I recently got into a discussion on twitter about the properties of the ICECAP instruments and what the zero on these mean. One particular point I made was that us “saying” the state of “no capabilities” must be zero didn’t necessarily make it so, at least in the eyes of a mathematical psychologist. They’d probably say it is not a ratio scale and might not even have good interval scale properties at the aggregate level (if there’s improperly-adjusted-for underlying heterogeneity).
I’m not too worried about these points, though I personally think better subgroup/heterogeneity analyses need to be done in future to address the latter point. But it did lead me to think about that old recommendation “don’t let the perfect become the enemy of the good”. This potentially gets a lot of extra-welfarism into hot water, where the maths psych people are concerned: instruments and valuation tweaks or even in some cases the whole valuation method (VAS and arguably TTO) have little in the way of theory (that that group would recognise) behind them. However, I remember one health economist summarising a discussion he had with clinicians and members of the public as to how scarce resources should be allocated in health care and they “naturally” came up with something that approximated a QALY with TTO scoring. This is fair enough and I am happy with the newer theories/concepts put forward to justify what health economists do in that particular area. After all, extra-welfarism doesn’t have the same assumptions and theories as traditional welfarist economics so why get bothered about what another discipline entirely thinks?
I guess I’m just naturally – having worked so long with a maths psych guru – very particular about getting scoring “right”, as in it satisfying one or more of the properties inherent in proper scales (absolute/ratio/interval/difference). So yeah I guess I may be guilty of being dissatisfied with just “good”…but in my defence, we are producing tariffs (sets of scores) here that are being increasingly used across the world – the Netherlands has already decided on a dual “QALY+” approach: more than one evaluative space seems to finally be accepted, to aid decision-making. We shouldn’t stand still, particularly as we know a lot more about the properties of the Best-Worst Scaling valuation technique now than we did back in the original UK ICECAP-O valuation exercise. Whilst it is gratifying that public interest and funders have agreed with us that areas like end-of-life care and (potentially) children need ICECAP instruments, we should not rest on our laurels with existing instruments.