The UK health reforms, among other things, have allowed some family pr
actitioners ('fundholders') to control part of the budget used to purc
hase secondary care for the patients registered with their practices.
Some feared that this change in organization might threaten the equity
of the National Health Service (NHS) by giving advantages to the pati
ents of fundholding general practitioners (GPs). This has led to the f
ormation of the first large GP 'Multifund' in the UK. Whilst the first
small fundholding co-operatives were conceived simply as a way of imp
roving the business efficiency of ordinary fundholding, the much large
r Kingston and Richmond (KandR) Multifund is a professional co-operati
ve of fundholding family practitioners who share a common management a
llowance, which controls pound 32.6 million per annum, about 20% of th
e public funds devoted to health care in its locality, and has a broad
er agenda. It sets out to end any two-tier system discriminating betwe
en the patients of fundholding and other practitioners within a locali
ty by making it possible for all GPs to become fundholders within an o
pen, democratic doctors' co-operative. Its declared aims are to mainta
in equity and clinical integrity, to plan health care and to undertake
scientific evaluation. Its overall goal is to preserve and improve up
on the economic and ethical success of the British NHS. A brief evalua
tion is included which suggests that Multifund members believe their o
rganization is currently meeting most of these aims. The essay argues
for GP-led Multifund systems centred on outcome evidence. A new genera
lized global tool for measuring outcome, is being developed to assist
in this process and it is briefly described here and is designed to fa
cilitate: (1) the selection of cost-effective interventions, (2) the n
egotiation of appropriate funding from government and (3) the demonstr
ation of ethical behaviour in rationing health provision within a limi
ted budget. It is envisaged that the Multifund's scope will be extende
d, to administer the majority of the funds available for health care,
by involving all local primary health providers and becoming a 'careho
lding' consortium. This would mean obtaining local accountability thro
ugh new community health forums and improving mutual trust through pub
lic professional affirmations. If all this is successful multifunding
could become a community-centred model for the management of health se
rvice money which may in turn have international application.