Learning from variations to increase value for money in the NHS

This content relates to the following topics:

With the pressure on the NHS budget, there has been much talk in clinical and policy circles about variations in medical practices. In the aftermath of the July 2010 White Paper, 'Equity and Excellence: Liberating the NHS', Chris Ham and John Appleby identified the reduction of unwarranted practice variation as the best way to achieve the necessary increases in value for money.

The NHS Atlas of Variation in Healthcare was launched just months later in November 2010, and the Fund published Variations in health care: The good, the bad, and the inexplicable in April 2011.

As the Fund's first International Fellow, I will be spending my time here getting a better understanding of the particular implications of practice variation for the NHS at this time of budget tightening and organisational restructuring. My experience in the United States and elsewhere tells me that this will be time well spent. Wherever investigators have examined the epidemiology of health care, they have found geographic variation in rates of procedures and other activities, as well as in the outcomes of care received. Variation in outcomes for patients with similar profiles suggests that some care provision is associated with better outcomes; understanding this variation better will give an opportunity to improve the quality of clinical decisions, resulting in improved efficiency in the deployment of health care resources.

The first opportunity for the NHS is to become better at doing things right. This approach is at the core of improvement science and has been instrumental in decreasing surgical mortality rates and other harms in the United States and elsewhere. The second opportunity is for the NHS to become better at doing the right thing –providing each person with the services they need and no less, the services they want and no more – which is the key to value for money in any health economy. Though variations in outcomes and in practice are universal, the opportunity to learn from this has not been realised anywhere on a large scale. Because of the UK's commitment to care for every citizen through the NHS, it is in a unique position to show the value of embracing variation as a source of new knowledge for improvement of health care.

That learning begins with making the distinction between unwarranted and warranted variation. The former reflects the limits of professional knowledge and its application; the latter reflects differences in the clinical circumstances and preferences of individual patients. Unwarranted variation naturally receives the most attention. Efforts to develop evidence-based guidelines (such as those conducted by NICE), are meant to reduce unwarranted variation, but they must be tailored to particular clinical circumstances. However, tailoring services to clinical circumstances is not sufficient: different patients feel differently about the same illnesses, treatments and states of health.

Shared decision-making has been shown to improve both patients' knowledge and the alignment between patients' preferences and the care received, while often reducing service utilisation and costs of care. The approach has been studied broadly in the UK, was highlighted in the NHS White Paper and is the subject of our recent publication, Making shared decision-making a reality: No decision about me without me. It is designed to reduce unwarranted variation by ensuring that clinician and patient have an accurate interpretation of the best evidence. But it is also designed to help the clinician understand what matters most to the individual. Informed by both professional and personal knowledge, the resulting choices can reveal a patient's preference. Applied across a population, shared decision-making could offer a measure of the value of specific services as judged by those who live with the consequences. Applied across different services, these measures could guide commissioning and investment decisions that determine short- and long-term capacity to deliver different services.

Shared decision-making programmes (and other tools such as Ann McPherson's award-winning charity Health Talk Online) have been developed to support clinicians and patients in improving the quality of medical decisions, but they have had a modest uptake. The stakes couldn't be higher. Too many decisions that have an impact on patients' lives are taken in a state of avoidable ignorance. And when patients are unable to reveal their preferences through informed choice, commissioners and other decision-makers are also avoidably uninformed. Let's hope that the current budgetary pressures produce the political will to embrace practice variation in all of its complexity, gaining the knowledge needed for an NHS that achieves value for money by delivering what is valued by patients.

Related events


Dr Hugh Rayner

Consultant Nephrologist,
Heart of England NHS Foundation Trust
Comment date
28 July 2011
You are so right to link these themes together. The big problem is overcoming the reluctance of doctors to change their well established ways of thinking and behaving. Change is more likely to happen if doctors believe their working lives will improve as a result and belief is more likely if they are shown credible examples. So I unashamedly publicise a paper that reports the impact of implementing a new model of care for patients with diabetes and kidney disease that incorporates reduction in variation in care, support for self-care and shared decision making.
Systematic kidney disease management in a population with diabetes mellitus: turning the tide of kidney failure
Hugh C Rayner, Lee Hollingworth, Robert Higgins, Simon Dodds
BMJ Qual Saf 2011;bmjqs-2011-000061Published Online First: 30 June 2011 doi:10.1136/bmjqs-2011-000061

Paul Kelly

Comment date
28 July 2011
Good posting. Need to remind, however, that it is much easier to add to the evidence base than to change paradigms that dictate the pattern of care. "Unwanted variation" may be a result of factors other than those associated with practitioners. Varying policies across PCTs/SHAs is one factor, as is the local political environment.

Add your comment