Only by swiftly adressing the issue of clinical practice variation and developing better evidence based practice can we stop the quality of NHS care eroding, writes Professor Alan Maynard.

The NHS faces formidable budget problems as patient demand and expectations rise in an era of at best flat funding.

The “solution”, offered by consultants such as McKinsey and Whitehall policy makers, is “efficiency savings” through the reduction of apparently large and unwarranted variations in clinical practice and the swift development of conservative, safe, evidence based practice. Is this approach likely to save the NHS or is it utopian nonsense?

Variation is evident in all human activities. In the US the top 10 per cent of companies are twice as productive as the lowest 10 per cent. Variations in productivity are even wider in India and China.

Variations in clinical activity have now become the “Holy Grail” of saving the NHS from bankruptcy and funding the implementation of the Obama reforms in the US. Policy makers in all healthcare systems believe that reductions in clinical practice variations are the solution to ever-expanding healthcare budgets. Are they right?

In the US, Dartmouth Institute researchers have analysed the federal Medicare programme for nearly 40 years. The academic Jack Wennberg asserted in a recent book that if conservative, safe practices were adopted, a saving of 40 per cent of the budget was possible. Translating this thesis to England, McKinsey asserts that a 20 per cent saving is possible: the Nicholson challenge.

The problem is that variations have a long history and healthcare systems have tolerated rather than reduced them. An analysis of tonsillectomy rates in England in the 1930s concluded large variations were a product of differences in “medical opinion”.

Another examination of the same condition in Scotland in the 1970s concluded the same thing. The Dartmouth Institute has produced similar explanations of US clinical practice variations for medical versus surgical interventions for hysterectomy.

These outcomes are a product of only half of medical care having an evidence base and where the evidence base exists, practitioners often choosing to ignore it, preferring to follow opinion rather than facts about the cost-effectiveness of competing interventions. This makes the use of practice guidelines as crucial as their audit by the medical profession. The rhetoric of medicine is now “evidence based” but variations are endemic, suggesting the rhetoric is not reflected in reality: you can lead a horse to water but you cannot make it drink.

Policy makers worldwide face the same problems: how to agree “best practice” and ensure that workers on the production line, be they at Nissan or the NHS, standardise their practices.

In the NHS, the National Institute for Health and Clinical Excellence is producing good guidelines which draw on the evidence base. Where that is thin, it uses opinion to achieve consensus about how to care for patients. However, we have little idea of the extent to which these NICE efforts lead to changes in practice.

With the Nicholson challenge, slow change is unacceptable. Miracles have to be performed in the next four years if £20bn is to be recycled and used to meet emerging patient demands. However, no healthcare system has a history of miracles. Even if we had evidence-based guidelines, we don’t know how to radically and swiftly alter the practices of clinical teams.

Are we doomed? Hopefully the current crisis will lead to a sea change in the ways healthcare is delivered. The closure of some accident and emergency facilities and concentration of activity in fewer facilities to get better outcomes and costs are essential small steps in this direction. More radical change will involve major shifts in activity and resources, particularly in the terms and conditions of employment.

Agenda for Change is inflexible, poorly managed with incremental creep, and unaffordable. 24/7 working will require staff to forego premium rates and work for standard rates at unsocial times. Cheaper inputs have to replace existing practitioners – GP list sizes may double or treble with nurses substituting for doctors.

Screams of protest are likely, but the choice is simple: deal with clinical practice productivity variations or face the consequences of an eroding NHS where the quality of care declines as managers struggle to balance the books.