A systematic review of hand hygiene improvement strategies: a behavioural approach
© Huis et al.; licensee BioMed Central Ltd. 2012
Received: 11 January 2012
Accepted: 23 August 2012
Published: 14 September 2012
Skip to main content
© Huis et al.; licensee BioMed Central Ltd. 2012
Received: 11 January 2012
Accepted: 23 August 2012
Published: 14 September 2012
Many strategies have been designed and evaluated to address the problem of low hand hygiene (HH) compliance. Which of these strategies are most effective and how they work is still unclear. Here we describe frequently used improvement strategies and related determinants of behaviour change that prompt good HH behaviour to provide a better overview of the choice and content of such strategies.
Systematic searches of experimental and quasi-experimental research on HH improvement strategies were conducted in Medline, Embase, CINAHL, and Cochrane databases from January 2000 to November 2009. First, we extracted the study characteristics using the EPOC Data Collection Checklist, including study objectives, setting, study design, target population, outcome measures, description of the intervention, analysis, and results. Second, we used the Taxonomy of Behavioural Change Techniques to identify targeted determinants.
We reviewed 41 studies. The most frequently addressed determinants were knowledge, awareness, action control, and facilitation of behaviour. Fewer studies addressed social influence, attitude, self-efficacy, and intention. Thirteen studies used a controlled design to measure the effects of HH improvement strategies on HH behaviour. The effectiveness of the strategies varied substantially, but most controlled studies showed positive results. The median effect size of these strategies increased from 17.6 (relative difference) addressing one determinant to 49.5 for the studies that addressed five determinants.
By focussing on determinants of behaviour change, we found hidden and valuable components in HH improvement strategies. Addressing only determinants such as knowledge, awareness, action control, and facilitation is not enough to change HH behaviour. Addressing combinations of different determinants showed better results. This indicates that we should be more creative in the application of alternative improvement activities addressing determinants such as social influence, attitude, self-efficacy, or intention.
Hospital-acquired infections (HAIs) burden patients, complicate treatment, prolong hospital stay, increase costs and can be life threatening . Recent studies in Europe have shown that HAIs affect 4.6% to 9.3% of the hospitalised patients [2–8]. In Europe, the estimated five million HAIs that occur annually have an assumed attributable mortality of 50,000 to 135,000 at a cost of €13 to €24 billion . In the United States, prevalence rates were estimated at 4.5% for 99,000 cases of excess mortality and an economic burden of US $6.5 billion in 2004 [10, 11].
Adequate hand hygiene (HH) among hospital personnel could prevent an estimated 15% to 30% of the HAIs [12, 13]. Numerous studies over the last few decades have shown that HH compliance rates are generally less than 50% of all the opportunities [14–16]. Many strategies have been designed and evaluated to address the problem of low compliance, but most of the effects are small to moderate and often short term [12–17]. This stresses the importance of a clear evidence-based strategy to improve HH routines [18, 19].
Explanation of terms
Determinants of behaviour change
The determinants targeted by a systematically developed strategy are those that have been identified for altering behaviours. Theoretically, the application of a chosen behaviour change activity as part of the HH improvement strategy will alter a specific behavioural determinant, which in turn will change behaviours
Behaviour change technique
Behaviour change techniques refer to the specific methods used to promote behaviour change
Activities refer to the operationalisation of behaviour change techniques
Overview of HH compliance rates
Teaching skills/specific instruction
Hand hygiene improvement strategy
A strategy consist of a set of one or more techniques (e.g., education, feedback, goal setting), intended to change specific determinants (e.g., education to increase knowledge, feedback to raise awareness, guided practice to enhance self-efficacy) of HH behaviour
Theoretically, the application of a chosen behaviour change activity as part of the HH improvement strategy (e.g., a meeting to educate staff on the World Health Organization five moments for HH) will alter a specific behavioural determinant (in this case, their knowledge on the five moments for HH), which in turn will change behaviours (in this case, HH behaviour in line with the five moments for HH) . We hypothesise that a HH improvement strategy targeting more different determinants of behaviour change will be more effective in increasing HH compliance than a HH improvement strategy targeting less different determinants of behaviour change.
The purpose of the present study is to offer sufficient conceptual clarity on the nature of HH improvement strategies by classifying their improvement activities on the basis of their determinants of behaviour change. In addition, we used the controlled studies of our review to explore the effectiveness of targeting different determinants of behaviour change.
First, we selected the 21 studies that Naikoba and Hayward reviewed . Second, we searched the databases of MEDLINE, CINAHL, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Database of Systematic Reviews, Database of Abstract of Reviews of Effects (DARE) from January 2000 up to November 2009, as well as the Current Controlled Trials, ClinicalTrials.gov, National Health Service Centre for Reviews and Dissemination (NHS-CRD): National Health Service Economic Evaluation Database (NHS-EED), and National Health Service Centre for Reviews and Dissemination Health Technology Assessment (NHS-CRD: HTA). The search was limited to studies of human beings, but no language restrictions were imposed. The search terms included the methodological filters of the EPOC combined with selected MeSH terms (handwashing) and free text terms (hand washing and hand hygiene) as used by Naikoba and Hayward . The search strategies used are outlined in Additional file 1.
Population: HCWs in hospital settings
Intervention: strategies aimed at improving HH behaviour
Comparison: HH behaviour before the introduction of the programme or strategy, or HH behaviour in a comparison group where another programme or no programme (usual care) was implemented
Outcome: all operationalisations of HH behaviour of HCWs.
Two reviewers (AH and TvA) independently reviewed the titles and abstracts of citations generated by the search to assess their eligibility for further review based on the selection criteria, and chose relevant articles for possible inclusion. Differences in selection were resolved by consensus or consultation with a third reviewer (MH or LS) in cases of doubt. From potentially eligible studies, the full text papers were subjected to the same evaluation strategy.
Methodological quality rating
Design of study or assignment rating
Experimental: RCT, random allocation; CCT, quasi-random allocation; three data collection points before and after the intervention
Quasi-experimental: CBA, comparable control sites
Quasi-experimental: nonequivalent control sites
Single group before-after tests with baseline measurement
Intervention is clearly described
Described and justified. An n per group sufficient to detect a significant effect (p < 0.05) with a power of 0.80 or reported calculation of power
Validity and reliability of instruments
Unobtrusive observations, rater procedure described and r > 0.80
Unobtrusive observations, rater procedure not described or r < 0.80
Obtrusive observations, rater procedure described and r > 0.80
Obtrusive observations, rater procedure not described or r < 0.80
Volume of soap or hand alcohol used
Test statistics are described
p Value or confidence interval is given
Two reviewers (AH and TvA) independently determined whether studies met the criteria set for methodological quality, and disagreements were again resolved by discussion. Studies with less than three out of seven points were removed. Studies that rated three points but failed to have a positive score for ‘instruments used’ were removed. Studies that rated three (with a positive score for ‘instruments used’) to five points were graded as moderate quality, and those with six or seven points were graded as high-quality studies.
Selection* of the most relevant techniques and their determinant with this overview
Behaviour change technique
Description of the activity in studies
Provide general information
Educational sessions or educational materials
Increase memory or understanding of information
Group discussion, answering questions, clarification
Information about risks of non adherence or inadequate hand hygiene (infection rates, costs)
Delayed feedback of behaviour
Overview of recorded hand hygiene behaviour
Direct feedback of behaviour
Using a system to make professionals aware of their hand hygiene behaviour soon after planned execution
Feedback of clinical outcomes
Overview of nosocomial infections
Provide information about peer behaviour
Information about peers’ opinions of correct hand hygiene
Provide opportunities for social comparison
Group sessions with peers in which discussion and social comparison of hand hygiene practices can occur
Mobilise social norm:
Exposing the professional to the social norm of important others (not peers) such as opinion leaders
Positive consequences of proper hand hygiene
Reinforcement of behavioural progress
Praise, encouragement, or material rewards
Use of a role model. Demonstration of proper hand hygiene behaviour in group, class, or team
Messages designed to strengthen control beliefs about the way of performing correct hand hygiene
Teaching skills and providing feedback. Specific instruction for correct hand hygiene behaviour
Plan coping responses
Identification and coping with potential barriers
Set graded tasks, goal setting:
Desired hand hygiene behaviour is achieved with a stepwise model
General intention information
Explanation of the goals and targets concerning hand hygiene
Agree to behavioural contract
Contract or commitment with formulated goals of hand hygiene behaviour
Use of cues
Following behavioural change
Facilitation of behaviour
Provide materials to facilitate behaviour
Supportive materials are provided for the healthcare workers
Continuous professional support
Involves service provided by infection control team or working group, and/or an additional nurse who attends the implementation
All reviewers participated in a four-hour training on identifying and coding behavioural techniques in line with the taxonomy. A coding manual (Additional file 2) guided the training. This manual contained comprehensive and detailed criteria for assessing the behaviour change techniques and their related determinants. These criteria and any ambiguities were discussed during the training. Then, we performed a pilot using three excluded studies to validate our scoring results. Finally, two pairs of reviewers (AH and TvA or LS and MH) used the taxonomy to independently code the complete range of improvement activities in the included studies into behaviour change techniques. The techniques identified were grouped under their related determinant. The same procedure was also applied to code ‘usual’ or ‘standard’ care provided to control groups The reviewers who coded the strategy were blinded for the study results, and vice versa. Differences in coding (i.e., <5%) were resolved through discussion. See Additional file 3 for a worked example of data extraction and coding.
Given the heterogeneity of the studies with regard to target groups, content and delivery of strategies, and opportunities/moments for HH, no formal meta-analysis was done. We describe frequently used strategies at the level of the nine categories of determinants within the classification of the Taxonomy of Behavioural Change Techniques by reporting the frequency with which the determinants were addressed across all studies included in this review.
We analysed the effectiveness at the level of the nine categories of determinants and compared studies addressing one or more determinants. To obtain methodological soundness, we only make inferences about effectiveness using data of the controlled studies (i.e., randomised controlled trials, controlled before-and-after studies, and studies with a cross-over design).
The overall effect size was determined by calculating the relative difference between the intervention and control groups in each controlled study. This relative difference represents the ratio of difference (in percentages) between the interventional and control groups. We obtained the value by dividing the difference between the post-intervention performance scores from the interventional and control groups by the post-performance test scores of the control group, multiplied by 100 (see Additional file 4). To combine findings across studies, we computed the median effect size and the range, representing the results of strategies related to determinants. We decided to report the median because it is less sensitive to extreme scores and provides a better estimate of what the ‘average’ is. Most of the studies included in this review evaluated short-term effects, so we only report results derived from measurements made directly after the interventions were completed.
Additional file 6 provides an overview of study characteristics in the 41 studies reviewed. Naikoba and Hayward had previously reviewed 11 studies that were published from 1986 through 1999, and the remaining 30 studies were published from 2000 through 2009. Twenty-eight studies had a before-after test design, seven had a controlled before-after design, three were randomised controlled trials, and three had a cross-over design. The study settings were predominantly intensive care units (n = 25), followed by medical or surgical wards (n = 10), emergency wards (n = 4), and 2 studies covered all hospital wards. Multicentre trials were conducted in three studies (two to four hospitals) and the number of participating wards varied from one to three per hospital. In 28 studies, the target population was specified as nurses, physicians, and other HCWs. Six studies targeted only nurses, while seven studies did not specify the type of HCW. The unit of analysis was defined as HH opportunities or moments for HH (n = 33), participants (n = 5), patients (n = 1) and number of dispenser activations (n = 2). Most studies (n = 39) reported HH compliance rates as a primary outcome measure. These data were collected by means of unobtrusive observations (n = 30) or by obtrusive observations (n = 9) in HCWs. One study measured HH performance by volume of soap and hand alcohol used, and one study identified HH episodes by using an electronic counting device. Six studies based their strategy on barriers identified by practice research such as skin irritation, workload, staff personal habits, and priorities. Eleven studies mentioned barriers derived from the literature. The rating of study quality resulted in six high-quality studies. Each of these studies scored six points on our rating scale. Two of the moderate-quality studies scored three points, 28 studies scored four points, and five studies scored five points. Identified quality limitations were: uncontrolled study design (n = 28), absence of sample size justification (n = 33), observations without a description of inter-rater reliability agreement (n = 31), and no description of test statistics (n = 3).
Content of strategies related to determinants of behaviour change
Studies n = 41
Determinants of behaviour change [studies]
Studies addressing one determinant (3 controlled and 6 uncontrolled studies)
Studies addressing two determinants (1 controlled and 4 uncontrolled studies)
Knowledge, Facilities 
Awareness, Action control 
Awareness, Social influence *
Studies addressing three determinants (3 controlled and 5 uncontrolled studies)
Knowledge, Awareness, Facilities *
Knowledge, Awareness, Attitude 
Knowledge, Awareness, Self-efficacy *
Knowledge, Action control, Intention 
Studies addressing four determinants (2 controlled and 4 uncontrolled studies)
Knowledge, Awareness, Facilities, Social influence *
Knowledge, Self-efficacy, Action control, Awareness 
Knowledge, Self-efficacy, Action control, Facilities 
Self-efficacy, Intention, Awareness, Social influence *
Studies addressing five determinants (3 controlled and 6 uncontrolled studies)
Knowledge, Awareness, Facilities, Attitude, Self-efficacy *
Knowledge, Awareness, Facilities, Self-efficacy, Action control *
Knowledge, Facilities, Self-efficacy, Action control, Attitude 
Studies addressing six determinants (1 uncontrolled study)
Knowledge, Awareness, Social influence, Attitude, Action control, Facilities 
Studies addressing seven determinants (1 controlled and 2 uncontrolled studies)
Knowledge, Awareness, Social influence, Self-efficacy, Intention, Action control, Attitude *
Knowledge, Awareness, Social influence, Self-efficacy, Intention, Action control, Facilities 
Knowledge, Awareness, Social influence, Self-efficacy, Action control, Attitude, Facilities 
Fourteen studies addressed three or four determinants and used combinations as described above, but seven studies also addressed determinants as social influence, attitude, self-efficacy, or intention. For example, Huang focussed on increasing knowledge (educational training programme and written information) and awareness (clarifying risks for blood pathogen exposure), but also enhanced the self-efficacy of nurses with one hour of practical demonstration of hand washing and using gloves . In Marra’s study, activities were also aimed at increasing awareness by providing feedback on infection rates. The nurse manager also provided opportunities for social comparison by showing each HCW the total number of times the dispensers were used in each patient room in which the HCW worked compared to the number of times that other HCWs used dispensers. In addition, the nurse manager explained the goals and targets of the HH improvement strategy twice a week, thus strengthening intention and self-efficacy .
All 13 studies addressing five or more determinants consisted of activities addressing multiple different determinants. For example, Trick et al. addressed determinants such as knowledge (educational sessions and distribution of educational materials to professionals), awareness (displaying HH adherence), action control (hospital-wide poster campaign), facilities (alcohol-based hand rub), and attitude (pointing out the benefits of using alcohol-based hand rubs) .
We found no differences in the extent to which determinants were targeted between the controlled studies and uncontrolled studies (Table 4).
See Additional file 7 for details of improvement activities and results in the 41 studies reviewed.
Effectiveness of controlled studies related to determinants of behaviour change
Determinants of behaviour change [studies]
All studies (n = 13)
R = relative difference between intervention and control$
M = median [range]
25.7 [-8.8 to 429]
Studies addressing one determinant
n = 3
M: 17.6 [-8.8 to 61]
Action control 
n = 1
n = 1
n = 1
Studies addressing two determinants
n = 1
M: 25.7 [25.7*]
Awareness, Social influence 
n = 1
Studies addressing three determinants
n = 3
M: 42.3 [19.5 to 82.7]
Knowledge, Awareness, Facilities 
n = 1
Knowledge, Awareness, Self-efficacy 
n = 1
R : 42.3
Knowledge, Action control, Facilities 
n = 1
Studies addressing four determinants
n = 2
M: 43.9 [14.8 to 73*]
Knowledge, Awareness, Facilities, Social influence 
n = 1
Self-efficacy, Intention, Awareness, Social influence 
n = 1
Studies addressing five determinants
n = 3
M: 49.5 [-8.6 to 429]
Knowledge, Awareness, Action control, Facilities, Attitude 
n = 1
Knowledge, Awareness, Facilities, Attitude, Self-efficacy 
n = 1
Knowledge, Awareness, Facilities, Self-efficacy, Action control 
n = 1
Studies addressing seven determinants
n = 1
M: 9.7 [9.7 *]
Knowledge, Awareness, Social influence, Self-efficacy, Intention, Action control, Attitude 
n = 1
No controlled study addressed six determinants. The only controlled study addressing seven determinants showed less impact on short-term effectiveness (relative difference 9.7). However, baseline HH rates in this study were higher in the intervention group than in the control group, probably because administrators were already planning and discussing the strategy during the baseline phase .
Improved HH behaviours among hospital personnel could have a considerable impact on HAIs, healthcare costs, and patients’ health and quality of life. Yet, reviews with detailed examination of the active content of strategies to promote HH are missing. In the present study, the content and effectiveness of a range of strategies to improve the HH adherence of HCWs were studied. By using a detailed coding taxonomy of behaviour change techniques targeting major behavioural determinants, we were able to obtain a detailed insight into frequently used HH improvement strategies and how they work. Analysing the content of the strategies at the level of determinants that prompt HH behaviour, it was found that those studies focusing on combinations of different determinants gave better results, which indicates that we should be more creative in the application of alternative improvement activities aimed at altering specific behavioural determinants change, such as social influence, attitude, self-efficacy, and intention.
Although the content of the strategies and related determinants varied greatly, most of the studies addressed more than one determinant (mainly knowledge, awareness, action control, and facilitation of behaviour). This is consistent with Naikoba and Hayward’s findings and previous systematic reviews of changing professional behaviour in which education (addressing ‘knowledge’), feedback (addressing ‘awareness’), reminders (addressing ‘action control’), and facilities (addressing‘facilitation of behaviour’) were the most frequently used improvement activities [12, 20, 21]. Twenty strategies addressed additional determinants that prompt HH behaviour such as social influence, attitude, self-efficacy, or intention. These specific determinants were especially targeted in comprehensive strategies that addressed at least four determinants. This provides new insight into the content of HH improvement strategies: half of the studies used a strategy targeting determinants not mentioned in previous reviews of HH adherence.
Most studies addressed determinants at the individual and institutional levels; specific team-oriented activities were hardly identified. Strategies including team-directed activities could, however, be valuable because HCWs (especially nurses) usually work in teams. Evidence for the effectiveness of team-directed strategies in other settings exists, but these strategies are rarely applied in studies of HH improvement [72, 73]. Surprisingly, activities directed at behavioural maintenance following behaviour change were not identified in the studies. Nonetheless, activities aimed at persistence should be part of the strategy for achieving sustainability of improved HH behaviour.
The effectiveness of the strategies varied substantially, but most controlled studies showed positive results. This is in line with previous review findings [74, 75]. If determinants such as social influence, attitude, self-efficacy, and intention are targeted within a strategy, the effect is larger than that of strategies consisting solely of a combination of determinants, such as knowledge, awareness, action control, and facilities. Apparently, these specific determinants provide an additional contribution to effectiveness. This finding is confirmed by results of previous studies where social influence, attitude, self-efficacy, and intention are considered relevant to successfully changing behaviour [26–29].
The median effect size increased when more determinants were addressed. In other words, there seems to be a dose response effect. This result deviates from Grimshaw et al.’s finding that there was no dose response relation between the number of improvement activities and the effects of multifaceted strategies . The lack of a rationale in the composition of a multifaceted strategy, such as mentioned by Grimshaw, may be a good explanation for the lack of a relationship between the number of improvement activities and the effect. An additional explanation for this discrepancy can be found in the framework chosen to classify the strategies for change. Grimshaw used the EPOC classification of strategies that is based on the form of performed improvement activities. We used an alternative approach that classed improvement activities on the basis of their determinants of behaviour change. By using the Taxonomy of Behavioural Change Techniques we collected information about triggers that encourage behaviour change rather than describing separate improvement activities. Thus, using multiple activities is not necessarily the same as addressing multiple determinants or vice versa. For example, the combined distribution of educational materials and provision of educational sessions constitute two different improvement activities. We would not label this strategy as multifaceted because both activities apply the same determinant (‘knowledge’).
Although we found a maximum effect in addressing five determinants, we cannot provide a ‘one-size-fits-all’ recipe for building a successful strategy. Previous recommendations from the literature have pointed out that an improvement strategy for HH behaviour should address existing problems and barriers [12, 73, 75]. Analyses of barriers and facilitators and linking improvement activities to these influencing factors are important steps in the design of a strategy and may be crucial to success. A multifaceted strategy with many improvement activities that are not precisely tuned to the existing barriers apparently misses the target; part of the components may be redundant or ineffective. For example, if there is no knowledge shortage, educational strategy components probably will not contribute to the effectiveness of the multifaceted strategy. Barriers also exits at other levels than the individual HCW. Barriers like negative role models, a poor social culture, and disinterested management can hamper good HH. Overcoming these barriers requires the use of alternative activities such as social influence, attitude, self-efficacy, or intention. Of particular interest is the HELPING HANDS study, currently performed in the Netherlands . In this study, improvement activities are directed at gaining active commitment and initiative of ward management; modelling by informal leaders at the ward; and setting norms and targets within the team. This team-directed strategy goes beyond individual and institutional only approaches, but rather addresses determinants at team level by focussing on social influence in groups and strengthening leadership.
In this review, it was not possible to check for this ‘appropriateness’ of determinants addressed within the studies because context and barrier analysis and the rationale regarding strategy selection were hardly reported. Therefore, for most of the studies, it was unclear how well the strategy fitted the context. In view of the effectiveness, but also feasibility and costs, we propose selecting appropriate determinants rather than addressing all determinants.
The Taxonomy of Behavioural Change Techniques was a valuable tool that led us to convert descriptions of improvement activities into well-defined determinants. We obtained a clear focus on theory-based determinants of behaviour change that were hidden in the improvement strategies. We consider this a crucial step in developing a theoretical understanding of the effectiveness of improvement strategies.
Although we succeeded in achieving substantial insight into the content and effectiveness of HH improvement strategies, some aspects should be considered further. First, the methodological weakness of the studies is still a major concern. Most of the studies were small scale; they lacked a control group comparable to the test group, and made no formal attempt to minimise bias. There is a risk that a positive relationship between the number of determinants targeted and the effect on HH compliance might be partly explained by an unknown confounder. This holds particularly true for the observational studies where wards were selected to receive an improvement strategy. In our review, we included studies that clearly described the content of the strategy and were at least of moderate quality. With methodological soundness in mind, we only used results from controlled studies when we reported effectiveness. However, the risk of confounding should be taken into account when interpreting our results. Methodologically robust research is still required to evaluate the effectiveness of interventions intended to improve HH compliance. Adequately powered cluster randomised trials or well-designed ITS studies would provide the optimal study design.
Second, our search included literature up to November 2009. Therefore, we cannot provide information on recently performed HH improvement studies The screening and analysis of the search results as reported in this review served as a starting point for the development of two HH improvement strategies, which were subsequently tested in a randomised controlled trial. The design of this study was published in 2011 .
Third, as in any systematic review of the literature, there may be publication bias. Most studies showed positive results; it is possible that studies with negative results have not been published. In our review we were unable to retrieve four articles; it is possible that they contained relevant data.
Fourth, the criteria used to determine when HH should be performed varied over the studies and were not always explicitly stated. This may have implications for the generalisability of the results of the studies.
Fifth, good reliability in coding the improvement activities was observed (>95%), suggesting that our instructions and definitions can be applied reliably after only brief training. Within all steps of the review process, validity was increased by using standardised methods and forms as well as multiple raters. However, once techniques and targeted determinants are well chosen, examining the actual exposure to the improvement activities was problematic. Studies did not or marginally report on how well the improvement strategy was implemented. Designating HH as hospital goal, for example, requires setting specific, realistic, and measurable targets . However, descriptions of the improvement activities in the studies provide insufficient detail to check for appropriate delivery as well as the actual exposure of the HCWs to this activity. Without sufficient information about implementation fidelity, it is hard to determine whether the impact of the HH improvement strategy is due to the implementation process or to the composition of the strategy itself, a so-called Type III error .
Finally, most studies did not describe, or only marginally described, the activities of the ‘usual’ or ‘standard’ care provided to control groups. Standard care practices may vary from site to site. Therefore, describing standard care is important for the interpretation and comparison of intervention effects. Given the combination of strengths and considerations, this review provides an original and valuable overview of various strategies for improving the HH adherence of HCWs.
By focussing on determinants of behaviour change, we found hidden and valuable components in HH improvement strategies. Addressing only determinants such as knowledge, awareness, action control, and facilitation is not enough to change HH behaviour. Addressing combinations of different determinants provided better results. This indicates that we should be more creative in the application of alternative activities addressing determinants such as social influence, attitude, self-efficacy, or intention.
A systematically designed strategy that targets various problems and barriers to change, with activities at different levels (professional, team, and organisation), is needed to achieve changes in HH behaviour. Currently, most strategies focus on the individual and the organisation, while group- or team-directed strategies are rarely used. Including team-directed techniques in a strategy is a promising development.
This study is funded by a research grant from ZonMw, dossier number: 94517101.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.