Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Washing hands reduces hospital-acquired infection

Unhealthy economics

Bandolier 73 examined the importance of hospital acquired infection to health services. In England it costs £1 billion a year, kills 5000 patients, and consumes the resources of 27 four hundred bed hospitals. Effective handwashing can reduce that substantially ( Bandolier 67 ), though there are few reports of effects over long periods. A long-term hospital wide study, demonstrating that an effectively implemented handwashing policy reduced hospital acquired infection by half and saving precious resources [1], is welcome.


The study was started in the mid-1990s in the University of Geneva Hospitals. Handwashing behaviour was monitored and the incidence of hospital acquired infection measured. After a baseline survey the interventions began in January 1995. At several times from 1994 on, trained staff observed health professionals at prespecified time periods throughout the hospital. An additional measure of efficacy was the amount of alcohol-based handrub solution dispensed by the pharmacy. Hospital acquired infection was measured as the annual prevalence identified by infection-control nurses using CDC standard definitions, and the number of new cases of MRSA infection.


The components of the interventions included:


The proportion of observed instances where necessary handwashing did indeed take place increased from 48% in 1994 to 66% by the end of 1997. Moreover, there was a change in behaviour, with hands being more often disinfected than just washed. This was demonstrated by a fivefold increase in the amount of alcohol-based handrub used over the period (Table 1). Hospital-acquired infection prevalence fell from 17% in 1994 to 10% in 1998 (Table 1), New cases of MRSA fell by half (Figure 1).

Table 1: Use of hand disinfectant and hospital acquired infections

Year Alcohol-based handrub L/1000 patient days New MRSA per 100 admissions Hospital acquired infection (% patients)
1993 3.5 0.50 n/a
1994 4.1 0.60 16.9
1995 6.9 0.48 17.5
1996 9.5 0.32 14.5
1997 10.9 0.25 9.0
1998 15.4 0.26 9.9

Figure 1: New MRSA cases before and after the hand-washing programme started


The cost of the programme was estimated at no more than SFr 380,000 (£155,000; US$270,000), including direct and indirect costs. The authors conservatively assumed that 25% of the observed reduction in infections was due to the hand washing programme, thereby preventing 900 infections. At an average cost estimate of about £1400 per infection, their estimate of overall savings of £1,260,000 far outweighed the costs.


A UK report ( Bandolier 73 ) estimated that a case of hospital acquired infection cost an extra £3150 and consumed an average of 14 extra bed-days. One of the main consequences is bed blocking, and massive inefficiency in hospital systems. Reducing hospital acquired infection by half, exactly as seen in a UK example ( Bandolier 67 ) would be equivalent to building, staffing and running 13 more 400 bed hospitals in England.

The characteristics of this study, apart from its typically Swiss thoroughness, are the same as examples in Bandolier 's sister publication, ImpAct . They are team work, ownership, audit, feedback, and a long-term commitment to improvement. That aside, there's nothing particularly clever being done. It really is easy, and it is done in most industries as a matter of course.

All hospital chief executives should read this paper several times. They should ask why their own institution is not already doing this. With this example, and in the full and certain knowledge that it saves resources, saves money, and is better for patients, there is no excuse not to follow the Swiss example. If they don't implement it the retirement cuckoo clock should be in the post.

Unhealthy economics

And that is where this particular article ended, until, that is, we met a man upon the stairs. This particular man was one of life's cynics whose attitude was something like - 'you never save anything in the NHS, so it's not worth bothering'. Hang on a minute: that is unhealthy economics, so the argument needs extending to demonstrate the benefits in even starker detail.

We think the argument goes something like this. The handwashing programme saved 900 infections. The average additional cost of a patient with hospital-acquired infection was £3154 and their average additional hospital stay was 14 days from a UK report ( Bandolier 73 ). So 900 infections would consume £2.85 million, and 12,600 bed days more than if these patients did not have a hospital-acquired infection. Ninety of the 900 patients would die.

We could spend that wasted £2.85 million on a package that would include:
  1. a hospital infection programme costing £155,000
  2. care for an extra 1575 patients without hospital acquired infection using the 12,600 bed days saved at a cost of £2.56 million (1575 times £1628, or 8 days, from Bandolier 73 for average patients without infection)
  3. £120,000 to spend in other ways (staff training or more initiatives to make the hospital better, perhaps)
  4. fewer deaths in hospital.

Now this is all back-of-envelope stuff. But even in this simple examination the concept of hand-washing programmes can be seen as being no-brainers. This type of analysis should make the man on the stairs go away.

But no. The man on the stairs laughes uproariously, and tell us that we would never get the £155,000 new money because there is no budget for it. He thinks that budget-based medicine rules.

It shouldn't. It is possible to treat more patients better with the same resources. If it is possible, it should happen. We can measure the effectiveness of handwashing programmes through infection rates, deaths, and patient throughput. It isn't as if we want the £155,000 all at once. It is spread over several years in a big institution with a huge budget.

One of the benefits of a thinking time directive would be to generate time to think through the implications of doing the simple things right - an extension of doing the right things right, and at the right time. Handwashing is but one example of how to do more with less.


  1. D Pittet et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet 2000 356: 1307-1312.
previous or next story in this issue