Performance targets have been very useful in many public services, but at their heart they are flawed. There is always a way of achieving a given target without the intention behind the target.
One of the best examples of the usefulness of targets is the NHS waiting time targets. These targets helped reduce the number of people waiting over nine months for admission to hospital from 175,000 in 1997 to 223 in 2004.
In contrast, the NHS Ambulance Service provides a very good example of flawed targets. The ambulance service arguably did not need a target for responding to calls – and certainly did not need the blaming and shaming culture that became associated with it. Any visit to an ambulance service before targets were introduced would confirm that ambulance crews, the people manning the telephones, and their managers were immersed in a culture that prioritised a speedy response – not the case with NHS waiting times.
Prior to the review of ambulance services, which began in 2013, the entire service was driven by two flawed targets. The first target was that 75% of ‘red’ (life threatening) calls should receive a response within eight minutes. But this refers to the first vehicle to arrive – even if it is a motorbike and the patient needs to be taken to hospital – and has resulted in one in four patients needing hospital treatment undergoing a hidden wait because the vehicle dispatched cannot transport them to hospital. In addition, since ambulance staff record whether a call is classified (or reclassified!) as an emergency call or not, there is the potential for further gaming when the stakes are high in terms of potential reputational damage.
The second target requires call handlers to assess calls and send out a vehicle within 60 seconds. However in practice it is not always easy to assess the most appropriate response in 60 seconds and therefore call handlers often send several vehicles to the same 999 call to be sure of meeting the eight-minute response target. The result is that one in four vehicles dispatched were stood down before arrival.
The ambulance service was also affected by the A&E four-hour target for patients and, while not all delays in handover were due to the target, 606,000 ambulance service hours were lost in 2015-16 due to handover problems in hospitals. This has contributed to a rise in the need for private ambulances, which has increased from £22m to £79m since 2011-12.
Finally, the fact there was no target for ‘green’ patients not in immediate need has resulted in poor service for many such people who sometimes have to wait several hours.
In reviewing the ambulance service, the national review team for urgent and emergency care had essentially two options: to improve the percentage achieving the targets or to improve the service and develop new targets around patient needs. They chose the latter option. Politically, removing targets altogether was not an option, as people would say that this indicated that the government did not care about the service.
The review team, following consultation with ambulance staff and groups representing patients, developed a new system of targets. They changed the target for life threatening and emergency calls from eight to seven minutes, basing this target on when the right vehicle arrives – not the first. They gave call handlers more time to assess calls where these were unlikely to be life threatening and introduced targets for ‘green’ patients. In addition, they introduced some targets across organisational boundaries. For example, 90% of stroke patients should receive thrombolysis or first CT scan within 180 mins of making a 999 call. They also recommended more use of ‘hear and treat’ when appropriate and more ambulances to help achieve the new targets.
After the new system was piloted in several ambulance trusts and received positive feedback from the independent evaluation, Sir Bruce Keogh, medical director of NHS England, concluded that “these changes will end the culture of hitting the target but missing the point. They will refocus the service on what actually counts: outcomes for patients”.
It remains to be seen whether the improvements seen in the ambulance regions that piloted the system will be sustained and be seen in other regions too. The new targets are still flawed, eg staff could turn up and do little but still achieve the targets. However, the approach taken by the review team has much to commend it.
There are lessons here for other public sector organisations. They should ensure that their aim is to improve the service and not just improve performance on flawed targets. Performance measures and targets must be based on outcomes or evidence-based drivers of outcomes and performance should be managed across organisational boundaries. Finally, they need to develop a performance management culture based on innovation, improvement and learning rather than a top-down blame culture.
To do this in a structured way, the author recommends the Public Sector Scorecard framework. This involves managers, staff and service users in a workshop-based approach and has three phases: strategy mapping, service improvement, and measurement and evaluation.
