Intended for healthcare professionals

Head To Head

Have targets done more harm than good in the English NHS? Yes

BMJ 2009; 338 doi: https://doi.org/10.1136/bmj.a3130 (Published 16 January 2009) Cite this as: BMJ 2009;338:a3130
  1. James Gubb, director of health unit
  1. 1Civitas, London SW1P 2EZ
  1. james.gubb{at}civitas.org.uk

    James Gubb argues that the focus on targets has ignored underlying problems important to patient care but Gwyn Bevan (doi:10.1136/bmj.a3129) believes it has resulted in real improvements in care

    One of the most pervasive beliefs in government is that quality in the NHS is a function of individuals who need buttons pressed and levers pulled by targets to deliver optimal performance.1 This is misguided. The most intractable problems in health care—the lack of communication, leadership, and teamwork; the lack of integration; and the lack of any meaningful, patient focused, quality framework—are systemic or cultural.2 3 4 5 And targets have only made them worse. If you treat people like knaves and pawns, they will behave like them.6

    Perhaps the most influential management revolution of the past century was led by Taiichi Ohno at Toyota.7 Systems or “lean” thinking saw Toyota become the world leader in manufacturing; applied to health care it is again showing the way. Flinders Medical Centre in Australia was one of the first to take up the principle and after two and a half years was doing 15-20% more work, with fewer safety incidents, on the same budget, using the same infrastructure, staff, and technology. More recently, Bolton NHS Trust, using less space and fewer resources, reduced its average turnaround time in pathology from over 24 hours to 2-3 hours.8 9

    Wrong focus

    Lean thinking exposes the fallacy of targets. Premised on value to the customer, local leadership, and looking at systems as a whole, lean thinking aims to improve flow and eliminate waste across a system by getting the right things to the right place, at the right time, in the right quantities.10 Targets do exactly the opposite: they devalue the customer (patient) by focusing attention on an arbitrary number, devalue local leadership by relying on central control, and break systems into silos by focusing attention on isolated parts rather than the whole.

    Emergency medicine is a case in point. Although official statistics show that 98% of patients were seen in under the four hour target,11 academics have used queuing theory—a mathematical analysis of waiting time statistics—to show this can have been achieved only by “the employment of dubious management tactics.”12 Well documented examples, confirmed in surveys by the British Medical Association,13 include moving patients to clinical decision units, making patients wait in ambulances, admitting patients unnecessarily, discharging people too early, and miscoding data.12 14 15 All are detrimental to patient care, yet politically charged league tables show only those that do not meet the target.16

    Critics will say that this was a badly designed target, or that audit should have been more effective.15 Indeed, studies have shown targets inducing change and delivering against their narrowly defined goals.15 17 Waiting times have fallen for inpatients, outpatients, in accident and emergency, and across the referral to treatment pathway; increases in staff numbers and facilities typically exceeded targets set out in the NHS Plan; and rates of meticillin resistant Staphylococcus aureus (MRSA) and Clostridium difficile infection are falling.

    However, it is the wider impact we should be concerned with. What you can’t see and measure doesn’t exist. The target is met and taken as evidence of good performance, but its true impact is concealed. After targets were introduced for inpatient and outpatient waiting times, median waits increased, waiting time was shifted to diagnostics,14 and bed occupancy rose to levels associated with excessive risk of infection.18 Yet the government’s solution was to search for the right target—and introduce further targets for 18 week referral to treatment and infection rates for MRSA and C difficile—rather than understand the extent of the problems they cause.19

    Undermining staff

    The most pernicious outcome of this has been on the ability of organisations to develop a self improving culture that truly puts patients at centre stage. Pressure to score short term goals has left measurement associated with spin, selection, and punishment, rather than the ability to learn.3 When only 26% of accident and emergency staff view figures submitted by their department as a fair reflection of performance,13 meaningful analysis of service development becomes incredibly difficult.20 The four hour target is met, but who is asking why the patient was in the emergency department in the first place, whether the patient is actually on the road to recovery, and whether the rapid transfer from accident and emergency has actually just left the patient waiting elsewhere (and for longer)?

    The preoccupation with hitting targets results in the actual journey an individual patient experiences becoming secondary; performance is determined against crude indicators, not the expectations and experience of those using the service.

    The American statistician W Edwards Deming once warned that 97% of what is important either isn’t measured or isn’t measurable.21 Good medicine is premised on values—on kindness, caring, good communication, honesty, and, above all, trust.22 When clinicians are seeing numbers, not the patient – and believe targets have undermined clinical decision making – such values are left in a parlous state at precisely the time when the most pressing challenge facing the NHS is to revive the first purpose of clinical medicine: to relieve human suffering.23 24 25 Measurement should be focused on the culture we are trying to create, not the culture we are trying to escape from. Targets suit politicians, not patients.

    Notes

    Cite this as: BMJ 2009;338:a3130

    Footnotes

    • Competing interests: None declared.

    References