Intended for healthcare professionals


Compliance becomes concordance

BMJ 1997; 314 doi: (Published 08 March 1997) Cite this as: BMJ 1997;314:691

Making a change in terminology produce a change in behaviour

  1. Patricia Dolan Mullen, Professora
  1. a Center for Health Promotion Research and Development, University of Texas School of Public Health, Box 20186, Houston, Texas 77225, USA

    At long last the “compliance problem” may be getting a new name and, with it, a new view of the patient's role in the doctor-patient relationship. A report published this week by the Royal Pharmaceutical Society of Great Britain's working party on medicine taking recommends that “concordance” should replace the term “compliance.”1 Although substitute terms have been suggested and used previously without much impact, this eloquent analysis of the importance of a new concept by a highly visible and distinguished panel may hold the promise of change. Moreover, the panel recommends a £1.8m ($2.7m) research budget to support analysis of the problem and training of health professionals.

    Compliance has long been criticised as denoting obedience-“following doctors' orders.” Although many researchers and practitioners have carefully avoided the term,2 the common alternatives-“adherence” or “cooperation”-do not take the user very far from compliance. One member of the working party, David Sackett, in his 1976 landmark publication, Compliance with Therapeutic Regimens, had already anticipated the approach advocated by the Pharmaceutical Society's report.3 Included in the book were sensible ideas such as the “tailored consensual regimen,” the need for a no fault approach to behaviour relating to following a regimen,4 and consideration of the effect of frequency of administration, side effects, delivery system, and the like-all aspects of the medication that affect compliance.5 A subsequent publication raised the idea of a clinically relevant definition of adherence, based on the properties of a particular drug and not solely on the doctor's instructions.6 Thus if seven days are sufficient to achieve the therapeutic effect of a drug then patients who stop the medication after “only” seven, eight, or nine days should not be deemed to be non-compliant even if it was prescribed for 10 days.

    Despite the predominance of the term compliance, interventions have not all been aimed at the patient. Manufacturers, for example, have responded with less complex delivery mechanisms such as patches, more convenient doses such as sustained release drugs, incentives for patients to fill their first prescription and to get refills, advertising to increase the perceived value of the drug, and direct patient education. In part these developments indicate that the manufacturers understand clearly the effects of non-compliance on drug sales. But they also suggest that clinicians value drugs with features that enhance compliance.

    Changes in drug trials reflect recognition of the effect of compliance on statistical power and interpretation of results.7 8 After initial resistance to including compliance experts, some investigators began to include them when planning the trial rather than attempting to fix problems later.9 Another approach has been pre-randomisation screening of potential participants, usually with a placebo. This approach assumes that non-compliance is a general characteristic of the person: thus a pretrial test of drug taking can reveal non-compliers.

    The change in terminology will have an impact only if the culture change that the working group is advocating succeeds and clinicians take a more egalitarian view “of the relationship between prescribing and medicine-taking, between patient and prescriber.”10 It is possible to envisage doctors and patients engaging in more productive discussion of medication regimens, but the barriers are substantial. A prescription is a traditional means of ending a consultation, after most of the time has been spent on diagnosis. Perfunctory questioning about the drug at the next visit may lead patients to assume that the doctor does not place high priority on drug taking. Clinicians may simply assume the patient's compliance and see any continuing symptoms as indicating the need for more or a different medication.

    More time spent should be spent assessing not only the best medication for a particular condition but also the best for a particular individual with a certain lifestyle and preferences. The concept of concordance suggests frank exchange of information, negotiation, and a spirit of cooperation. Compared with the US, conditions in Britain favour this approach. Patients and practitioners are more likely to have known one another for longer, dispensing is less impersonal, and ancillary personnel are available for follow up.

    Moreover, evidence of effectiveness is available from rigorous trials, and no single method of improving compliance appears to be inherently superior. 11 With coaching and a non-judgmental attitude from the prescriber, patients are more likely to describe drug taking truthfully. Patients can be informed about dosing options and asked what would work best for them. Initial prescriptions can be regarded as a trial, not only of the drug's effect but also of the feasibility of taking it. Treating the patient as a decision maker is a fundamental step away from the compliance model.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.