Article Text
Abstract
Background Effective hand hygiene has long been recognised as an important way to reduce the transmission of bacterial and viral pathogens in healthcare settings. However, many studies have shown that adherence to hand hygiene remains low, and improvement efforts have often not delivered sustainable results. The Children's Hospital at Westmead is the largest tertiary paediatric hospital in Sydney, Australia. The hospital participated in a state-wide ‘Clean hands save lives’ campaign which was initiated in 2006.
Intervention Strong leadership, good stakeholder engagement, readily accessible alcohol-based hand rub at the point of patient care, a multifaceted education programme, monitoring of staff, adherence to recommended hand hygiene practices and contemporaneous feedback of performance data have significantly improved and maintained compliance with hand hygiene.
Results Hand hygiene compliance has increased from 23% in 2006 to 87% in 2011 (p<0.001). Sustained improvement in compliance with hand hygiene has been evident in the last 4 years. A decline in a set of hospital-acquired infections (including rotavirus, multiresistant organism transmission, and nosocomial bacteraemia) has also been noted as hand hygiene rates have improved. Monthly usage of alcohol-based hand rub has increased from 16 litres/1000 bed days to 51 litres/1000 bed days during this same period.
Conclusion This project has delivered sustained improvement in hand hygiene compliance by establishing a framework of multimodal evidence-based strategies.
- Hand hygiene
- hand washing
- accreditation
- audit and feedback
- chart review methodologies
- evaluation methodology
- continuous quality improvement
- antibiotic management
- cluster trials
- MRSA
- nosocomial infections
- infection control
Statistics from Altmetric.com
- Hand hygiene
- hand washing
- accreditation
- audit and feedback
- chart review methodologies
- evaluation methodology
- continuous quality improvement
- antibiotic management
- cluster trials
- MRSA
- nosocomial infections
- infection control
Background
Hospital-acquired infections (HAIs) pose a very real and serious threat to patients who are admitted to or treated at hospitals. Many bacterial and viral pathogens are readily transmitted on healthcare workers' hands, either directly or via an intermediate stage of environmental contamination, and hand hygiene substantially reduces this transmission risk.1 Some pathogens also pose a risk to healthcare workers themselves. Adherence to hand hygiene practices is particularly important in the paediatric setting due to the frequent contact among patients, their environment, parents, visitors, other admitted patients and healthcare workers. This high frequency of contact further raises the risk of pathogen cross transmission.
The international literature suggests that despite a decade of publicity and efforts to improve compliance, hand hygiene practices remain suboptimal in many settings. A systematic review published in 2010 showed a median hand hygiene compliance of 40%, and this was lower among physicians, in intensive care units and prior to patient contact.2
The ‘Clean hands save lives’ campaign, launched in March 2006 by the Clinical Excellence Commission in New South Wales, Australia, was designed to reduce HAIs through improving hand hygiene adherence. The pre-campaign state-wide compliance rate was 47%.1 The Children's Hospital at Westmead (CHW) is a 250-bed tertiary referral paediatric hospital with approximately 3000 staff. Prior to participation in the ‘Clean hands save lives’ campaign, hand hygiene was considered an institutional priority but the focus was on staff education at the time of recruitment. An initial audit of hand hygiene practice at CHW in 2006 showed a compliance rate of only 23%. It was hypothesised that factors contributing to this low performance included unavailability of alcohol-based hand rub (ABHR), lack of systems to capture and feedback hand hygiene compliance data, and a lack of actionable performance data for the hospital management to review. Therefore, CHW decided to participate in the state-wide hand hygiene campaign with the aim of achieving a sustainable improvement in hand hygiene rates and reducing key HAIs.
Method
Multidisciplinary and multifaceted interventions were planned and implemented to improve adherence to recommended hand hygiene practices. The planning occurred based on the guiding principles of minimising the evidence-to-practice gap,3 acknowledgement of local organisational culture and cognisance of working conditions in the paediatric healthcare environment. Hospital quality improvement ethics approval was obtained prior to commencement.
Governance
A multidisciplinary project team was established to oversee the project which had high level executive sponsorship. Hand hygiene became a regular agenda item on the hospital's peak quality and safety committee, the Health Care Quality Committee, which is chaired by the Chief Executive.
Alcohol-based hand rub
The WHO guidelines for hand hygiene reflect evidence that accessibility to alcohol-based hand rub near patient locations is key to assisting staff in decontaminating their hands.4
Different hand-cleaning products were trialled at the hospital and extensive staff feedback was sought which informed the final product selection. Being a paediatric hospital, initial placement of ABHR was problematic as staff were very concerned about children being able to access the ABHR, but widespread discussion led to a consensus on placement. Brackets were installed at a certain height (1.3–1.5 m) close to the point of care to avoid children accessing them and also at a distance from electrical equipment or cables due to product flammability. The final location of the brackets was decided in consultation with the staff of each area, the vendor, and the maintenance and engineering department.
Education
Education sessions using fluorescent hand dye revealed by ultraviolet light (‘Bugsy’, Medtex Medical Division, Notting Hill, Victoria, Australia) were conducted for all staff to enhance their understanding of what constituted good hand hygiene technique. The education sessions also focused on the type of patient-care activities that can result in hand contamination, and the advantages and disadvantages of the various methods used to clean hands.
A number of promotional media were produced, from posters to a display in the front foyer. A brochure called ‘Stopping the spread of germs in hospital—information for parents and carers’ was developed and made available in all clinical and waiting areas. The brochure was designed with input from infection control experts and patients and their families.
Presentations
Regular presentations on the initiatives implemented and observed compliance rates were made to senior management, senior and junior medical staff, nursing staff, allied health, support service staff and consumer groups. This helped keep staff engaged with the project and involved them in the plans for future interventions and sustaining the improvements made.
Audits
Every month about 1000 observations are conducted to monitor staff adherence to recommended hand hygiene practices. A standardised overt observation tool was initially developed by the Clinical Excellence Commission. This overt observation tool allowed a small number of trained auditors to record, over a 20 min period per audit, whether healthcare workers who touched patients or their environment had decontaminated their hands before and after patient care. The tool differentiated between high, medium or low-risk activities that can cause infection transfer, and the clinical discipline (craft group) of staff being observed was also recorded. These audits were completed in high acuity wards every month and on a rotational basis in other inpatient areas. Hand hygiene Australia has since adopted the WHO's ‘5 moments for hand hygiene’ audit tool for use in Australia. Since March 2010, this standardised tool has been used by CHW to monitor compliance to hand hygiene. The information is entered into a database to generate reports which can specify compliance by ward, craft group, and hand hygiene ‘moment’.
In 2009, a joint audit was completed at CHW, comparing the assessments of an independent external auditor from the Clinical Excellence Commission using the WHO's ‘5 moments for hand hygiene’ audit tool with those of a CHW auditor using the older ‘Clean hands save lives’ campaign audit tool. The results were remarkably consistent, with only a 1% difference noted in the overall compliance rate. Although the primary objective of this joint audit was to assess the compatibility of the two audit tools, a high compliance rate noted by an external auditor using the new tool enhances confidence in inter-rater reliability, and the credibility of our internal audit results.
Information sharing and acknowledgement
The audit results are regularly shared with hospital senior management and clinical leadership. Comprehensive information on hand hygiene compliance rates is also forwarded to all clinical areas on a monthly basis allowing them to compare their progress against other areas, and their own results over time. The best performing areas are rewarded with a thank you, chocolates, a certificate and a visit from a member of the executive team. This sharing of information and acknowledgement by senior management has been a major factor in improving hand hygiene practices at CHW and sustaining the initial improvement.
Analysis
Hand hygiene rates over time were presented graphically. If appropriate (for instance, when both pre-intervention and post-intervention rates were stable), statistical comparisons between pre-intervention and post-intervention compliance rates were performed using Pearson's χ2 test with Minitab 15 software.
Results
Since the project was implemented hand hygiene compliance has increased from 23% in August 2006 to 87% (p<0.001) in 2011 (figure 1). Major improvements were noted in February 2007 after installation of ABHR throughout the hospital, and then in November 2007 when the performance feedback system was formalised. Since 2008 the compliance rate has been consistently over 80%. The comprehensive dataset now contains well over 35 000 observations over the last 5 years.
Medical staff compliance has increased from 37% at the beginning of the project to a monthly average of 71% in 2011 (figure 2). Specific strategies targeted at medical staff are detailed in the discussion section. While these strategies to engage medical staff have had a positive impact, opportunities for further improvements still exist. In 2011, a drop in medical staff compliance was noted in comparison to 2009, but this is statistically non-significant (p=0.2). In response, we have enhanced the compliance feedback mechanisms to medical unit heads, and continue to monitor this outcome.
Following the introduction of the ‘5 moments for hand hygiene’ audit tool in March 2010, observations are recorded for each ‘moment’ of hand hygiene. An improvement for each of the moments has been noted between March 2010 and June 2011. Overall compliance for moment 1 (before touching a patient) improved from 80% to 86%, for moment 2 (before a procedure) from 86% to 98%, for moment 3 (after a procedure) from 91% to 96%, for moment 4 (after touching a patient) from 84% to 89%, and for moment 5 (after touching a patient's surrounding) from 57% to 68%.
The use of ABHR also provides a good indication of compliance with hand hygiene. Currently over 400 litres of ABHR is used each month, in comparison to 100 litres/month in 2007 and 250 litres/month in 2008. This equates to an increase from 16 litres/1000 bed days to 51 litres/1000 bed days (figure 3).
In consultation with infectious diseases and microbiology, an outcome measure has been developed to assess the overall effectiveness of the project interventions in terms of the ultimate goal of infection prevention. A set of HAIs for which hand contamination was felt to be particularly important was prospectively developed and has been closely monitored. This set includes rotavirus, multiresistant organism (MRO) acquisition, and nosocomial bacteraemia. A decline in these HAIs has been noted as hand hygiene rates have improved (figure 4). Due to small numbers of monthly infections the statistical significance of this reduction could not be established, however the decline is obvious, and improvements in this highly clinically significant endpoint provide important encouragement to project participants, as well as persuading senior clinicians of the importance of the project.
Discussion
The ‘Clean hands save lives’ campaign, developed in response to community and health system concerns about HAIs in health facilities, produced substantial improvements. However, it was not until several additional strategies were added to better suit our hospital's local environment that consistently high hand hygiene compliance was achieved. Our approach included establishing project governance, staff education and training, regular audits and feedback, visual reminders, use of a multidisciplinary team, patient and family involvement, and a reward and recognition system. We achieved success with a multimodal approach to hand hygiene promotion that is consistent with the published literature.5
Managing the project as a clinical governance activity and emphasising the risk management perspective enhanced the engagement of senior clinical and administrative leadership who are ultimately accountable for driving the quality improvement agenda and providing the necessary resources. Communication of a clear vision and project aim by the chief executive was crucial to the project. The provision of regular, meaningful, accurate and actionable data fulfilled the needs of clinical and administrative leadership and assisted in echoing the project aim, as well as the support of both governance and content experts, down the hierarchy. Process and outcome measures were communicated regularly to demonstrate short-term wins and longer-term progress. As well as reporting to the hospital leadership, contemporaneous feedback to each ward area was provided in a graphical format, with comparator data from other wards. This was linked to the incentives, rewards and recognition mechanism which also demonstrated the commitment of hospital leadership to the hand hygiene programme and its priority on the organisational agenda. The identification of comparator wards in the performance feedback graphs created an immediate sense of competitiveness, since no area wanted to be at the bottom of the performance list. Systematic literature reviews have shown that audits followed by comparative, contemporaneous feedback on performance are generally effective for stimulating improvement.6 7 The WHO guidelines also consider that reporting results of hand hygiene observations to healthcare workers is an essential element of multimodal strategies to improve hand hygiene practices.4
Much of the healthcare literature shows lower hand hygiene adherence among physicians than other craft-groups.8 Cantrell et al reported that physician compliance with hand hygiene guidelines has often been suboptimal.9 We applied a different set of strategies for medical staff than other healthcare workers, using content experts such as a clinical microbiologist to present peer-reviewed evidence, paired with the visual demonstration of bacterial growth from swabs of hands and medical equipment. We believe that this tailored approach contributed to the improvements in medical staff compliance.
The project also focused on easy availability of the ABHR that staff preferred. ABHR rollout was clearly associated with a large early increase in the hand hygiene compliance rates. Bischoff et al10 also found that introduction of easily accessible dispensers with an alcohol-based waterless hand rub led to significantly higher hand washing rates among healthcare workers.
Collecting reliable observation data requires a highly structured method of observing care and documenting data.11 Studies have also reported very different hand hygiene rates for the same units, depending on the role and training of the observer.12
The overt observation method (as used by healthcare workers) of monitoring compliance with hand hygiene has the inherent limitation that people may improve their behaviour while being observed—the ‘Hawthorne effect’. Sax et al13 suggest that even if this bias cannot be avoided entirely, desensitising healthcare workers through repeated exposure and unobtrusive conduct of observers can attenuate it. Another limitation of our data is that while HAI transmission in a paediatric hospital may be due to patients, family members or other visitors as well as healthcare workers, only the hand hygiene practices of staff were directly monitored. Product usage is considered an indirect approach to assessing adherence to hand hygiene guidelines.12 The volume of ABHR ordered each month by the hospital was adopted as a secondary measure of hand hygiene activity, and a steady increase in the volume of ABHR was noted (figure 3). Concordance between this additional measurement method and the overt observation data provides confidence that the observed compliance rate is a good reflection of actual practice. Furthermore product usage (unlike observed compliance) also reflects use by patients and visitors as well as by healthcare workers.
One weakness of the study is that since multiple interventions were implemented in quick succession (figure 1), it is difficult to attribute success to individual interventions. This is analogous to a ‘bundle of care’ approach, in which multiple evidence-based interventions are implemented together for pragmatic reasons and hoping to achieve maximal results, yet the individual contributions of each intervention cannot be inferred.
The most clinically compelling outcome of a quality improvement project to improve hand hygiene practices would be a reduction in HAIs. However, HAIs are multifactorial, reflecting not only breaches in hand hygiene but case mix, community ‘importation’ rates, seasonality, environmental design and cleaning, device use and care, and (for MROs) antibiotic selection pressure, Despite these known limitations, it was important to be able to show improvements in a clinically important endpoint (outcome measure) as well as process measures. For this reason, a composite set of selected HAIs was developed, the transmission of which was most likely to be reduced by improvements in hand hygiene practices. Regular dissemination of this outcome indicator proved to be a very powerful tool to link the results of better hand hygiene to the more tangible benefits of HAI reduction and improved patient safety. However, other factors may have contributed to the observed decrease in HAIs: community vaccination has reduced rotavirus infections in the community during the study period; and the transmission of MROs may have been affected by an antimicrobial stewardship programme which commenced in January 2008. Despite these potential confounders, the project continued to report the same set of HAIs to maintain consistency.
Conclusion
As hospitals fight to reduce infection rates and face the challenge of improving hand hygiene practices, the work done in this area by the CHW confirms that a multimodal and multidisciplinary framework is the best approach to sustainably improve hand hygiene compliance rates. The hospital managed to achieve its aim of 80% of hand hygiene compliance rate within 2 years of project commencement. These improvements in hand hygiene have since been maintained for over 3 years. We believe we substantially eliminated the gap between theoretical best practice and the reality ‘at the coalface’ for frontline clinical staff. Establishing a governance framework for the project, regular sharing of meaningful performance data with clinical and administrative leadership as well as individual wards, easy availability of ABHR, targeted information strategies for different professional groups, and acknowledgement of improved performance has helped CHW in keeping all levels of staff engaged to achieve and sustain a high level of hand hygiene.
Acknowledgments
The authors would like to acknowledge the support and guidance of the Clinical Excellence Commission for this project. We also congratulate the staff of the Children's Hospital at Westmead as without their participation, this true hospital-wide multidisciplinary project would not have been possible.
References
Footnotes
This report employed the SQUIRE publication guidelines for reporting healthcare quality improvement research.
Funding All authors are salaried employees of The Children's Hospital at Westmead. A part-time project officer (one day a week) was funded by the Clinical Excellence Commission for an eight-month period in 2006.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.