Implementation of MRSA Infection Prevention and Control Measures – What Works in Practice?

There have been increasing numbers of media reports about careless behaviour by healthcare workers, mainly involving insufficient cleaning practices and the absence of hand hygiene measures (Boyce, 2009). Although adherence to infection prevention and control measures has received a lot of attention in the media and in scientific literature, surprisingly little attention has been given to the implementation of the infection prevention and control strategies in healthcare practices. In the medical literature the focus is on the availability of national or regional MRSA surveillance data and guidelines for prevention and control. To date hardly any data has been made available about the kinds of interventions that have been successful in implementing infection prevention and control.

). Such a strategy should include interventions aimed at different levels: the management of healthcare institutions, the behaviour of healthcare workers and the quality of the infection control guidelines. However, what empirical evidence exists for a multi-faceted implementation strategy? And how successful are these strategies? To investigate this, we conducted a systematic literature review. This review will be used to develop an implementation strategy that fits the habits and culture of hospital-based healthcare workers (HCWs) in hospital care settings. In this review, we searched for empirical studies to investigate and identify effective implementation strategies for improving adherence to MRSA prevention and control measures. The following questions guided our review of the literature: - What implementation strategies are used? -What is the foundation of these strategies (theories, experience, etc.)? -What research designs were used to measure the effects of the implementation strategies? -What effects are reported? -On adherence to the measures? -On the reduction of costs? -On the reduction of MRSA?

Method of the systematic review
The York protocol for systematic reviews (Centre for Reviews and Dissemination, 2001) was used to guide the review process. Literature searches were carried out in the online databases Scopus, ISI Web of Knowledge and the Cochrane Library. In addition, we handsearched the indexes of the Journal of Hospital Infection (JHI), the American Journal of Infection Control (AJIC) and Clinical Microbiology and Infection (CMI) for relevant publications. We searched for studies describing the implementation of MRSA prevention or control measures. The publications were included in the review if they met the inclusion criteria listed in Table 1. Most important was that the publications described an implementation strategy and implementation outcomes. Two independent reviewers (NdJ, JW) applied the inclusion criteria to the publications in a title screening round, followed by an abstract and a full-text screening round. After each round, the reviewers compared their judgments and resolved discrepancies through discussion. The included studies are summarized in a data table, and the study features and results are summarized and compared. Due to the heterogeneity of the data and the limited number of included studies, no meta-analysis was performed.

Article screening
The search strategy resulted in 661 potentially relevant publications (after duplicates were removed). The screening process and outcomes are shown in Figure 1; 29 publications were included in the review. The characteristics of these publications are summarized in Table 2.
The characteristics and outcomes of the included studies are discussed in the following sections. The numbers we cite correspond to the publications summarized in Table 2. www.intechopen.com

Implementation of MRSA Infection Prevention and Control Measures -What Works in Practice?
95

Inclusion Criteria
Publication Type (Scientific) Journal article, published between 2005-2010

Scope of Studies
Implementation of an evidence-based MRSA prevention or control measure. The implementation strategy must be described.

Study Settings
Primary-/secondary-care facilities, long-term care facilities, nursing homes

Outcome measure
Implementation outcomes (mostly behavioural) must be given.  Behavioural (e.g. adherence to implemented measure, knowledge)  Clinical (e.g. prevalence rates, infection rates, deaths)  Organizational (e.g. changes in Length of Stay (LoS), expenditures, costs)

Behavioural
Significant reduction in cephalosporins use. Significant increase in ciprofoxacin use. Clinical Significant reduction in MRSA due to intervention. An increase in susceptibility to piperacillin/ tazobactam obse in P. aeruginosa isolates ceased after the change in antibiotic Increase in susceptibility to ciprofloxacin in K. pneumoniae i although an abrupt change in the percentage of susceptible is intervention. Decrease in susceptibility to piperacillin/tazobactam and im an increase in susceptibility to ciprofloxacin in P. aerugionos surveillance period, with no significant changes due to interv 3: Burkitt, et al., 2010, United States Educational meetings Reminders Mass media Technology-supported Implementation foundation Theoretical: lack of measurement of actual changes in knowledge due to education Infection control measure Hand hygiene Personal protective equipment Patient screening Patient isolation Design Before and after design Behavioural Questionnaires Significant increase in proportion of respondents who report than soap and water to clean their hands. Significant increase in mean number of knowledge questions Significant increase in proportion of respondents who agreed problem in their unit. Significant increase in proportion of respondents who report staff about proper hand hygiene and contact precautions. Significant decrease in job satisfaction. Significant increase in reported using prevention practices. Significant increase in proportion of respondents reporting a hygiene, primarily because they feared that hand rubs or soa forgot to perform hand hygiene.

Observations in person
Overall hand hygiene compliance improved at 4 months, and months. In individual sentinel areas, compliance rates improved sign months post-intervention in all areas. Use of ABHRS products increased in all sentinel areas. Clinical MRSA colonization rates did not change in any of the sentine worker MRSA colonization did not decrease in sentinel areas Environmental contamination did not change significantly d Significant decline in total clinical MRSA infections per 100 p For patient episodes of MRSA bacteremia, the monthly rate d period was static, but fell significantly in the post-interventio monthly rate of MRSA bacteremia had decreased. Total clinical isolates per month of ESBLs increased during th had fallen by more than 90% by the 36th month of OCS. Inclusion of axilla and groin sites did not affect the MRSA dete discharge, both overall and when the ICUs were analysed indi Between groups: the rate of MRSA colonization detected by A discharge was higher in SICU than MICU. No significant difference in MRSA infection rate pre-and post when analysed individually. Less variability in MRSA rates post-intervention; the 95% CI at that at pre-intervention. Septic shock at ICU admission was more common in MRSA-co patients.

Financial
Detection of MRSA at any point was associated with longer pr antibiotic therapy, and longer ICU length of stay.

Behavioural
Within groups: medical and nursing staff reported that they 87% of cases in the control period, and in 96% of cases in th Medical and nursing records were flagged significantly mo the control period. The set of four organizational measures was implemented m than in the control period. The same observation was made separately. When considering only ICUs and rehabilitation units, i.e. w significant increase in the implementation of isolation preca often, use of gowns increased, use of dedicated materials in increased, and proportion of MRSA patients in private room There was no significant increase in the proportion of health status of patients or in the proportion of flagged records. 29: Thomas, et al., 2005

MRSA prevention and control measures
Different measures were implemented to prevent or control MRSA. In some studies a single MRSA prevention or control measure was implemented, in others a bundle of measures was implemented. Hand hygiene was implemented as a stand-alone measure in seven studies (4,5,7,10,13,20,29) and as part of a bundle of measures in eleven studies (1,3,6,8,11,14,16,17,19,21,24). Environmental hygiene was implemented as a standalone measure in one study (12) and as part of a bundle of measures in two studies (1, 17). The use of personal protective equipment such as gloves or gowns was implemented as part of a bundle of measures in four studies (1,3,16,21); it was implemented as a stand-alone measure in none of the studies. Medication, or the correct use of antibiotics, was implemented as a stand-alone measure in six studies (2,9,15,22,23,25); in none of the included studies was it part of a bundle of measures. In two studies (26, 27), patient screening was implemented as a stand-alone measure, and in six studies (3,8,11,16,17,19) it formed part of a bundle of measures that was implemented. HCW screening was implemented only as part of a bundle of measures, in one study (17). Patient isolation was implemented as stand-alone measure in one study (28), and was part of a bundle of measures in five studies (3,6,16,18,19).

Implementation strategies and their foundation
Various strategies were used to implement the MRSA prevention and control measures. Most implementation strategies are set up because of the empirical observation of nonadherence to clinical guidelines, thus creating an impediment to successful MRSA control. The theoretical foundation of the chosen strategies is often unclear, or not specified.

-
Clinical multidisciplinary teams were used in five studies (2,6,20,22,25) to guide the implementation of a MRSA control measure. Via cooperation or consultation these teams supported the measures taken, for example by approving antibiotic prescriptions. - Educational outreach was carried out in five studies (6,10,15,25,27) to teach HCWs onsite and sometimes on demand how to apply the implemented measure. -Rewards for correctly performing the implemented measures were given in two studies (4, 16), either to individuals directly after observing correct behaviour, or to groups based on periodic adherence results. -A patient-mediated intervention was implemented in one study (10); patients and visitors were actively addressed to perform the desired hand hygiene behaviour and motivate adherence among staff. -AB permission/formulary was applied in one study (2) where permission to use a certain antibiotic was required.

Outcomes
We classified the reported effects into three categories: adherence to the measures, reduction of costs and reduction of MRSA.
Acquiring a hospital-associated infection (HAI) results in a longer length of stay for the patient and poses many additional costs. Therefore, reductions in length of stay are an important outcome associated with decreased MRSA infection rates. Cost savings, or at least cost-neutral intervention effects, were observed in four studies (5,10,15,27). On the other hand, increased isolation and increased expenditure also posed costs, as described in one study (18). However, in this study, these increased costs were not compared to possible savings due to prevented infections. In another study (19), improved screening led to increased lengths of stay (pre-ICU and ICU), because MRSA detection increased.

Conclusion and discussion
The results of our review show that in most cases hygiene experts or an infection control team (nurse, infectologist, microbiologist) are the developers of implementation strategies. These strategies are driven by empirical observations and audits. The theoretical foundation of the chosen strategies is often unclear. No references to theories and models of human behaviour are made. However, some articles indicated that a literature search was carried out.
When looking at the implementation strategies, we can conclude that in most cases a multifaceted strategy was carried out. This strategy entails a combination of several activities: -Education or training modules for HCWs, sometimes mandatory, taking various forms (DVDs, PowerPoint presentations, posters, meetings, brochures) to improve hand hygiene and compliance with protocols. -Inspections of the adherence to the safety programme and of hand washing behaviour via audits, on-site instructions, and observations by hygiene experts or trained auditors. Results were communicated to management and demonstrated via feedback meetings. -Environmental interventions (red lines at the entrance to high-risk wards, talking walls) to remind HCWs to behave safely in that particular area and to provide antibiotic policy support via guidelines and cards.
The implementation pathway consists of education-inspection-feedback rounds; unfortunately it is unclear who is responsible for the management of the intervention strategies and who invests in these activities. No business model seems to underpin the entire implementation strategy.
To answer the research question about the effect of the implementation strategies, we reviewed the research designs that were used to measure their effects. In general, quasiexperimental designs (before and after and time series designs) underpin the research activities. Implementation outcomes are usually measured in a before-and-after design, where they do not concern antibiotic use, and therefore provide little insight into temporal changes in implementation results or adherence. HCWs are the main target group in the research designs. It is unclear who these designs seek to manage (researchers, HCWs, management) in their execution or whether a project manager is responsible for this. Trained nurses or infection control teams are sometimes used. In most cases quantitative instruments are used to measure the effects on knowledge and behaviour (questionnaires, self-reporting of behaviour, material use, and hand hygiene) and on a reduction in MRSA and antibiotic doses (lab statistics). The effects on cost/benefits were sometimes measured, addressing utilizations such as reduced length of stay. In general the outcomes are promising. However, the extent to which the outcomes are related to the implementation strategies is not clear, except for the routine screenings and reduced MRSA rates. The outcomes on cost-savings are especially hard to analyse. It remains unclear what is measured, how it is measured and to what purpose. Long-term effects are almost never addressed.
Due to several shortcomings in research designs, the overall impact of the implementation strategies could not be measured sufficiently. Shortcomings in the research designs include, for example, the one-sided focus on HCWs. We know from prior research (Verhoeven et al., 2009) and from behaviour change models that not only is a multifaceted strategy needed to change safety behaviour, but that a multi-perspective stakeholder view (HCWs, infection experts, patients, the safety policy of the management of the organization) is necessary to obtain insight into the cost/benefits of the implementation strategy and to discuss the long-term implications of the strategy for the organization and workflow (Kukafka et al., 2003). This requires a theory or innovation-driven approach that grounds the implementation strategy, enabling an assessment of which activities are successful for whom (patient, HCWs, management) and what the interaction effects of the different components of the strategy are.
Another shortcoming concerns the chosen study designs. Authors of the included studies refer to the difficulties in matching control and intervention groups, the high rates of dropouts and the low volume of included respondents, and confounding factors that cannot be excluded. These shortcomings are well-known impediments related to RCTs and the selfreported behaviours. In fact, these shortcomings cannot be avoided due to the study of realtime behaviours and contextual factors that influence these behaviours. Therefore, these factors should not be regarded as nuisances, as the authors do; they are the key issues that are important in implementation studies aimed at changing culture and behaviour. For example, some authors reported problems in implementing the activities due to a lack of resources (a result of the economic downturn) to manage the implementation and problems with measuring the effects of each component of the implementation strategy due to financial constraints. A lack of transparent funding models and lack of management support made the participation of different institutes or wards in the research projects problematic, resulting in only small pilot projects being carried out. These financial barriers should not be reported as shortcomings; rather, these factors should be determined by the key stakeholders and considered as critical factors for changing behaviour and the culture of safety in hospitals or other institutions.
In addition, some authors reported a lack of commitment on the part of nursing personnel to participate in the implementation projects. It appeared that some personnel were uncertain about the implications of several measures. For example, they were concerned that patients would not feel as clean after being washed with wipes instead of soap and water. The level of commitment of HCWs and management is one of the main conditions for success in programmes for innovation or change. The impediments indicate that the implementation strategies are expert-driven rather than stakeholder-centred. Changing safety behaviour in hospitals is first and foremost a cultural problem of management and staff, which requires that implementation strategies should address that level.
How to improve the implementation strategies? Given the fact that the implementation strategies influenced the attitude and knowledge of HCWs in a positive way, that intentions to behave safely increased, and that MRSA rates decreased in several studies, the question is how to boost the impact of the implementation strategies. Education-inspection-feedback rounds could be one way to do this.
Based on prior experience in infection management control and on information gathered from other studies of innovation management (Cain & Mittman, 2002;Rogers, 2003), we argue that the participation of staff and management is crucial to the development and implementation of interventions, to increase applicability, accountability and ownership and to create a fit between the proposed activities and the culture of the organization (Van Gemert et al., in press). In addition, both positive and negative incentives are needed to encourage staff to do the right things at the right times. Change agents and demonstration of best practices will improve the incorporation of safety behaviour. To enhance the transparency of the implementation programme and strategies, communication of results or key factors for success should be available to staff. Communication should include insights into results related to infection management (prevalence and incidence rates of MRSA, identification of increasing/decreasing trends), the business model underpinning the programme (resources, investments, additional costs) and benchmarking (how are we doing and what are others doing?). It is also important to demonstrate to the management and staff that the investment costs of the intervention can be less than the costs of not adopting an MRSA-infection control programme.
Another point of attention is the use of media to implement the strategies. Even though evidence of the usefulness and effectiveness of computerized decision support or reminders exists (Grimshaw et al., 2004), it is not often used. We found that in ten studies DVDs, PowerPoint presentations, educational programmes available online or on CD-ROM, and electronic alerts or reminders were used. This is rather remarkable in our Internet-driven world. Web-based communications systems in particular can increase staff knowledge and provide access to accurate, adequate and easy to understand information (Kreps & Neuhauser, 2010). In prior and on-going research projects aimed at cross-border infection control (MRSA-net; EurSafety Health-net) we developed stakeholder-driven, web-based communication systems, based on national infection control standards, to support staff and patient behaviours (see for example Verhoeven et al., 2009). This resulted in fewer errors, time savings and also appropriate behaviour by HCWs.