Editorials

Redesigning health care

Radical redesign is a way to radically improve

A decade or so ago car manufacturers completely transformed the way they made motor cars: they stopped stockpiling components and supplied them "just in time"; they drastically reduced the types of components such as subframes and engines; they worked closely with their suppliers on quality and timeliness instead of beating them down on price; and they eliminated waste by making all workers responsible for quality, not just inspectors. As a result new models took less time to develop, their quality improved, and cars could be made to order for each customer. At the same time productivity improved and costs fell.¹ Since then people in health care have realised that their service has a long way to go to match the performance of other manufacturing and service industries. Report after report, from country after country, has documented the size of the gap between the best evidence based care that could be delivered to patients and the care they actually receive. Health systems fail to provide treatments that are known to work, persist in using treatments that don't work, enforce delays, and tolerate high levels of error. Healthcare leaders are now recognising, as car manufacturers did, that the healthcare system needs radically redesigning. ^{2 3}

Such redesign of an important part of the healthcare systemclinical office practice (in UK terms ambulatory care by mainly primary care physicians)was the focus of a meeting in Atlanta earlier this month. Though the meeting was run by the Boston-based Institute for Health Improvement (and all the participating practices were using its methods), not all the successes were American. The UK's primary healthcare collaborative⁴which now includes about 20% of English practicesalso showed impressive improvements in patient access and evidence based care.

The organisations that have been taking part in the IHI's "Idealised design of clinical office practices" project⁵ range from community health centres to academic medical centres and have shown improvements in patient satisfaction, staff satisfaction, revenues and reduced costs, and also in clinical outcomes. They have made these improvements using the familiar methods of quality improvementunderstanding the problem, identifying possible changes, testing those changes to see whether the change is an improvement, and then incorporating that change and moving on to try another. Quality practitioners are strong on data, because it's crucial to know whether a change has improved things, but the key is pragmatism, using "good enough," easy to get data. So for example, they emphasise measuring small representative samples, combining quantitative and qualitative data, and building data collection into daily work. Using these methods some practices have, for example, seen their proportions of diabetic patients who have had their eyes examined rise from 65% to 85% and their glycosylated haemoglobin measured from 47% to 80%. And among the first wave practices in the UK collaborative the percentage prescribing aspirin for over 80% of their patients with coronary heart disease has increased from 23% last July to 50% this April. That means, pointed out John Oldham, head of the primary care collaborative, "that there are people walking around today who wouldn't have been if the teams hadn't done the work."

Don Berwick, president of IHI, explained that it had chosen clinical office practice because it's the heart of health care, where doctors and patients meet. The vision is the patient's statement: "They give me exactly the help I want (and need) exactly when I want (and need it) . . . while maintaining and improving a joyful work environment and a financially viable organisation." The point of such a goal, said Berwick, is that it is "psychotic": it will not be achieved simply by tinkering, the system has to be redesigned (see box). As with many visions, it packs a lot of meaning into an almost banal statement

Unpacking it involves working on access, interaction, reliability, and vitalityand the improvements in access are perhaps the most immediately striking. The target is to give patients an appointment the day they ring upand many of the providers in the IHI project, together with many of the practices in the UK collaborative, have almost achieved that aim. For example, 16 primary care sites in Cambridge, Massachusetts, reduced the number of days for a new appointment from 4.3 in 1999 to 1.8 in 2001. Seven English practices reduced it from 8 days to 2 days over five months.

Mark Murray, who has led the work on access for the IHI, explained that the underlying aim was to "do today's work today, not next month." He pointed out that most demand is predictablethe waiting time has probably been stable for years, and its pattern is largely predictable. Once that is understood, it's a question of working off the backlog, matching capacity to demand, reducing the number of appointment types (having different types with different queues lengthens overall waiting), and trying to do more within each consultation.⁶ Work can also be managed in other ways: patients like email (and doctors can learn to love it too), and there was great interest at the meeting in group visits. Instead of being invited for an individual review, patients with chronic diseases come to a session with other patients, where there's an opportunity for education and peer support, as well as routine check ups and individual advice from the doctor or nurse if wanted. Through such measures practices have seen the average number of visits per doctor drop: they are replaced by telephone calls, emails, group visitsor they simply disappear because patients see their own doctor and know they can always get an appointment with him or her that day if needed (demand drops by about 15% by this alone, claims Murray).

Once you've got the patient seeing the doctor (or nurse) the system has to ensure the quality of the interaction with the patient and that the technical care is reliable. If all this is right then vitality, the fourth component, should followstaff will be happier and eager to innovate and finances healthier. The trick, as Ed Wagner, leader of a parallel project on improving chronic disease care,⁷ put it, is to get systematic. And here British general practice already has many of the elements that help ensure good interaction and reliability: a registered population, an emphasis on consulting in ways that elicit patients' expectations and needs, some degree of electronic health record in nearly every practice, disease registers, access to summary evidence such as Clinical Evidence,⁸ and practice teams with nurses running chronic disease clinics.

But, even if the elements are already there, none of this happens by accident. It's hard work working off a backlog, it's hard work enthusing people. Some practices in these projects have fallen by the wayside. And there's still a way to go. The practices doing this work need to extend their own improvementsand the improvement and redesign need to extend beyond the pioneers. A specific aim of the UK primary care collaborative is to ensure the systematic transfer of knowledge about improvement to all practices in England, with project leaders and exemplar practices in all primary care groups. And in the United States IHI has just received $21m for a further collaborative project to improve whole systems of health care.⁹ But maybe the core ingredient is leadership: as one speaker pointed out, the best leaders have a strong belief that they can change anything they want.

Jane Smith, deputy editor. 

BMJ