Perhaps 20, possibly 25 or even 30, years ago I had breakfast (or was it lunch?) with Geoffrey Marsh, a GP from Teesside, and he told me that all GPs should have 3000 patients. I think he was right, but since that time average list sizes have steadily shrunk until they are now under 1600.

Marsh’s argument was simple. GPs should work in teams, and each member of the team should do the work best suited to his or her skills. The GPs’ skills were primarily clinical, and, Marsh argued, if they didn’t have at least 3000 patients they wouldn’t maintain those skills because they wouldn’t see a broad enough range of clinical problems. Others in the team—nurses, counsellors, physiotherapists, social workers, and others—would use their skills, preforming better, argued Marsh, than a GP trying to do their work, perhaps counselling unhappy couples. But the team would be a team, working closely together and respecting each other.

Finance wasn’t part of Marsh’s argument. His argument was about giving the best possible service to patients and GPs and others maintaining their skills and perhaps satisfaction rather than dabbling ineffectually in each other’s work. But as we recognise that the care services will have to cope with a much heavier burden without more money and hear calls for 30 000 more GPs we need to revisit Marsh’s argument.
I think that his core argument is more relevant than ever because of the financial constraints, but I also have much more experience than I did when I dined with Marsh of health systems where doctors are rare. I work with a team from the University of Cape Town who have developed a package of guidelines, training, nurse prescribing, and a cascade system of scaling up that allows nurses to provide high quality primary care in rural areas where there are no doctors.

I work with researchers in Asia, Africa, and Latin America who have shown what good results can be achieved by community health workers with minimal training. Sometimes dismissed by medical unions as “providing second class care for second class patients,” community health workers can not only do what more highly trained health workers can do but also have advantages. For a start, they are there—in rural areas and urban slums where doctors are rare. But more importantly they are close to the patients, understand their needs and speak their language in a way that health professionals from far away find it hard to match.

Increasingly community health workers are being used in developed countries, and they could play a bigger role in Britain—because the advantages they have in rural Africa can also apply in parts of Britain, particularly the poor parts. I learnt this through attending the conferences of the Association of General Practices in Urban Deprived Areas (AGUDA), which may now be extinct as I can’t find it on the web. Listening to patients and watching role plays, I discovered that for many patients living in deprived areas general practices felt as forbidding and remote as hospitals. I noted too that between the first and the 10th anniversary meeting most of the doctors had disappeared. It’s the scandal of British general practice that the inverse care law, described by a British GP, applies much more strongly in general practice than in hospital practice.

Many GPs, I suspect, might be happier working with bigger lists and bigger teams, “doing more medicine and less social work.”

Clearly the current model of general practice is under great strain. I urge the leaders of general practice to revisit Marsh’s thinking and studies: the route to the future may lie in the past.

Richard Smith was the editor of the BMJ until 2004 and is director of the United Health Group’s chronic disease initiative.