Intended for healthcare professionals


Prescribing and taking medicines

BMJ 2003; 327 doi: (Published 09 October 2003) Cite this as: BMJ 2003;327:819
  1. Giselle Jones, papers editor (gjones{at}
  1. BMJ

    Concordance is a fine theory but is mostly not being practised

    About half of the medicines prescribed for people with chronic conditions are not taken.1 This is assumed to have large personal, economic, and societal costs,2 but are patients really any worse off for not taking their medicines as prescribed? Doctors say that not taking drugs means poorer health outcomes, but patients argue that only they can know what works for them and what doesn't. Bridging the gap between the agendas of patients and doctors changed the emphasis from the authoritarian concept of compliance to the more inclusive concept of concordance. Concordance means shared decision making and arriving at an agreement that respects the wishes and beliefs of the patient. What it should not be is a more gift wrapped version of compliance.3

    The concordant model of shared understanding and prescribing should improve health outcomes. Why then has it proved so hard to put into practice? One problem is that most health professionals either have not heard of the term, or they don't understand it. This is not surprising when so little evidence exists on the information needs of patients and prescribers, on how to improve communication skills, and on whether better communication does improve outcomes.4 5

    Another problem is that shared decision making based on shared understanding relies on the provision and exchange of accurate, relevant, and accessible information. The advent of the internet has vastly increased the availability of information, but often what it offers is untailored, incomplete, irrelevant, and plain wrong. Information can be contradictory—even simple instructions from a doctor on how and when to take medicine can be contradicted only 10 minutes later by a pharmacist. High quality information is essential if concordance is to succeed6 but in itself will be useless without the ability to share and use it.

    The perception of patients and doctors of the risks and benefits from medicines differs, especially when medicines are being used to prevent rather than treat symptoms.7 The study by Lewis and others shows that patients and health professionals find the concepts of risk and benefit difficult to understand, and in deciding whether to start preventive treatment doctors were more likely to accept smaller benefits than their patients.7 That patients do not persist with regimens that cause adverse effects balanced against only hypothetical benefits is hardly surprising. Evidence based guidelines on treatment will be useful only if the subjective element of patients' preferences and values is acknowledged and explored.

    Concordance poses ethical and legal challenges when a decision is made that is ultimately detrimental to a patient's wellbeing. At what point does accepting such a decision fall outside of a doctor's foremost duty to do no harm? This is further complicated when a patient's dissent to treatment has the potential to threaten the health of others—for example, with tuberculosis.8

    All the challenges to concordance are most difficult when the patient is a child—communicating well enough to allow an equal contribution to decision making is difficult because third parties—parents—are involved.9 When can children take responsibility for their medication? Should more weight be given to the parents' wishes or the child's? What if the parent, child, and doctor all disagree? Although we can go some way to answering these questions, little evidence is available.9 But we should not assume that the same rules for achieving concordance in adults apply to children.

    So what next for concordance? For it to be a success a change in practice is required. Because concordance is an evolving concept and the sum of many parts it is a difficult subject to research in its entirety. The emphasis should turn to patients' needs and wants and how they influence the way they take medicines. More evidence on the specific elements of the prescribing process and how they interact is required. Patients need information that speaks in clear terms that reflect what matters to them, and prescribers need practical tools to use information in the consultation. The biggest challenge for concordance and the most difficult to research will be a change in values.


    • Competing interests None declared.