Intended for healthcare professionals


Hand hygiene

BMJ 2001; 323 doi: (Published 25 August 2001) Cite this as: BMJ 2001;323:411

Use alcohol hand rubs between patients: they reduce the transmission of infection

  1. Louise Teare, chair, Hand Hygiene Liaison Group,
  2. Barry Cookson, director,
  3. Sheldon Stone, senior lecturer
  1. Chelmsford Public Health Laboratory, Chelmsford CM2 0YX
  2. Hospital Infection Laboratory, Public Health Laboratory Service, London NW9 5HY
  3. Academic Department of Geriatric Medicine, Royal Free Hospital, London NW3 2QG

    It is two years since the hand washing liaison group, a group of professionals interested in reducing the transmission of infection, drew attention in the BMJ to the importance of hand washing in reducing hospital acquired infections.1 The issue has again received prominence in Britain with the recent publication of the “Epic” evidence based guidelines on hand hygiene, commissioned by the Department of Health.2 The challenge now is to ensure implementation of the guidelines in daily practice. In recognition of the fact that washing with soap and water is not the only (or even the most effective) way of reducing the transmission of organisms our group has changed its name to the hand hygiene liaison group and has some practical recommendations on easy ways of improving hygiene.

    Publication of the Epic guidelines on preventing hospital acquired infection follows reports 3 4 documenting the seriousness of hospital acquired infection and antimicrobial resistance in the NHS. Hospital acquired infections in the United Kingdom cost around £1bn a year3 and affect nearly 10% of patients, causing over 5000 deaths a year (more than deaths on the road) and taking up thousands of bed days. Methicillin resistant Staphylococcus aureus, a surrogate marker for hospital acquired infection, is now responsible for 47% and 68%, respectively, of all cases of S aureus bacteraemia and surgical wound infection.5 The National Audit Office report suggested that the incidence of hospital acquired infection could potentially be cut by 15% and that hand hygiene recommendations should be implemented as part of the NHS's national plan.3

    Systematic review evidence, appraised and used by the Epic guideline developers, identified several well designed studies showing that patient contact resulted in contamination of health care workers' hands by pathogens.2 For example, staff dressing wounds with methicillin resistant S aureus have an 80% chance of carrying the organism on their hands for up to three hours. Another study showed that 40% of all patient-nurse interactions on an intensive care unit resulted in transmission of Klebsiella to the nurse's hands, even after minimal contact such as touching a patient's shoulder. Organisms remained on hands for up to 150 minutes. Similar data are available for Clostridium difficile. Hand washing removed the organisms.

    Formal handwashing with soap and water is required when there is soiling. When there is none the hand hygiene liaison group now advocates that staff should use an alcohol-glycerol hand rub between patients. Alcohol hand rubs are quick to use (10-20 instead of 90-120 seconds) and can be used while walking and talking. Thus they overcome objections to hand washing, including lack of time, lack of sinks, and skin damage. Indeed, a recent study has shown that such hand rubs cause less irritation than soaps.6 The Epic systematic review would appear to support this strategy because it shows that, though liquid soap and water decontaminate hands, 70% alcohol or an alcohol based antiseptic rub decontaminates hands more effectively for a wide variety of organisms, including S aureus, Pseudomonas aeruginosa, Klebsiella spp, and rotavirus.

    The evidence that hand hygiene reduces infection is strengthened by our own review of studies (excluding before and after observational studies and those that did not measure or enforce handwashing).7 This found nine studies (three randomised controlled trials, five controlled trials, and one multiple crossover trial) showing major reductions in infection related outcomes across a wide range of clinical settings. The effect is so great (commonly reported odds ratios and relative risks of 0.4) that if “hand hygiene” were a new drug it would be accepted without question.

    Mathematical modelling suggests that even small increments in hand hygiene may be highly effective in controlling, for example, endemic methicillin resistant S aureus. The risk of transfer on carers' hands is proportional to the power of the number of times a patient is touched.8 Given that chance plays a strong part in events on a small ward, it is apparent that even small increments of frequency of effective hand hygiene should reduce the risk.

    The issue is no longer whether hand hygiene is effective, but how to produce a sustained improvement in health workers' compliance. Our group and others 2 3 have recommended trials of behavioural and educational interventions that might achieve this goal. Indeed, feedback may be more effective than educational interventions, and the influence of senior staff is likely to be critical. A recent study improved compliance by 20% using feedback and encouraging the use of alcohol handrubs.9

    Where do we go from here? Firstly, all healthcare workers need to be aware of the current evidence underpinned by the new national guidelines2 and our own review.7 The need to integrate effective hand hygiene into clinical governance in association with risk management has been highlighted recently10: every trust should have as a standard that alcohol handrub is available at every bedside, and hospital acquired infection should be one of the key performance indicators because it is an important marker of the quality of patient care. Long term change in behaviour requires that all staff, especially senior staff, take responsibility for ensuring that hand hygiene becomes an every day part of clinical culture.


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