Intended for healthcare professionals


Commentary: Best practice in primary care

BMJ 2006; 333 doi: (Published 20 July 2006) Cite this as: BMJ 2006;333:173
  1. Pippa Oakeshott, reader in general practice (oakeshot{at},
  2. Phillip Hay, reader in genitourinary medicine1
  1. 1 St George's Hospital, University of London, London SW17 ORE
  1. Correspondence to: P Oakeshott

    A well done (though necessarily unblinded) trial from the United States shows that women attending for cervical smears feel less vulnerable and have less physical discomfort if a method that doesn't require stirrups is used.1 The quality of the smears did not differ, and around half the women were from ethnic minority groups. This trial should change practice in the United States, where many women may be unaware that there is an alternative to using stirrups in cervical screening.

    By contrast, in the United Kingdom most speculum examinations for routine cervical smears are done in general practice or family planning clinics and stirrups are not used. Use of stirrups is mainly confined to hospital colposcopy and genitourinary clinics, and leg supporter boards are increasingly preferred.

    “For women, the vaginal speculum has loomed large, and has long signified a kind of scrutiny and intrusion [that] they have feared.”2 An unpleasant experience of vaginal examination for a first smear may make women extremely reluctant to attend for cervical screening in future. Examination should always be done by a doctor or nurse who is skilled, sympathetic and gentle.3 All health professionals should practice the basic principles of respect, privacy, explanation, and consent for intimate examination (box). These principles are increasingly incorporated in medical and nursing education.4

    Suggested guidelines for conducting vaginal examinations in primary care3

    • Explain the reason for doing a vaginal examination and obtain verbal consent

    • Offer to find a chaperone and record this in the notes

    • Provide privacy to undress and use drapes to maintain the patient's dignity

    • Use a closed room and avoid interruptions during the examination

    • During the examination: be gentle, explain what you are doing, be alert to indications of distress, avoid personal comments

    In the UK, cervical screening rates have been shown to be better in practices that have a female partner.5 Improved coverage in deprived areas has also been associated with an increase in the number of practice nurses, who are often the main providers of cervical screening in general practice. Uptake tends to be lower in practices with more patients who are socially deprived or from ethnic minority groups,5 and non-responders may be at increased risk of cervical cancer.

    There are alternatives for women who find a conventional speculum examination unacceptable. An Australian study of women attending family planning clinics found that 67% (133/198) agreed to insert their own speculum, and of these, 90% would choose to do it again. The main barrier was women feeling unsure how to self insert a speculum.6 In future, screening might be based on detection of specific human papillomavirus (HPV) subtypes and additional biomarkers for risk of cervical cancer. This might allow the use of self-taken vaginal samples, which could be done either in the clinic or at home. Although response rates might be low, non-responders to cervical screening could be sent postal swabs, and women who are found to have persistent infection with HPV 16 or 18 could be invited to attend for further evaluation.

    The paper by Seehusen and colleagues should change clinical practice away from the routine use of stirrups. If cervical screening becomes more user friendly, this could lead to increased coverage. The study also highlights the need for doctors and nurses to respect the patient's integrity when doing vaginal examinations, and shows how trials can be used to assess issues that are important to patients.


    We thank Phyllis Moore and Sima Hay for advice.


    • Funding BUPA Foundation.

    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    View Abstract