American guidelines on managed care are not the answer

EDITOR,—If, in their editorial on clinical guidelines in the independent sector, Gillian Fairfield and Rhys Williams are suggesting that the guidelines used by American managed care organisations to authorise care are evidence based then they are misguided.1 In a national poll Group Health in Seattle was recently acknowledged to be “one of the best [managed care companies] at monitoring health, risk behaviors and disease outcomes.”2 Group Health has clinical guidelines that are strongly based on the effectiveness literature. They total just 13. Some health maintenance organisations offer incentives for limiting referrals and certain procedures through reviews of use. These “referral guidelines” have existed for a long time and are seldom evidence based; their intent is cost containment rather than cost effectiveness. If it is these to which the authors are referring then it would be a retrograde step for the NHS to adopt such a practice.

The authors correctly predict that private patients in Britain would not accept rationing by their insurers. Enrolment is dwindling in traditional health maintenance organisations with their own staff, which are renowned for their gatekeeping function. The insurers are busy developing alternatives that are more patient friendly and have names such as “options” or “alliant plus”; these even allow the choice of physicians paid on a fee for service basis—whose costs, ironically, managed care was set up to control.

Private insurers have no advantage over the NHS in the provision of useful data on outcomes. The independent health company BUPA only recently entered the primary care market, where the bulk of high cost and high volume long term care will occur in a primary care led NHS. Group Health has been host to many patient outcome research teams; despite the existence of data back to 1977 the lack of an appropriate coding system that links processes with outcomes is the barrier to more sophisticated research into outcomes. Read codes, which the Somerset morbidity project used,3 or new diagnosis codes may help.4

Surely the only place to experiment with evidence based guidelines is in the NHS. It is there that the real clinical controversies exist as doctors struggle with whether treatment A will be cost effective given patient B's age, sex, and socioeconomic and disease status, rather than in the private sector, where the concern is mainly elective surgery in people of higher socioeconomic status. Techniques of adjusting for case mix are increasingly being used to aid this complex process.5 Perhaps the private insurers' funds could best be used to support such an effort in the NHS if the interest is truly in collaboration.