairment in either eye decreased from 21.4 per cent to 17.4 per cent, and the percentage who ever developed visual impairment in the better-eye decreased from 7.0 per cent to 5.8 per cent.
A limitation of this study is the fact that the risk reductions reported over five years in the AREDS is assumed to continue for as long as the patient takes the nutritional supplement.
European perspective
Another study published in 2004 modelled the cost per QALY of population-based screening and treatment of early AMD with the AREDS formulation, compared with no screening and no treatment.17 The project was based on the results of the AREDS randomised controlled trial12 and the Australian Blue Mountains Eye Study,18 but the authors believe that the model outputs can be generalised to European populations.
The decision analytic model assumed optometrist-based screening (performed using dilated fundoscopy) to identify the number of people with early AMD, who were then treated with the high dose zinc and antioxidant formulation. Incremental acuity events were used to calculate the QALYs associated with the treatment arm, and incremental neovascular events were used to calculate the incremental cost savings from the reduced number of laser and PDT treatments. Costs were converted to British pounds through purchasing power parity calculation this means that exchange rates were adjusted to eliminate differences in price levels between countries. A discount rate of 6 per cent was applied to both QALYs and costs in the UK a discount rate of 3.5 per cent is applied. Estimates of incidence and prevalence were taken from the Blue Mountains Eye Study,18 and the Australian Bureau of Statistics provided census-based population projection data19 and Australian life tables.20
In the 65 years and over age group, the cost per QALY was £22,722, decreasing to £18,948 when the savings accrued from reduced numbers of PDT treatments were taken into account.
Conclusions
Both studies described here involved cost-utility analyses, which are relevant to NICE decision-making but do not account for costs which are not directly related to the intervention. Examples of such costs might include the time burden on informal care givers, rehabilitation and training or lost productivity due to visual impairment. It should also be remembered that all the people within the models received guideline-recommended standards of care.
The difference in the cost per QALY, or ICER between the two studies (£10,717 for the US perspective and £22,722 for the European perspective) may be caused by the fact that the European model included the cost of separate 'screening' visits when in fact many people would attend their optometrist or ophthalmologist on a regular basis anyway. The US model was based on diagnosis and prescription of the AREDS formulation during routine care. The potential positive impact of a specific AMD screening programme on the detection of other ocular diseases such as diabetic retinopathy and glaucoma was not taken into account in the European model. The investigators of this model also suggested that the price of the AREDS formulation is likely to decrease should such a screening programme be established. Taking these factors into account is likely to lead to a reduction in the reported cost per QALY.
Both cost per QALY values are below the maximum threshold reported to be accepted by NICE. It has been argued that people at risk of AMD should be screened and that people who already have AMD in one eye should be treated with the AREDS formulation.17 The authors of the US-based study support the use of vitamin therapy for all indicated patients diagnosed with AMD and aged 50 years or older.13 Contraindications should be taken into account, such as smoking for beta-carotene.21 ?
References
A full list of references is available from the clinical editor: william.harvey@rbi.co.uk
Dr Hannah Bartlett is a lecturer in Aston University Department of Life Sciences and has no commercial association. This paper was supported by Bausch & Lomb