Many have argued that good quality health care includes involvement of patients in decisions about their care. Furthermore, care should only be judged as appropriate if, as well as meeting professional and societal safeguards and concerns, patient preferences and patient values have been incorporated within the decision making process.¹ Few would disagree with this in principle. However, despite the importance of patient views and the value of engaging patients in all aspects of care being acknowledged within mainstream health policy, the changes needed in clinical practice and delivery of health care that will move policy from lip service to a reality have yet to be made.

Indeed, the changes needed will challenge some of the assumptions of health care and will raise many complex questions. For example, patients who do not wish actively to engage in decision making, particularly in taking responsibility for decisions on their treatment, may be inadvertently harmed despite our best intentions. Active debate about understanding patient preferences and how to allow real choice within routine practice needs to be supported by good research if patients are to benefit from being properly and appropriately engaged in decisions about their care, and not be exposed to the risks of an oversimplistic approach.

Furthermore, truly involving patients in treatment decisions will require fundamental review and reappraisal of some of the stated objectives of health care. The aggregate effects of informed or shared decisions on traditional health outcomes are hard to predict, but consider the case of hypertension treatment. The commonly stated objective of treating hypertension in a population of patients is to reduce the morbidity and mortality from conditions such as stroke and coronary heart disease. The traditional approach is to identify those at risk and to target treatment at those with a blood pressure over a defined threshold. Previously symptomless patients are labelled as “ill” and asked to take medications with common side effects on the basis of arguments centred on improving the health of the population, with health defined in this example as prevention of cardiovascular disease. Indeed, this is the approach promoted in recent national service frameworks²³ and British Hypertension Society guidelines.⁴

But what if patients themselves were to decide on their own treatment after receiving full information on the risks and benefits? Would they make the same decisions as their doctors would advise? Quite probably not, as shown by recent work quoted by Montgomery and Fahey in this supplement (page 39) which shows that patients were significantly less likely to want antihypertensive therapy than physicians, particularly when the baseline risk was low. If patients were then to make different decisions, what would be the effects of these decisions? Perhaps we would see patients with hypertension who are better informed, more satisfied with the decision made, less anxious, with greater autonomy and self-esteem, but collectively more prone to stroke and coronary heart disease. To accept that would require a major shift, not only in the way we interact with patients, with implications for training of clinicians and changed consultation methods, but also in the fundamental goals of health services. It also implies another major shift in the willingness of patients to take a greater responsibility for choosing between alternative courses of action and hence, ultimately, for the outcomes of treatment.

The papers in this supplement address some of the questions central to understanding how patient preferences can be genuinely incorporated into routine practice in a way that will benefit them. Most of the papers were initially prepared for a workshop sponsored by the MRC Health Services Research Collaboration (HSRC) which sought to pursue the whole issue of patient preferences by bringing together key researchers from across the UK. In addition, other writers have provided papers on related topics not covered explicitly in that workshop—for example, on the ethics of shared decision making (page 29). We thank all of the authors for their hard work in developing workshop discussion statements into papers for this supplement, and we thank the MRC HSRC under the leadership of Professor Paul Dieppe for sponsoring both the original workshop and this supplement. The patient perspective is a key component of the HSRC's work programme and this collection of papers will be an important contribution to it. We believe that this supplement, about a subject central to the delivery of good quality care in modern health services, is both timely and provocative. We hope you enjoy it and that it stimulates you to think about how to incorporate patient choice into health care.