Skip navigation

Now wash your hands

The relevance of hand washing to hospital acquired infection has long been of interest to Bandolier . Clearly it is considered an issue of importance, and a BMJ editorial [1] pointed out how infrequent and sporadic hand washing is in health care workers. One classic paper quoted showed that while doctors estimated that they washed their hands 73% of the time before patient contact, the observed frequency was just 9%.

That hands can be a repository of nasty germs, and that hand washing can reduce infection rates has been known for a long time. A cracking little paper [2] showed that Klebsiella species can live quite happily on hands for up to two and a half hours, and that instituting hand washing regimens can reduce infections in patients by a significant amount (down from 23 to 16%). Hand washing with chlorhexidine regularly gave 98-100% reductions in hand counts.

There is even a systematic review [3] looking at compliance issues with hand washing and barrier precautions, which is a useful source of literature. It highlighted two other studies which showed that increased compliance with hand washing before and after patient contact resulted in large (50% or more) decreases in infection rates. Three more recent papers of interest are examined here.

Iowa [4]

Over eight months a multiple cross-over study was conducted in which two hand washing systems, chlorhexidine and soap plus alcohol rinse, were compared. The primary outcome was the patient nosocomial infection rate.

By observation, the proportion of opportunities for hand washing (after one patient and before the next) in which hands were actually washed was 42% during chlorhexidine use and 38% for soap/alcohol. With chlorhexidine during 4001 patient days there were 152 infections. With soap/alcohol during 3984 patient days there were 202 infections. This was statistically lower.

London [5]

Following cases of Clostridium difficile and methicillin-resistent Staphylococcus aureus in three acute medical wards for elderly people, infection control measures were monitored to examine the effect on infection, and use of cephalosporin antibiotics. The interventions were:

The result of the infection control policy was to reduce the amount of cephalosporin drug use by more than two-thirds. Compared with the nine months before the intervention, the nine months following the intervention resulted in a 42% drop in C difficile infections and a 51% reduction in MRSA infections (Figure). For MRSA this did not include a period when an isolation unit was in action during March to June 1995.

One year later this policy had had a significant impact on ward closures and bed availability. There was a net gain of 5 beds a day.

Geneva [6]

So just how contaminated do hands get during episodes of routine patient care? A study in Geneva answered this question by examining the previously-washed hands of health care workers by looking at the number of colony forming units (CFU) from the five fingertips of a dominant hand. There were 417 episodes, lasting a median of five minutes.

The total number of counts ranged from 0 to 300 CFU, with a mean 100 CFU. As well as normal skin flora, 11% were contaminated with Staphylococcus aureus and 15% with gram negative bacteria. Gloved hands acquired only 3 CFU per minute, while ungloved hands acquired 16 CFU per minute. The longer the episode of patient care, the larger the number of bacteria acquired.

For ungloved hands, direct patient contact, respiratory care (endotracheal tube), handling bodily secretions and episodes of care interrupted by a telephone call, for instance, were associated with highest rates of acquisition, from 16 to 20 CFU per minute. Most types of hospital setting were associated with similar rates of acquisition except septic orthopaedic and paediatric intensive care. Prior hand washing with antiseptics containing chlorhexidine was associated with lower rates of bacterial acquisition.


The bottom line appears to be simple, that hand washing with agents containing chlorhexidine really does lower the rate at which bacteria get onto the hands, and that hand washing protocols properly enforced reduce hospital acquired infections. It's more than that, though. There would appear to be clear evidence of effectiveness, and of benefits of quality and cost-effectiveness. It may even be a clinical governance issue.

Bandolier and ImpAct would love to hear from people who have found ways in making a policy work, in addition to that from London above.


  1. Editorial. BMJ 1999 318: 686.
  2. M Casewell, I Phillips. Hands as a route of transmission for Klebsiella species. BMJ 1977 2: 1315-7.
  3. E Larson, EK Kretzer. Compliance with handwashing and barrier precautions. Journal of Hospital Infection 1995 30 (Supp): 88-106.
  4. BN Doebelling et al. Comparative efficacy of alternative hand-washing agents in reducing nosocomial infections in intensive care units. New England Journal of Medicine 1992 327: 88-93.
  5. SP Stone et al. The effect of an enhanced infection-control policy on the incidence of Clostridium difficile infection and methicillin-resistant Staphylococcus aureus colonization in acute elderly medical patients. Age and Ageing 1998 27: 561-568.
  6. D Pittet et al. Bacterial contamination of the hands of hospital staff during routine patient care. Archives of Internal Medicine 1999 159: 821-6.
previous or next story in this issue