As the NHS has become sophisticated about the need to measure outcomes, it inevitably raises questions about what is actually measured, and therefore what matters. By necessity, the focus in the early years of this government has been on the utterly tangible - mortality rates, waiting lists and so on.
But there is now an increasing emphasis - driven in part by greater awareness of long-term conditions and also of the link between physical and mental health - on the issue of well-being. In short, to what extent is it the NHS's role to improve the public's happiness as well as its health?
It is a question being asked of public services in general, in part driven by the Treasury's recent review of measuring public service outputs more subtly. But it has special resonance for the NHS, as our cover feature explores this week.
The problem is how do you actually measure it, a task being tackled by the Office of National Statistics but drawing the attention of a range of academics and commentators. Most of the data available is about activity rather than outcome - but this is anathema to well-being. What is the use of counting the number of words in a joke without finding out if it's funny or not?
It seems intuitive that statistics and happiness do not go together, but the temptation is to write off well-being as something so unclassifiable that you cannot even try. But already thinking is emerging about factors such as 'quality change', which give an indication of improvements in what is delivered.
As we report, public health researchers in Scotland are pioneering work on universally applicable questionnaires. to identify those at most risk of low. well-being. That work holds the real promise of not just interpreting society's happiness, but of changing it.