Intended for healthcare professionals

Editor's Choice

Tackling practice variation

BMJ 2011; 342 doi: (Published 23 March 2011) Cite this as: BMJ 2011;342:d1884
  1. Fiona Godlee, editor, BMJ
  1. fgodlee{at}

Our enthusiasm for tackling the topic of practice variation has led us to commission five articles and a podcast. There’s an Analysis from “Jack” Wennberg, the undisputed father of practice variation, in which he deftly summarises the evidence of underuse, overuse, and misuse of health care in the US and the UK (doi:10.1136/bmj.d1513). There’s a commentary from Gert Westert and Marjan Faber from the Netherlands showing that variations are less extreme in what was, until recently, a centrally planned health system (doi:10.1136/bmj.d1429). We have a review by Chris Ham of Wennberg’s latest book (doi:10.1136/bmj.d1757), and a review of one of his earliest, the Dartmouth Atlas of Health Care, now a medical classic (doi:10.1136/bmj.d1756). Richard Smith compares it with Darwin’s On the Origin of Species because of its rigorous accumulation of data and for fundamentally changing our view of the world. Finally, an editorial from Nick Mays discusses the atlas’s first direct offspring, the NHS Atlas of Variation in Healthcare (doi:10.1136/bmj.d1849). Published late last year by the Department of Health, this shows how primary care trusts vary in their care of patients, their costs, and their outcomes.

Is this coverage overkill—or, in the language of practice variation, supply sensitive overuse? Given the importance of the topic to health systems and patients around the world, I don’t think so. The evidence that more care does not mean better care is overwhelming, and the scope for savings from more rational pathways of care is huge, especially with chronic conditions. All five articles are worth reading because they present the story, and the concepts behind it, in different ways. And by the time you’ve read them all and listened to the podcast (, you will easily be able to answer the following questions.

Who was the first person to document variations in clinical practice, and what did he find? How did he explain his findings? How did Jack Wennberg tackle the same problem in Vermont 40 years later? What are Wennberg’s three categories of care in relation to practice variation? Which ones are “OK” and which are not? What is Roemer’s law? How much would the US save if the whole country followed practice patterns of organised systems such as Mayo Clinic? What is Jack Wennberg’s main solution to unwarranted practice variation? (Clue: the Salzburg declaration is just published on (

The first reader to get all the answers right will get a pat on the back. The first health system to get to grips properly with unwarranted practice variation will get a high quality, low cost health system, which must be everyone’s goal. Sadly, if Nick Mays is right, the NHS is heading in the wrong direction. General practitioner consortiums may do better than primary care trusts, he says, because of stronger incentives to be efficient and to reinvest savings in services. But other changes—the scrapping of the Audit Commission, the emphasis on localism, and the blurring of geographically defined populations—may bring more not less variation. And they won’t help the collection of data or the integration of care for chronic conditions.


Cite this as: BMJ 2011;342:d1884