Some methodological problems in the assessment of complementary therapy

The increased interest of the public and the medical profession in complementary forms of therapy introduces an urgent need for proper assessment of efficacy. This growth in popularity suggests that complementary therapies are effective in some circumstances, but without objective assessment neither the public nor the medical profession can be sure. Rigorous scientific assessment is required, but the nature of complementary therapy is such that double‐blind randomized controlled trials, as usually conducted, are rarely applicable, and alternative approaches are needed. At its best, complementary therapy is designed individually for each patient, and few treatments can readily be applied ‘blind’. These are the main problems in trial design but complementary therapists are also aware of the importance of body‐mind interactions: they see the ‘placebo effect’ as an integral part of treatment which needs investigation, and they question the validity of clinical trials in which the patient's will is not fully engaged through lack of information, and of those not asking the patients if they feel better. The problems are examined and alternative approaches suggested. The paper raises the question of whether they are only problems for complementary medicine.