« Was Ricardo right? | Main | Immigration and cheap preferences »

June 14, 2007


Igor Belanov

You're coming close to the sort of philistine argument that people have that says hospitals shouldn't spend money on art projects because it should all be going into care. You've descended into the emotivism you so often criticise.


Igor, IMO if people are dying or going blind due to unaffordable treatment then the government should spend NO MONEY on art or sports at all.

Igor Belanov

So what about research into incurable diseases? Should we stop treating ingrown toenails and divert all attention to developing a cure for cancer?

I think Trident should go a long time before culture and sports myself.


Would there not also be a case then for spending vastly (and I mean vastly) more on overseas aid?


The mis-allocation of spending is surely not the fault of NICE, which has no choice but to work within the constraints set by the Department of Health. It's not clear how abolishing NICE would lead to any change in the health budget; it might just lead to the same budget being spent in a less effective way.

Can you say more about the case for abolition of NICE?


Igor - I don't think I'm being emotivist. I just suspect that most people's utility function is such that they'd rather their taxes were spent on insuring them against partial blindness than on some other aspects of government activity. But the actual allocation of spending doesn't reflect this preference. And there's no good mechanism in place to ensure it does.
Gdr - you're right. I'm not questioning how good a job Nice does, given its guidelines.
The case for abolition is one of honesty. If government spending is irrational, let's see it as such, rather than pretend otherwise.
An alternative - a better one - would be to widen Nice's remit to cover all Health Department spending. So it can assess the cost-benefit case of a spin doctor's salary against (say) 3 courses of Lucentis.


The money needed for buying the drug for those poor souls who're going blind could easily be raised by selling tickets and TV rights for the public execution of Toni.


Chris, isn't there a problem of bounded rationality here? On the one hand, it may be possible given the expertise that actually exists to optimise a fixed level of expenditure across different types of treatments, as Nice does. On the other hand, it might be impossible to rationally optimise those decisions when all possible other ways of spending that money are taken into account. The problem is that to optimise across different types of treatment you only need expertise in different types of treatment, to optimise across all ways of spending money you need to know about everything.

William McIlhagga

What Dan said.

The problem of optimizing across every variable is eased by a) allocating money to broad areas e.g NHS then b) optimizing within the area. NICE is working in b), but you criticize it for not working on a). It's Gordo's fault that he spends too much on other things & not enough on health. NICE is just caught doing the best they can (which is a lot better than exposing the irrationality of spending decisions).

In addition, you're probably lapping up a hype produced by newspapers, who may not know just how effective Lucentis really is, but are ready to use it as a stick to beat their chosen enemy.

Maynard Handley

Igor, IMO if people are dying or going blind due to unaffordable treatment then the government should spend NO MONEY on art or sports at all.

Oh don't be stupid. People are, for the foreseeable future, always going to die, and it is always going to be the case that spending more will keep them alive a little longer. Should society spend $1million on heart/lung/liver transplants for someone who will then die three days later from a fscked up brain?

I have to agree. The post is
(a) a mindless appeal to emotion
(b) blaming the wrong person --- The head of NICE only gets to determine how the NICE budget is spent. I expect he would agree with you in a flash that all the money spent on Iraq should be funelled to NICE, and it is beyond stupid to blame him for the fact that the money is being wasted in Iraq.


wouldn't this budgeting rule have the implication that whenever any efficiency gain is made anywhere in the government, the NHS should be able to slurp it up for whatever treatment was the most recent rejection by NICE?

And in any case, why should the potential sources of finance for this treatment be restricted to efficiency gains? Perhaps the tax rate should be raised.


How about the old suggestion that taxpayers can nominate in what area a certain percentage of their tax can be spent? Say you have discretion over what happens to 25% of your income tax - that might be a pretty good way of ensuring more money to the NHS and less to the military than just abolishing Nice.


The man spending his money on tat could be entirely rational if he feels he gains more benefit from that than he does from food, and that's entirely his choice. We elect politicians to make those choices for us over services provided by or through the state. How much should we spend on health? How much on education? How much on the military? How much on the Olympics? Etc. There is always an emotive case for more to be spent on certain things, in particular health, but the maximum public utility is not best served by spending all our public money on health. We would all be healthier, maybe, but less educated, less safe, less interesting all round. Money spent on, say, art or the Olympics will mean less for the NHS and that means some people will suffer more or even die earlier as a consequence. It could be you or me. But in total utility terms, greater overall benefit may come from the spending on something other than health. Tough choices however you organise it.


I take you points regarding the generality of how choices are and could be made, but that's not how NICE actually operates.

As best I understand the process - and this is complicated (for me - it may make perfect sense to you, Chris/DD) - the starting point for all this is to evaluate the benefical effect of a new treatment in terms of Quality Adjusted Life Years (QALY), which 'measure' the the number of years of ‘perfect health’ added to the lives of individuals by the treatment - you'll see why I put measure in quotes a bit later.

This is then used to conduct a Cost Utility Analysis (CUA) which estimates the ratio between the cost of the treatment and the benefit in terms of QALYs, which goes into the calculation of its Incremental Cost Effectiveness Ratio (IECR) which measures the change in cost of using the treatment relative to an alternative (usually the current best available alternative treatment). Incremental, here, does not have it usual economic meaning, i.e. cost per additional unit, but rather measures the cost of switching to the new treatment.

Finally the IECR is compared with a monetary threshold defined by NICE (which is thought to be £30,000 per QALY) with those treatments with an IECR exceeding that figure being unlikely to be approved on cost grounds.

And if that wasn't enough, weightings in the QALY calculations are based on largely subjective criteria derived from population surveying. The three main methods of deriving the weightings are:

Time Trade-Off - those surveyed are asked to choose between being ill for a period of time or being restored to full health as the expense of a loss in life expectancy.

Standard Gamble - given choice of being ill for a period or taking a treatment that has a chance of restoring full health of killing them.

Visual Analogue Scale - people are asked to rate a state of ill-health on a scale of 0 (death) to 100 (full health)

This introduces all the usual problems of bias - when asked to rate a particular illness those who don't have it tend to over-estimate its impact on quality of life against those who do and weightings may be skewed by perceptions of prevalence and, therefore, degree of personal risk, personal prejudices (the calculations tend to weight physical conditions considerably higher than psychological ones) and so on...

In general, it looks to me to something between bullshit and voodoo - a way of pretending you're making rational, objective decisions when you could just as easily pull numbers of your arse and arrive at a better answers.

The comments to this entry are closed.

blogs I like

Blog powered by Typepad