Traditionally, adherence to medical advice has been conceptualised as compliance. This is now seriously out of fashion. Indeed, an otherwise sympathetic academic reviewer of our proposal for this book was appalled that we intended to use the term ‘compliance’ at all. There are a number of good reasons to reject compliance as a desirable objective. Compliance implies a passive submission to the doctor's will. It is based on the assumption that medical advice is necessarily good advice, which is not always the case (in the fullness of time, accepted medical wisdom can prove to be wrong, for example, the use of bed rest for low back pain, or prefrontal leucotomy for schizophrenia). Even when advice is sound, individual circumstances can make it inappropriate or impossible to implement. In any case, the gap between the level of education and information of doctors and patients has narrowed. The authoritarian, all-knowing doctor directing the deferentially compliant patient was never a good model, and it is untenable in the present day.

Much of the modern literature concerning adherence to treatment plans suggests that the concept of compliance should be replaced by concordance. Concordance means that the clinician and patient agree on a treatment plan and adhere to it. When the patient fails to take prescribed medication, the problem can be understood in a number of different ways, which prominently includes failure to achieve true agreement between psychiatrist and patient over the treatment plan in the first place.