You don't need to be signed in to read BMJ Group Blogs, but you can register here to receive updates about other BMJ Group products and services via our Group site.

James Raftery: Moving to value based pricing—adjusting costs

20 Nov, 12 | by BMJ

With the introduction of value based pricing scheduled to apply to new drugs when the current prescription pricing regulation system expires in 2014, the Department of Health has been working on the details of how it might apply. This blog reports on an invited workshop held by NICE on how the cost impact of health technologies might be extended to reflect a societal (termed “wider social benefits”) as opposed to the (current NICE) NHS and personal social care perspective. Chatham House rules applied (the discussion can be reported but not attributed).

The current cost effectiveness analysis was characterised as quality adjusted life years (QALY) maximisation, subject to a) all QALYs being the same, and b) the costs linked only to the care the patient receives.  The challenge of VBP was seen as extending this to include the costs of care to family and other carers and users of public services. Costs were seen as having two dimensions: production and consumption of resources. A patient able to return to paid or unpaid work sooner than otherwise due to a new treatment would increase production, earning more so lowering net cost. Such a patient might also require less formal and informal care by family and other members of society, again lowering the net cost of treatment.

The attempt to operationalise the measurement of these changes relied mainly on how improved quality of life due to a new medical treatment might influence production and consumption.  Operating on a general as opposed to a case by case basis (with risks of opacity and heterogeneity), the aim was to provide a ready reckoner (or look-up table) that would build on the present system  by adding the resource implications of  wider social benefits under the seven headings shown in Table 1.

The first headings had to do with work, both paid and unpaid. A valiant attempt was made to estimate the gains due faster return to work through linking quality of life to the “sick rate.” This relied on a newly commissioned survey of hospital patients—health improvement and patient outcome (HIPO)— which contained data on quality of life, and days of work, and taking into account age, gender, and disease (at international classification of disease chapter level). The same approach applied to both paid and unpaid labour. It was valiant because major assumptions were required to link the survey results to the actual (much lower) sick rate. Work was reported to be continuing on this. The monetary gain of increased hours at work used the “human capital” as opposed to the “frictional” approach. The discussion of how work effects were being estimated was probably more critical than the other headings but that may be because it was first. Difficulties included lack of relevant data, the use of the human capital as opposed to the frictional method given current unemployment levels, and the linked assumption of perfect labour (and indeed other) markets.

Moving on from effects on work, changes in the consumption of informal care (including child care, informal health related care, and other unpaid production such as domestic work) were also linked to quality of life, based on the same HIPO survey which asked patients about such consumption in the previous six weeks. This showed the expected gradients. Discussion focussed on the link between ill health and child care. Would improved quality of life lead to reduced informal child care requirements (or vice versa) and should this be valued at replacement cost. This echoed debates in social security reform to do with whether children should be considered as consumption or production (really!). The approach attempted to distinguish between the costs of caring for children when ill and the benefits of being with them when well, something all parents, let alone grandparents will have experienced.

Changes in the consumption of formal care were based on estimates from a variety of sources showing the probability of receiving (as opposed to requiring) such care could be  related to quality of life and disease. Discussion focused on the importance of dementia in care in formal residential care, challenging the need for a disease by disease approach. The implication could be a premium for treatments which made patients with dementia less likely to require such care.

The effects of improved treatments on other members of society included personal paid consumption (housing, food, travel), personal unpaid consumption (such as domestic work) and consumption of government services. The attempt to link these to changes in quality of life proved the most challenging with only data linking each to age available. It was not clear it was plausible to link these to quality of life.

The overall approach is summarised in Table 1, which indicates that quality of life could be linked to only 4 of the 7 headings, with the remainder varying only by age.

Table 1:

Wider Social Benefits: list of elements and drivers
Changes in Driven by quality of life changing the:
Paid labour Sick rate
Unpaid labour Sick rate
Informal care consumption Care received
Formal care consumption Probability of requiring care
Personal paid consumption Age effect only
Personal unpaid consumption Age effect only
Consumption of government services Age effect only

What to make of this?

First, the approach of providing average across-the-board estimates of the resource implications of wider social benefits based the improved quality of life due to treatment is  ……   courageous (As Sir Humphrey defined it in Yes Minister).

Second, the attempt to include age, gender, and disease implies scope for a premium for each of these, provided robust estimates can be made. How this might fit with equality legislation remains to be  explored. Ministers would have to consider mitigating measures.

Third, the approach relies heavily on microeconomic theory, which has a range of relevant if esoteric concepts, but whose usefulness assumes perfect (or working) markets, rational maximising behaviour, and perfect (or very good) information.

The discussion threw up some other big issues. One was that of the opportunity cost to the NHS of NICE recommending new health technologies needed to be measured in terms of what treatments are displaced. Although this was widely agreed, it had not been made explicit. The proposed new approach appeared to assume it should apply and was being pursued in separate work. However, the cost of such displaced services should be similarly adjusted for “wider social benefits,” raising further complications, which it was planned to address.

Another had to do with the choice of relevant factors to go beyond age, gender, and disease, such as social class or socioeconomic group.

Finally the point was made repeatedly that if value based pricing is to proceed as planned, estimates, of the sort that were presented are needed. Unless better alternatives were suggested, this was the way forward. This left me wondering if those who clamoured for value based pricing knew what complications it would unleash.

James Raftery is a health economist with several decades experience of the NHS. He is Professor of Health Technology Assessment at Southampton University. A keen “NICE-watcher,” he has provided economic input to technical assessment reports for NICE but has never been a member of any of its committees. The opinions expressed here are his personal views.

By submitting your comment you agree to adhere to these terms and conditions
You can follow any responses to this entry through the RSS 2.0 feed.
BMJ blogs homepage

BMJ.com

Helping doctors make better decisions. Visit site



Creative Comms logo

Latest from BMJ.com

Latest from BMJ.com

Latest from BMJ.com podcasts

Latest from BMJ.com podcasts

Blogs linking here

Blogs linking here