The recent release of data about how well hospitals perform when it comes to ensuring staff wash their hands drew widespread media coverage.
In the article below, Professor Lyn Gilbert, an infectious diseases physician and clinical microbiologist with a strong interest in preventing healthcare-associated infections, provides some of the wider historical context.
Making sense of hand hygiene data
Lyn Gilbert writes:
The MyHospitals website posted hand hygiene compliance data, by hospital, for the first time, on March 6th. It’s interesting reading but easily misinterpreted without some background. Hand hygiene is a simple concept but compliance data are complex.
Everyone knows (more or less) that hand hygiene by healthcare workers can prevent transmission of germs and reduce infections – and the more the better.
A quick survey of MyHospitals data from 22 large public hospitals (with more than 500 beds), in all states, shows reported compliance rates between 61% and 83%.
The national benchmark is 70%, but many believe it should be higher. What do these rates actually measure and why is the benchmark not 100%?
Although most people accept the concept that hand hygiene can reduce infection risk, there are widely differing views about when and how often it is necessary.
The concept goes back at least to the mid-19th century and predates the germ theory of disease. At the Vienna General Hospital, the professorial assistant in Obstetrics, Ignác Semmelweis, was troubled by the fact that women whose babies were delivered in the clinic staffed by doctors and medical students were three times more likely to die from childbed fever than women admitted to the midwives’ clinic.
The only significant difference between clinics was that medical students, but not midwives, performed autopsies on women who had died from childbed fever. Semmelweis suspected that they carried “cadaverous particles” on their hands to the women in labor, thus spreading infection. He ordered the students to wash their hands in chlorinated lime (calcium hypochlorite) when they left the mortuary to go into the labour ward and the childbed fever rate fell within a month to one similar that in the midwives’ clinic.
The evidence was not well received and Semmelweis was angrily ridiculed by many of his colleagues, who could not conceive that they could be responsible for causing the deaths of patients they cared for. He was not the first to suggest that childbed fever was contagious, but no one before had produced evidence that it could be prevented by hand washing.
During the 20th century, with the development of vaccines and antibiotics, infectious diseases were thought to be a thing of the past.
Asepsis in the operating theatre, including the surgical hand scrub, continued, but the general idea of hospital hygiene – spotless wards staffed by nurses in spotless uniforms overseen by a matron in a starched veil – lost currency and hospitals became more informal, accessible and “friendly”.
Hand washing by healthcare workers was still expected, of course, especially after examining a patient with an infection or if hands were obviously soiled.
More than that was difficult, even for the most conscientious healthcare workers; hand basins were sparsely distributed, so hand washing after contact with every patient disrupted workflow and took time that no-one had to spare in a busy ward.
In 2001, the Institute of Medicine’s report “To Err is Human” pointed out that more people died in the USA, from preventable medical mishaps in hospitals, than from motor vehicle accidents, cancer or AIDS.
The report made little mention of hospital-acquired infections (HAI), but quoted Centers for Disease Control estimates, that 2 million patients developed HAI in acute care facilities in the USA, annually, at a cost to the health system of US$3.5 billion1. Moreover, it was estimated that at least a third (or up to 70%, depending on the type of infection2) were preventable by simple, inexpensive infection control measures, such as hand hygiene.
At the same time there was mounting alarm about increasing rates of antibiotic resistant bacteria, causing HAI and deaths in the most vulnerable patients – who are immunosuppressed, severely injured or recovering from major surgery.
Treating these infections was increasingly difficult and expensive; some were resistant to all available antibiotics. One effective way to reduce the risk was to protect patients from being exposed to these bacteria, by stopping their spread on the hand of healthcare workers.
In Australia, a number of reports by working parties and expert groups on antimicrobial resistance and surveillance of HAI were commissioned by the Australian Government, between the 1980s and early 2000s, but there was little action until the Australian Commission on Safety and Quality in Healthcare (ACQSHC) was formed in 2005 and determined that HAI would be one of its priority areas for 2007-10.
An ACQSHC publication on surveillance of HAI3, highlighted hand hygiene as a major strategy to reduce the estimated 200,000 HAIs occurring in Australian hospitals each year.
The renewal of interest in hand hygiene was timely. In 2000, researchers in Geneva led by Professor Didier Pittet demonstrated that the easy availability of alcohol-based hand rub (ABHR) significantly increased compliance with hand hygiene. This was associated with a fall in the rates of nosocomial infection and transmission of antibiotic resistant bacteria4. AHBR is as effective as soap and water, more convenient to use – because it can be placed at each bedside – and causes less skin irritation.
These findings were the basis of the WHO’s “Clean Care is Safer Care” program, a major component of which is the “SAVE LIVES: Clean your hands” campaign, which aims to have 15,000 healthcare facilities signed up to improve hand hygiene, by May 2012 (so far 14,825 have enrolled).
Hand Hygiene Australia was engaged by the ACQSHC, in 2009, to implement the National Hand Hygiene Initiative (NHHI), to improve hand hygiene and reduce HAIs in Australian public acute healthcare facilities. To do this and to show it was being done, required standardisation of hand hygiene practice, a consistent, accurate way to measure compliance and an objective marker of HAI incidence.
The WHO’s “5 Moments of Hand Hygiene” (see Figure) was chosen as the standard hand hygiene system for Australia, which was one of the first countries to sign up to the WHO program.
The rate of healthcare-associated Staphylococcus aureus bloodstream infection (SABSI)5 was chosen as an objective, easily measured marker of compliance with infection control practice, including hand hygiene.
Implementing the NHHI required all States and Territories to sign up to the program. Some were initially reluctant to do so because they already had well-established hand hygiene programs and while they have all now agreed, implementation of the program has been delayed in some States.
The next step was to train “gold standard auditors”, to ensure that compliance would be measured consistently and data comparable between facilities. This has been a mammoth task involving face-to-face workshops in all major cities and significant time commitment by many “frontline” staff. These auditors are responsible for training ward auditors, who perform audits and report a designated total number of “moments” regularly to Hand Hygiene Australia.
The data reported on the MyHospitals website recently were the results of audits reported in the third audit period for 2011, from 589 hospitals, covering more than 300,000 “moments”. There is still variation between States and Territories and individual facilities, in the numbers of “moments” audited by different hospitals and rates of compliance.
The overall national compliance for this period was 73%, a significant improvement on the first reporting period in 2009, when it was 64%. More detailed analysis shows considerable variation in compliance between “moments” – for example it is generally higher after than before contact. There are also differences between professional groups – from 58%, overall, for doctors, to 77% for nurses.
Healthcare-associated SABSI rates were first reported on the MyHospitals website in September last year, for the period June 2010-June 2011. They include SABSIs acquired 48 hours or more after admission to hospital (hospital onset) and healthcare-associated outpatient SABSI (defined by a series of criteria including a recent admission or procedure or the presence of some type of indwelling device, such as a venous catheter).
The reliability of these early data is patchy. There are agreed case definitions, but anecdotal evidence suggests that they are interpreted differently. It requires a detailed chart review, by an experienced clinician, to decide whether a SABSI in an outpatient is truly healthcare-associated and, if so, which facility it “belongs” to.
Until hospitals become more experienced with reporting, some of these may be misclassified as community-associated and not counted in calculation of rates based on the numbers of occupied bed days.
The greatest value of these “report cards” is for the facilities themselves, to see how they compare with similar hospitals. There are many factors that determine SABSI rates, apart from hand hygiene (although it is clearly important).
The challenge, for every clinician and every administrator in every facility, is to understand what factors contribute to the occurrence of SABSIs on their watch and act to ensure that none occurs that could have been prevented.
For Health Departments, politicians and the public the important information is not the rate for one time period, but trends. As the data are better understood and local responses implemented, they should improve. Already the National Hand Hygiene Initiative has improved overall hand hygiene compliance in Australian hospitals6.
Demonstrating a reduction in infection rates will be more difficult – a reported reduction in methicillin resistant Staphylococcus aureus (MRSA) infection is promising, but occurred in some facilities even before the NHHI was implemented and has been offset by increases in methicillin susceptible SABSI, which are easier to treat, but still serious.
Nevertheless, public disclosure of these data is a powerful incentive for improvement and a step towards an important goal – to make hospital infection prevention and control “everybody’s business”, not just the responsibility of a few infection control specialists.
• Lyn Gilbert is Clinical Lead, Infection Prevention & Control, Western Sydney Local Health Network, and Director, Centre for Infectious Diseases & Microbiology-Public Health
- Kohn LT, Corrigan JM, Donaldson MS (Eds). To Err is Human, Building a Safer Health System. Institute of Medicine, Washington DC, USA. 2001
- Umscheid CA, Mitchell MD, Doshi JA, Agarwal R, Williams K, Brennan PJ. Estimating the proportion of healthcare-associated infections that are reasonably preventable and the related mortality and costs. Infect Control Hosp Epidmiol, 2011; 32:101-114.
- Cruickshank M, Ferguson J (Eds). Reducing harm to patient from health care associated infections: the role of surveillance. Australian Commission on Safety and Quality in Healthcare. Sydney, Australia. 2008. http://www.safetyandquality.gov.au/internet/safety/publishing.nsf/Content/con-pubs-hai-report
- Pittet D, Hugonnet S, et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet, 2000;356;1307-12
- Collignon, P. J., I. J. Wilkinson, et al. Health care-associated Staphylococcus aureus bloodstream infections: a clinical quality indicator for all hospitals. Med J Aust. 2006; 184: 404-406.
- Grayson, M. L., P. L. Russo, et al. Outcomes from the first 2 years of the Australian National Hand Hygiene Initiative. Med J Aust 2011; 195: 615-619.