Health care absorbs a large proportion of governments’ budgets across the OECD, so it’s no surprise that governments are constantly looking for ways to increase productivity in the health care sector. When the Blair administration came to power in the United Kingdom in 1997, they maintained a focus on supply-side reform but turned their back on the previous administration’s use of competition as a way of increasing productivity in the National Health Service (NHS). Instead they chose to implement a system of targets for senior managers.1
The most important targets were those for waiting times for non-emergency care. Long waiting times for such care dogged the British NHS – and its politicians – for over a decade. In an attempt to reduce them, the Blair administration established a system of targets accompanied by strong managerial sanctions for failure to meet these targets. In 2000 it was announced that waiting times in England were to progressively fall over the next 5 years, beginning with an eighteen-month maximum that was to fall to six months by 2005. Managers of hospitals failing to meet these targets were subject to a battery of penalties including “naming and shaming”, loss of autonomy, and loss of their jobs. Indeed, such was the focus on these targets and the strength of sanctions that the system was dubbed ‘”targets and terror”.
In contrast, the Scottish Parliament, which assumed responsibility for the NHS in Scotland on devolution in 1999, downplayed the use of targets, preferring to promote cooperation and collaboration instead. This policy variation, in a system that had been common until devolution, provides an opportunity to evaluate whether targets work.
Did targets cut waiting times?
Our research examines whether the targets achieved their goals of reducing waiting times. Economists don’t tend to favour the use of targets, arguing that they engender behaviour that ”hits the target but misses the point”. Commentators in the NHS also argued that targets would simply result in ”gaming”, with undesirable responses including the stacking up of patients at waits just shorter than the target, a diversion of activity away from other aspects of patient care, and more ill patients being made to wait longer in order that scarce capacity could be used to treat those who had been waiting a long time.
We found little evidence to support these fears. The policy reduced waiting times. Average waiting times fell in England by twelve days more than in Scotland and by around fifty-five days for those who had to wait the longest. Patients did not get stacked up at the maximum waiting times. Figure 1 shows the distribution of waiting times in the two countries for elective treatment. The solid line shows the pre-policy distribution (which is the same in each country specific panel). The dotted line shows the post-policy distribution, where there is one for each year for each country. The vertical dotted lines mark the waiting times targets in operation in England. Comparing pre- and post-policy distributions, it is clear that the effect of the policy in England was not just to reduce the waits of those who would have breached the target but to shift the distribution to the left, therefore reducing not only waits that were greater than the target but also waits below the target. In contrast, in Scotland the distribution moved rightwards, increasing the number of longer waits and reducing the number that waited below the target set for England.
Figure 1. The distribution of waiting times in England and Scotland
We found no compelling evidence that the order in which patients were treated was altered to meet these targets. Nor were shorter waits achieved at the expense of patient treatment. There were no changes in differences in mean length of stay pre- and post- policy. Nor did care outcomes – crudely measured by death rates post-hospitalisation – worsen in England post-policy: in fact, if anything, these appeared to have improved in England after the terror and targets regime was introduced. We did find some evidence that hospitals that were at greater risk of breaching their targets did remove or suspend more patients from the waiting list than those who were farther from their targets. This might either be interpreted as “good waiting list management” or it could be seen as an attempt to classify patients in a manner that avoided them being counted towards targets. However, this re-classification did not appear to damage patient outcomes.
Why did targets work?
Our conclusion is that – at least in this instance – targets linked to strong managerial sanctions worked. The question is why? Our tentative answer is that these targets were directed towards something on which there was considerable consensus. Long waiting times were seen as undesirable by almost everyone in the United Kingdom – patients, tax-payers, clinicians, hospital administrators and politicians – and by many as a blot on the international reputation of the NHS. Targets enabled mangers to tackle organisational issues that resulted in long lists; sanctions gave this desire added bite and the long-term nature of the policy might have allowed the idea of a service run with shorter waiting lists to become embedded in the organisational psyche.
In addition, this was an era of unprecedented growth in resources for the NHS. While differences in resources did not account for the success of the policy (resource growth was similar if not higher in Scotland), bringing about organisational change may be easier in an era of generous funding. Managers, for example, might have been able to pay staff to undertake extra operations.
We do not advocate targets as the answer to all prayers of a politician: however, in this case they appear to have delivered the desired service improvement.
Propper, C, Sutton, M, Windmeijer F and Whitnall, C (2008), “Did `Targets and Terror' Reduce Waiting Times in England for Hospital Care?” , The BE Journal of Economic Analysis and Policy Vol. 8 : Iss. 2 (Contributions), Article 5.