The concept of 'trade-offs' is well understood in health – benefits of drugs traded against side-effects, for example, but many find it harder to juggle the trade-offs of change, says Raj Persaud.

At the heart of the clinical practice of medicine is the idea of a ‘trade-off’ – we prescribe a drug juggling the downside of the side-effects with the upside of the treatment benefits.

No medical school would be doing its job if it did not teach the complexity of offering treatment. Tortured debates ensue as to the value of screening, for example. What is the point of worrying thousands and inconveniencing them if the only benefit is the marginal improvement in quality of life in the small number of cases that are actually picked up?

Perhaps one source of conflict between managers and doctors is a common misapprehension over this notion of the ‘trade-off’. To doctors, for example, managers may seem to have difficulty grappling with the trade-offs implicit in any change.

This can be seen in the impoverished way the health service has tried to measure patient preferences to date. Yet the measurement of attitudes and values associated with healthcare will inevitably continue to be a critical issue in terms of providing care which meets with patient approval.

The scientific approach to an 'attitude' and its formal measurement derives from social psychology, where an 'attitude' is measured as 'a psychological tendency that is expressed by evaluating a particular entity with some degree of favour or disfavour'. So patients prefer a nice waiting room and charming nurses and this contributes to their ‘attitude’ to the hospital or clinic or treatment.

But the problem is that most attitude measurement in the NHS simplistically probes what patient attitudes are without considering two crucial social psychological issues – patients may be loathe to tell you directly what they really think, and their attitudes only make sense in terms of an appreciation of the implicit trade-offs involved. Charming nurses may not, for example, turn out to be quite so vital to patients if they understood that this benefit could only be bought at the expense of numbers of nurses.

This more realistic concept of preferences comes from economics, where preferences are defined as people's 'utility' for consuming healthcare goods and services. The assumption is that we aim to maximise our utility. Economic theory argues that utilities or benefits can be measured in money equivalence. In other words the real strength of preferences is the monetary compensation that would leave people indifferent between having a given utility change and not having the change (that is, willingness to pay or willingness to accept).

If you could be given a money substitute for charming nurses, how much would compensate you for their absence? This figure, compared with the monetary value for what you would accept to forgo, for example, clean waiting rooms, tells us which you really prefer.

In the commercial sector the idea that you cannot have everything seems pretty self-evident – but in the NHS it looks as though we are striving to chase our tails so often because patients’ true attitudes and preferences have not been properly measured or understood.

The technique currently popular in social psychology which takes this point on board is called ‘conjoint analysis’ and it assumes that individuals make trade-offs within a resource constraint. In contrast, the most commonly used approaches to measuring attitudes in the NHS typically do not require patients to make choices within a resource constraint.

Attituderesearchers have also long been sceptical of assuming that our reports of our own attitudes are not tempered by concerns over impression management. So attitudes towards immoral or illegal behaviour need to be probed in a more sophisticated manner than just eliciting attitudes on simple scales that measure intensity of evaluation.

In a recent review of attitude measurement Laurie Rudman, a social psychologist based at Rutgers University in the US, points out that response latency measures are now the buzz approach and they yield evaluations that are unlikely to be consciously controlled or censored

This is because they are taken from reaction time tasks that measure people's attitudes or beliefs indirectly (without asking people how they feel or think). That is, people's attention is focused not on the attitude object, but on performing an objective task, and attitudes are then inferred from systematic variations in task performance.

For example, in the Implicit Association Test, automatic pro-white bias is indicated when people show faster performance categorising pleasant words and white (and unpleasant words and black) together, compared with categorising unpleasant words and white (and pleasant words and black) together. Thus, implicit attitudes can be characterised as the automatic association people have between an object and evaluation (whether it is good or bad).

Another venerable principle in social psychology is that people prefer consonant (as opposed to dissonant) evaluations of related attitudes. For example, according to this principle, if I like myself and I am male, then I should also like men.

Yet when it comes to healthcare, consistency might break down and need to be measured properly in-depth. For example, we may place a high value on fitness yet derive pleasure from smoking.

Sometimes patients hold inconsistent attitudes, preferences they are unclear about or may even vacillate over. Attempting to provide a service which ticked all the boxes in these conditions would be impossible. Sometimes managers should be measuring attitudes more carefully or instead providing leadership and judgement – perhaps we only discover what our attitudes and preferences really are when they have been articulated by an insightful other.

Managers may want to exercise judgement and leadership to innovate first and check later what patients think or prefer – rather than always go laboriously the other way round. Especially if you understand how difficult it is to get at what patients really think about your service.