Intended for healthcare professionals


Non-attendance at general practices and outpatient clinics

BMJ 2001; 323 doi: (Published 10 November 2001) Cite this as: BMJ 2001;323:1081

Local systems are needed to address local problems

  1. Deborah J Sharp, professor of primary health care ({at},
  2. William Hamilton, general practitioner and research fellow (w.hamilton{at}
  1. University of Bristol, Bristol BS8 2PR
  2. 12 Barnfield Hill, Exeter EX1 1SR

    “Non-attendance at NHS outpatient clinics and at general practices is more common in deprived populations. It results from the organisation of these services, which currently puts the needs of staff before those of patients. Discuss.” Proposing this debate at your next management away day should enliven proceedings. What evidence exists for the various parts of the statement?

    The national figure of 12% for non-attendance at outpatient clinics in the United Kingdom hides large variations between specialties and between regions. Studies report figures that range from 5% to 34%. 1 2 Much less research has been done on non-attendance in general practices, though figures of 3% and 6.5% have been reported. The first figure comes from an unpublished doctor-patient partnership survey in 1998 and an unpublished survey (by WH) of 500 non-attenders in Exeter. The higher figure is from a study of 221 000 appointments in practices in Sheffield.3

    Different methods have been used to assess deprivation, such as extrapolation from postcode data, information from interviews or postal questionnaires, and validated indices of deprivation.47 The main associations with hospital non-attendance are reported as being male sex, youth, the length of waiting time for the appointment, and deprivation. 1 48 Multivariable analysis of the 2072 patients referred to outpatient clinics in our prospective study showed these four factors to independently predict non-attendance at outpatient clinics (unpublished data).9 Non-attendance in general practices is associated with youth and deprivation (assessed by the referring practice's Townsend score) but not sex.3

    Non-attenders are less likely to own a car or a telephone and are more likely to be unemployed. 4 6 7 10 The finding about telephones may be pertinent, because some non-attendance arises from an inability to cancel the appointment, either because the hospital's system for cancelling or changing appointments is poor or because the patient has no access to a telephone. 7 11 Non-attendance is not thought to be related to the severity of the patient's condition, except in the case of psychiatric illness, where non-attendance may be a marker of severity of illness. 6 8

    Questionnaire surveys of non-attenders provide further evidence for the link with deprivation. 7 8 The commonest reasons cited for missing an appointment, after forgetting it, are family or work commitments. Patients with lower paid jobs may have difficulty in getting time off work or arranging childcare. These reasons also partly explain the peak age range of 20–30 in non-attenders, as this is the usual age for raising a family. 3 4

    The organisation of clinics and surgeries gives further insight. The strategy of overbooking appointments to allow for anticipated non-attendance may be counterproductive. Attendance of 100% puts pressure on both patients and staff. Also, overbooking means that the appointment time is rarely met, so patients have to clear their commitments for the whole morning or afternoon. For some people this is impossible, and for others the difficulty may be enough to tip the balance towards not attending.12

    The key seems to be to allow the patient to select a suitable time and date—indeed, such flexibility may largely explain the lower non-attendance rate in general practices, although evening surgeries and shorter time intervals to the appointment probably also contribute. No hospitals have reported their experience of offering evening or weekend clinics. Several hospitals have instituted systems offering patients a choice of time and date. 11 13 Some clinics offer telephone reminders, although the issue of confidentiality of telephone calls has not been examined. 2 11 These strategies have resulted in reductions in non-attendance of up to 60%.13 No formal economic analyses have been published—and they are necessary—but some hospitals report that the additional staff and telephone costs outweigh the efficiency gains.11 This view may be short sighted, as simply adding up the costs misses two crucial gains. Firstly, the strongest predictor of non-attendance is the time interval to the appointment; reducing non-attendance reduces waiting times, which further reduces non-attendance, creating a virtuous circle.4 Secondly, systems that require input from the patient before the appointment takes place identify a number of patients who fail to reply. This failure may arise either from administrative errors such as wrong addresses or because patients are particularly disadvantaged by deprivation or illness.6

    However, increasing the flexibility and therefore the complexity of appointment systems carries a risk: patients most in need may be disempowered. Patients who cannot read English, because of learning difficulties or cultural background, or patients with sensory disabilities may struggle with newer systems. As long as these local factors are known, individual solutions can be tailored, such as avoiding making appointments for Muslim patients on important religious dates or in the wrong language.2

    So, how should the managers at the away day answer our provocative question? Increased consumerism in the NHS means that current systems are stale, at best. No single solution will work across the NHS and in outpatient clinics as well as general practices. Local trusts in primary and secondary care should be able to devise local systems to allow convenient access for their patients. Any new systems should be the subject of research and development. If some of these measures are adopted non-attendance should fall, though it will never disappear—we are all human.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract