ANNOTATED BIB WRTG 391

ww.sciencedirect.com

Journal of Hospital Infection 119 (2022) 33e48

Available online at w

Journal of Hospital Infection

journal homepage: www.elsevier.com/locate/jhin

Review

Hand hygiene compliance in the prevention of hospital- acquired infections: a systematic review

V. Mouajou a, K. Adams b,c, G. DeLisle c, C. Quach a,c,d,* a Department of Microbiology, Infectious Disease and Immunology, University of Montreal, Montreal, QC, Canada b Department of Epidemiology and Biostatistics, McGill University, Montreal, QC, Canada c Research Centre, CHU Sainte-Justine, Montreal, QC, Canada d Infection Prevention and Control, CHU Sainte-Justine, Montreal, QC, Canada

A R T I C L E I N F O

Article history: Received 28 May 2021 Accepted 22 September 2021 Available online 25 September 2021

Keywords: Healthcare-associated infections Hand hygiene Hand hygiene compliance Infectious diseases Infection prevention and control

* Corresponding author. Address: Departme Universitaire Sainte-Justine, 3175 ch. Côte-Sa 2358.

E-mail address: c.quach@umontreal.ca (C

https://doi.org/10.1016/j.jhin.2021.09.016 0195-6701/ª 2021 The Healthcare Infection S

S U M M A R Y

Background: The hands of healthcare workers (HCWs) are known to be a primary source of transmission of hospital-acquired infections (HAIs). Thus, both practising hand hygiene (HH) and adhering to HH guidelines are expected to decrease the risk of transmission. However, there is no consensus on the optimal hand hygiene compliance (HHC) rate for HCWs. Aim: To systematically review the published literature to determine an optimal threshold for the HCW HHC rate associated with the lowest HAI incidence rate. Methods: This systematic review was performed in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines. Online databases were searched using comprehensive search criteria for randomized controlled trials and non-randomized controlled studies, investigating the impact of the HCW HHC rate on HAI incidence rates in patients of all ages within healthcare facilities in high-income countries. Findings: Of the 8093 article titles and abstracts screened, 35 articles were included in the review. Most studies reported overall HAIs per 1000 patient-days and device- associated HAIs per 1000 device-days. Most studies reported HHC rates between 60% and 70%. Lower HAI incidence rates seemed to be achieved with HHC rates of approx- imately 60%. The studies included in this review were not originally designed to assess the impact of HHC on HAI incidence rates, but risk of bias was assessed in accordance with the predetermined exposure and outcome criteria. Eleven (31%) studies were deemed to have low risk of bias. Conclusions: Although HHC is part of the HCW code of conduct, very high HHC rates are difficult to reach. In observational studies, HHC and HAIs had a negative relationship up to approximately 60% HHC. Due to flaws in the study design, causality could not be inferred; only general trends could be discussed. Given the limitations, there is a need for high- quality evidence to support the implementation of specified targets for HHC rates. ª 2021 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.

nt of Microbiology, Infectious Diseases and Immunology, University of Montreal, Centre Hospitalier inte-Catherine, Montréal, Québec, H3T 1C5, Canada. Tel.: þ1 514 345 4931, 7430; fax: þ1 514 345

. Quach).

ociety. Published by Elsevier Ltd. All rights reserved.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e4834

Introduction

With the recent coronavirus disease 2019 pandemic, hand hygiene (HH) has been publicized and highlighted like never before for both healthcare workers (HCWs) and the general public. The search trend for ‘handwashing’ peaked in March 2020. HH has long been regarded as a cornerstone of infection prevention, marketed as an inexpensive strategy, and one of the most effective, for reducing hospital-acquired infections (HAIs) [1]. HAI, defined as ‘an infection acquired in a health care facility in a patient in whom the infection was not present or incubating at time of admission’ [2], continues to be among the most common adverse events affecting hospi- talized patients, and is among the top five killers in the USA [3]. Over the years, several infection prevention measures have been implemented to curb the acquisition of HAIs, among which is HH. The hands of HCWs are believed to be the primary vector of transmission of bacterial agents responsible for HAIs [4,5], hence the popular belief that excellent HH compliance (HHC) can reduce HAIs dramatically. Unfortu- nately, HHC rates have been historically low in healthcare settings [6, 8]. HAIs are often seen as a direct consequence of low HHC and fault is often put upon the negligence of HCW. However, strong evidence of a direct causal link between HHC alone and HAIs is missing. The persistent underlying message is ‘the higher, the better’, with little clinical evidence to support this claim [4]. It is currently unknown whether a 60% level of HHC reduces HAIs significantly compared with an 80% level of HHC. A modelling study by Beggs et al. suggested that an HHC rate of 50% was optimal to reduce the transmission of meticillin-resistant Staphylococcus aureus (MRSA), but no clinical studies have corroborated this finding [9]. This review aimed to determine the HHC rate associated with the lowest HAI incidence rate in an effort to determine an optimal HHC rate and provide evidence for the implementation of realistic HHC targets.

Methods

Search strategy and study selection

This systematic review was conducted and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. The review protocol was registered with PROSPERO (ID: CRD42019138278). The research question followed a PECO format; ‘P-Population’ was healthcare personnel (HCWs) in a high-income country, ‘E- Exposure’ was any intervention to improve HHC, ‘C-Com- parator’ was the baseline HHC and the corresponding HAI incidence rate, and ‘O-Outcome’ was the post-intervention HHC rate and corresponding HAI incidence rate. The search strategy, undertaken in collaboration with a medical librarian, included a combination of Medical Subject Heading terms and keywords pertaining to HHC and HAIs. English and French articles published on HHC and HAIs prior to July 2019 were identified through the electronic databases PubMed, OVID Medline, EMBASE, ICTRP, PsycINFO, Google Scholar and ClinicalTrials.gov. The initial strategy was created in PubMed and adapted to other databases. Only studies that contained terms related to both HHC and HAIs reported in a temporal manner were included. The reference lists of systematic

reviews were hand searched to identify all relevant papers. Identified articles were uploaded into DistillerSR (Evidence Partners, Ottawa, Canada) to track study selection. The full search strategy is available in Appendix A (see online supple- mentary material). Once retrieved, titles and abstracts were screened independently by two reviewers (VMF and KA). Arti- cles included by at least one reviewer were included in the next stage. A second round of title and abstract screening was performed with stricter inclusion definitions; articles needed to be included by both authors to move to the next stage. Disagreements were resolved by consensus.

Study eligibility

Studies that evaluated HCW HHC, were conducted in a high- income country as defined by the World Bank, and reported HHC rates and HAI incidence rates within the same time frame were included in this review. Grey literature, pre-prints, research reports and modelling studies were excluded. Stud- ies that reported proxy measures for HHC (e.g. hand-rub con- sumption) were excluded. Studies reporting HHC alone or HAI incidence rate alone were excluded. Studies that did not report HHC and HAIs in a temporal manner were also excluded. No restrictions were made regarding study design or time period. A table of the inclusion and exclusion criteria is available in Appendix B (see online supplementary material).

Data extraction

A full-text review was performed by two independent reviewers (VMF and GD). Conflicts were resolved through dis- cussion until consensus was reached. Data extraction was per- formed independently by two reviewers (VMF and KA) and conflicts were resolved by a third reviewer (CQ). The following information was extracted: country, year of study, study design, HHC assessment method, hospital ward, hand hygiene guideline used, HHC rate measured by ward or by moment of hand hygiene, intervention assessed, type of HCW observed, type of HAI, unit of HAI, and effect measure of HAI (incidence rate, incidence risk).

Quality assessment

Risk of bias was assessed independently by two reviewers (VMF and KA) using an online template of the Robins-I tool for observational studies, and the updated ROB2 tool for randomized controlled trials [10,11]. Most studies were not designed to evaluate the impact of HHC rates on HAIs, so quality assessment was evaluated as per the exposure (HHC) and outcome (HAI incidence rates) identified by the present authors, notwithstanding those for which the study was designed. Studies were deemed to be at high/critical risk of bias if they had one or more of the following flaws: presence of co-interventions that differed before and after the inter- vention to increase HHC, thus making it impossible to isolate the impact of HHC alone; HAI measurement that differed before and after the intervention; and HAI assessment not defined, did not use a standardized definition (US Centers for Disease Control and Prevention, National Healthcare Safety Network) for HAI surveillance or did not mention what sur- veillance protocol was used to measure HAIs. Studies were considered at serious risk of bias if HAI incidence rates were pooled across different ward types with no adjustment for

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e48 35

patient characteristics. Studies were deemed to be at moder- ate risk of bias if HHC was also measured in visitors, if it was clearly specified that patient characteristics differed before and after the intervention and no adjustment was made during data analysis, or if the study contained some missing data that could not be addressed during data analysis.

Data analysis

A qualitative summary of included studies was under- taken. Principal study characteristics and major findings were summarized into tables. Several scatter plots were created to observe any emerging trends. Trendlines were fitted using the ‘trendline’ function in R, and the line with the highest R2 value was retained. Due to poor study reporting, such as lack of number of HHC opportunities and lack of sample size for HAI incidence rate calculations, and overall heterogeneity in how studies were conducted and reported, it was not possible to perform descriptive statistics and create forest plots.

Results

Characteristics of studies

In total, 8093 articles were identified through the database search. After removal of duplicates, 6657 citations were

8093 records identified through database searching

PubMed (2179)

EMBASE (3722)

ICTRP (7)

Cochrane trials (2185)

6657 records screened for title and abstract

298 full-text articles assessed for eligibility

35 studies included in qualitative synthesis

Figure 1. PRISMA flowchart. HHC, hand hygiene compliance;

screened, among which 35 articles were included in the sys- tematic review (Figure 1) [12e46].

Location and study design

HHC and HAIs are a topic of interest as evidenced by the wide range of high-income countries conducting research on the topic. This research was predominantly conducted in North America (49%, N¼17) and Europe (29%, N¼10). Eight studies were conducted in Asia and the Middle East (Table I). Most studies had a beforeeafter study design (20/25); only two of 35 studies were randomized controlled trials [13,32] (Table I).

Setting

Most studies were conducted in hospital facilities, among which the intensive care unit (ICU) setting was most common, accounting for 46% (N¼16) of clinical settings where research was performed. Two studies were conducted exclusively in a long-term care facility [32,35]. Among hospital facilities, 10 studies were conducted across the entire hospital, three studies were conducted in single hospital units, and three other studies were conducted across multiple wards within the same hospital (Table I).

1436 duplicates removed

6359 records excluded based on

• Setting (middle- or low-income

country)

• Study population (dentist or

veterinarian)

• Study design (grey literature, reports)

• Only HHC rates reported

• HHC and HAI incidence rates not

reported over the same time frame

263 full-text articles excluded, with reasons

• Grey literature (118)

• Article could not be found (15)

• Research protocol/report (16)

• Duplicate article (9)

• Non high-income country (6)

• Simulated data (9)

• HHC not measured on HCWs (38)

• HHC and incidence/prevalence rate not

reported (29)

• Lack of temporality (23)

HAI, hospital-acquired infection; HCW, healthcare worker.

Table I

Summary of study characteristics

Characteristic Number of studies References

Study location 35 North America 17

Canada 2 [24,26] USA 15 [12,17,19,20,22,23,28,30,36,37,40e42,44,46]

Europe 10 France 1 [39] Italy 2 [14,16] Germany 1 [13] Netherlands 2 [21,33] Switzerland 3 [34,38,45] Spain 1 [29]

Asia 5 Hong Kong 1 [32] Japan 1 [25] Taiwan 3 [18,35,43]

Middle East 3 Israel 1 [31] Saudi Arabia 1 [15] Kuwait 1 [27]

Study design Beforeeafter 20 [15,17e20,22,23,25e27,29,34,35,37,38,40,41,43,45,46] Cross-sectional 1 [24] Interrupted time series 4 [16,21,28,33] Observational studies 4 [14,31,36,39] Quasi-experimental 4 [12,30,42,44] Randomized controlled trial 2 [13,32]

Setting Hospital 33

Intensive care unit 15 [13,16,20e22,25e27,33,36,40,41,43,44,46] Single hospital ward 4 [15,30,38,42] Multiple hospital wards 3 [12,14,39] Multiple hospitals 1 [19] Hospital wide 10 [17,18,23,24,28,29,31,34,35,37,45,56]

Long-term care facility 2 [32,35] Hand hygiene guideline

Direct observation 33 Observational study 1 [41] Electronic monitoring 1 [42]

Type of infections Clostridium difficile infection 7 [12,14,22,24,27,29,37] Catheter colonization 1 [42] MRSA colonization 2 [29,39] VRE colonization 1 [40]

Class of HAI Bloodstream infection 7 [17,18,24,32,42,43,45] Catheter-associated urinary tract infection 4 [15,20,27,36] Central-line-associated bloodstream infection

7 [16,19,20,23,27,36,41]

Lower respiratory tract infection 2 [27,42] Respiratory tract infection 4 [18,29,42,43] Urinary tract infection 5 [18,29,32,42,43]

Resistant pathogens Any resistant pathogen 1 [22] Highly resistant enterococci 1 [21] MDRO 3 [13,21,41] MRSA 7 [17,21,22,24,25,27,45] VRE 3 [21,22,40]

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e4836

Table I (continued)

Characteristic Number of studies References

Specific pathogens Acinetobacter baumannii 1 [27] Escherichia coli Klebsiella pneumoniae Pseudomonas aeruginosa Coagulase-negative staphylococcus 1 [44]

Non-specified pathogens Hospital-acquired infections 17 [12,15,18,22,26e30,33e35,38,42,43,45,46]

Risk of bias Critical 11 [12,15,18e20,22,23,28,34,37,41] High 1 [32] Serious 9 [17,24,31,33,35,36,39,45,46] Moderate 3 [21,26,42] Low 11 [13,14,16,25,27,29,30,38,40,43,44]

MRSA, meticillin-resistant Staphylococcus aureus; VRE, vancomycin-resistant enterococci; HAI, hospital-acquired infection; MDRO, multi-drug- resistant organism.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e48 37

Risk of bias

The included studies were very heterogenous in terms of study reporting, data analysis methods and overall study design. Only 11 (31%) studies were deemed to have low risk of bias as per the risk-of-bias assessment (Table I).

Summary of findings

Hand hygiene compliance

The HHC rate was determined mainly by direct observation; however, one study used an electronic device [42] as the method of measurement to document compliance. Most stud- ies reported overall HHC rates alone, while one study reported HHC after patient contact alone [42], and two studies reported HHC by moment of hand hygiene alone [20,24]. Reported HHC varied widely, ranging from 19% to 100% in the hospital setting and from 9% to 33% in long-term care facilities. Among all 35 studies, a total of 111 HHC rates were reported; the majority of HHC rates were <60% (50/111), while very few HHC rates >80% were reported (18/111). Forty-three of 111 HHC rates were between 60% and 79% (Figure 2).

Hospital-acquired infection

A wide range of HAI types were reported; seven studies reported specific types of HAI, while some studies reported HAIs caused by specific pathogens. The most common patho- gens reported were MRSA (9/35) and Clostridium difficile (7/ 35), but most studies reported overall HAIs without further stratification (Table I). In terms of types of HAI, the most common types reported were central-line-associated blood- stream infection (CLABSI) (7/35) and urinary tract infection (UTI) (5/35) (Table I). HAI incidence rates were reported with variable units of measure, among which the most common were per 1000 patient-days (22/35) and per 1000 device-days (9/35) (Table II).

In total, 22 studies reported a wide range of HAIs per 1000 patient-days (Table II). When all HAI incidence rates were plotted against HHC rates, a negative trend emerged.

However, above an HHC rate of approximately 60%, there did not appear to be any additional improvement in the decreasing rate trend for HAIs, with the dispersion of HAI incidence rate point estimates similar from HHC rates of 60% upwards (Figure 3); this was labelled ‘beneficial’. Studies were further stratified, and graphs were produced for HAI types that had a sample size �10. The beneficial HHC rates were 60% for resistant pathogens (Figure 4) and 60% for non-specified HAIs (Figure 5). The second most reported unit of measure was catheter-days for catheter-associated urinary tract infection (CAUTI), CLABSI and UTI, and ventilator-days for lower respi- ratory tract infection (LRTI), all combined under the heading ‘device-days’. A scatter plot was produced for these HAIs with no clear beneficial threshold above which HAI incidence rates were lower (Figure 6).

Graphs for all remaining HAIs (% rate, 100 days at risk, 1000 admission-days, 1000 bed-days, 1000 inpatient admissions) were not produced because they had �10 data points, or the data points were derived from fewer than five studies (Table II) (Table III).

Discussion

The objective of this systematic review was to determine the HHC rate associated with the lowest HAI incidence rate. This question has been raised in several reports [47,48], but to the authors’ knowledge, this is the first systematic review to address this question. In this review, the ICU was the most common clinical setting where research was conducted. This may suggest that HAIs are more prevalent in ICUs, given vul- nerable patients at risk for HAIs and a higher use of invasive procedures [7], all of which underscore the importance of high HH compliance. In addition, the most frequently reported HHC rate was between 60% and 70%, showing that despite the rate of 80% recommended by the World Health Organization (WHO), this target is extremely difficult to reach, even in developed countries where access to sanitary facilities, handwashing sinks and alcohol-based hand-rub stations are not limited. Despite an HHC rate of 100%, a study reported an outbreak of multi- drug-resistant organisms [22]. The authors argued, among other things, that at such high HHC rates, environmental

30

25

20

15

10

10 20 30 40 50 60 70 80 90 100

5

0

HHC rate (%)

N u m

b e r

o f

ti m

e s

H H

C r

a te

w a s

re p o rt

e d

Figure 2. Histogram of all hand hygiene compliance (HHC) rates reported in the studies included in this review.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e4838

cleaning takes on a more important role in the prevention of HAIs. Furthermore, this perfect HHC score may not reflect reality. It would be interesting to compare HHC monitored through direct observation with electronic surveillance, in which the latter might be better correlated with real HHC rates, enabling better appreciation for the correlation

Table II

Characteristics of reported hospital-acquired infections (HAIs)

HAI unit type HAI type Nu

% rate BSI, MRSA, HAI, RV 100 days at risk HRE, MRSA, VRE 1000 admission-days BSI, HAI, RTI, UTI 1000 device-days CAUTI, CLABSI, UTI, LRTI 1000 inpatient admission MDRO 1000 patient-days AB, any resistant pathogen, BSI,

catheter colonization, CDI, CONS EC, EC and Group B streptococcus MDRO, MRSA, MRSA bacteraemia, MRSA colonization, non-specified, KP, PA, RTI, UTI, VRE, VRE bacteraemia

AB, Acinetobacter baumannii; BSI, bloodstream infection; CAUTI, catheter CLABSI, central-line-associated bloodstream infection; CONS, coagulase- enterococci; KP, Klebsiella pneumoniae; LRTI, lower respiratory tract infe Staphylococcus aureus; PA, Pseudomonas aeruginosa; RTI, respiratory vancomycin-resistant enterococci.

between the HHC rates obtained and the HAI incidence rates observed).

From the studies included in this review, an HHC rate of 60% seemed to be the point at which low HAI incidence rates were observed for most studies. Of course, this comes with major caveats given poor study designs and reporting, and the

mber of point

estimates

Number of

studies

References

11 5 [26,31,34,39,45] 9 1 [21] 8 1 [18] 40 9 [15,16,19,20,23,27,29,36,41] 2 1 [41] 153 25 [12e15,17,18,21,22,24

e30,32,33,35,37e39,42e45]

-associated urinary tract infection; CDI, Clostridium difficile infection; negative staphylococcus; EC, Escherichia coli; HRE, highly resistant ction; MDRO, multi-drug-resistant organism; MRSA, meticillin-resistant tract infection; RV, rotavirus; UTI, urinary tract infection; VRE,

60

50

40

30

20

10

0

20 40 60 80 100

HHC rate (%)

H A

I in

c id

e n c e r

a te

/1 0 0 0 p

a ti

e n t-

d a y s

y = 60.2 e–0.0753x +1.98 R2 = 0.165

Figure 3. Scatter plot of hand hygiene compliance (HHC) rate and all hospital-acquired infections (HAIs) reported per 1000 patient-days.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e48 39

presence of concurrent infection prevention and control (IPC) strategies that are not always considered when results are reported. The studies included in this review were not initially designed to find the optimal HHC rate; most were designed to evaluate the impact of an intervention to increase HHC, with the impact on HAIs being a secondary objective. As such, these studies were not designed to answer the present research question. Nevertheless, they provided an observed

8

6

4

2

0

50 60 70

HHC

In c id

e n c e r

a te

o f

re si

st a n t

p a th

o g e n s/

1 0 0 0 p

a ti

e n t-

d a y s y = 0.00223 x

2 – 0.299 x + 9.95

R2 = 0.271

Figure 4. Scatter plot of hand hygiene compliance (HHC) rate and incid drug-resistant organisms’, ‘meticillin-resistant Staphylococcus aureus days.

trend for the association between HHC and HAIs. However, causality was impossible to assess. For studies reporting on CAUTI, CLABSI and LRTI per device-days, an HHC rate of 50% seemed to be the point at which low HAI incidence rates were observed. It is important to note that several aseptic meas- ures and specific IPC strategies are usually implemented for patients with invasive devices to minimize the risk of infection [2,49]. The apparent beneficial impact of this lower HHC rate

80 90 100

rate (%)

ence rate of resistant pathogens (‘any resistant pathogen’, ‘multi- ’, ‘vancomycin-resistant enterococci’) reported per 1000 patient-

60

30

40

50

20

10

0

20 30 40 50 60 70 80 90

HHC rate (%)

In c id

e n c e r

a te

o f

n o n -s

p e c if

ie d H

A I/

1 0 0 0 p

a ti

e n t-

d a y s y = 0.0185 x2 – 2.56 x + 92.6

R2 = 0.729

Figure 5. Scatter plot of hand hygiene compliance (HHC) rate and non-specified hospital-acquired infections (HAIs) reported per 1000 patient-days.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e4840

is likely confounded by different concomitant IPC measures in effect. In addition, patients with invasive procedures and devices are at increased risk of being infected by their endogenous flora. HH would thus have limited impact on device-associated infections [50]. Furthermore, a lack of information was available on compliance with ‘Moment 2’ of the WHO indications (before aseptic task), which would be the most important indication for the prevention of device-

15

10

5

0

30 40 50 60

HHC

H A

I in

c id

e n c e r

a te

/1 0 0 0 d

e v ic

e -d

a y s

y = –4.15 In(x) + 20.8 R2 = 0.13

Figure 6. Scatter plot of hand hygiene compliance (HHC) rate and hos infections, central-line-associated bloodstream infections and lower r

associated infections. Most of the studies included in this review reported either the overall HHC rate [measured either during ‘Moment 1’ (before patient contact) or ‘Moment 4’ (after patient contact)] or the HHC rate measured at Moment 4 alone. Unfortunately, many studies on HHC record HH opportunities upon room entry (used as a proxy for Moment 1) and/or room exit (used as a proxy for Moments 4 and 5) [51,52], hence creating uncertainty regarding whether the

70 80 90 100

rate (%)

pital-acquired infections (HAIs) (catheter-associated urinary tract espiratory tract infections) reported per 1000 device-days.

Table III

Characteristics of included studies

Year, first author,

country [ref]

Setting Study design Intervention Assessment of bias

Hand hygiene compliance

Hospital- acquired infection

2019

Boyce et al.,

USA [12]

Multiple wards Quasi-

experimental

Automated HH

system and

other

educational

strategies on HH

promotion

Critical P1: 28.0

P2: 32.2, P<0.01

P3: 40.6, P<0.01

P4: 51.8, P<0.01

CDI

P1: 7.9/10,000 PD

P2: 6.1/10,000 PD,

P¼0.41 P3: 8.3/10,000 PD,

P¼0.90 P4:12.6/10,000 PD,

P¼0.05

Non-CDI

P1: 5.7/10,000 PD

P2: 3.7/10,000 PD,

P¼0.29 P3: 7.2/10,000 PD,

P¼0.60 P4: 2.5/10,000 PD,

P¼0.84 2019

Von Lengerke et al.,

Germany [13]

Multiple ICUs Randomized

controlled trial

PSYchological

optimized hand

hyGIENE

promotion

Low Control unit

P1: 55.0

P2: 68.0, P<0.01

P3: 64.0

Intervention unit

P1: 54.0

P2: 64.0, P<0.01

P3: 70.0

MDRO control unit

P1: 0.691/10,000 IpD

P2: 0.605/10,000

IpD, P¼0.90 P3: 0.669/1000 IpD

Intervention unit

P1: 0.845/10,000 IpD

P2: 0.585/10,000

IpD, P¼0.015 P3: 0.348/1000 IpD

2018

Ragusa et al.,

Italy [14]

Multiple wards Retrospective

cohort

NA Low ICU

P1: 58.0

P2: 60.0

Gen med unit

P1: 47.0

P2: 44.0

Surgical unit

P1: 72.0

P2: 65.0

Paediatric unit

P1: 78.0

P2: 65.0

CDIeICU

P1: 0.39/10,000 PD

P2: 0.37/10,000 PD

CDIegen med unit

P1: 1.13/10,000 PD

P2:1.64/10,000 PD

CDIesurgical unit

P1: 0.06/10,000 PD

P2: 0.189/10,000 PD

CDIepaediatric unit

P1: 0.19/10,000 PD

P2: 0.55/10,000 PD

2017

Al Kuwaiti,

Saudi Arabia [15]

Single ward Beforeeafter Hand hygiene

intervention,

staff education

Critical P1: 51.2

P2: 66.1

P3: 71.8, P<0.05

CAUTI

P1: 3.72/1000 CD

P2: 2.21/1000 CD

P3: 1.75/1000 CD,

P>0.05

HAI

P1: 3.37/1000 PD

P2: 2.42/1000 PD

P3: 2.55/1000 PD,

P<0.05

2017

Musu et al.,

Italy [16]

ICU Interrupted

time series

Surveillance and

educational/

training

programme

Low Unit 1

P1: 42.3

P2: 83.8

Unit 2

P1: 37.9

P2: 93.5

CLABSI unit 1

P1: 6.7/1000 CvcD

P2: 3.0/1000 CvcD

CLABSI unit 2

P1: 7.9/1000 CvcD

P2: 6.2/1000 CvcD

2017

Rupp et al.,

USA [17]

Hospital wide Beforeeafter Contact

precautions

Serious P1: 93.5

P2: 91.4

MRSA bacteraemia

P1: 0.18/1000 pd

P2: 0.18/1000 pd,

P>0.05

MRSA

P1: 0.55/1000 pd

P2: 0.48/1000 pd,

P>0.05

VRE bacteraemia

P1: 0.07/1000 pd

P2: 0.08/1000 pd

VRE

P1: 0.45/1000 pd

P2: 0.32/1000 pd

2016

Chen et al.,

Taiwan [18]

Hospital wide Beforeeafter Alcohol-based

hand-rub

dispensers,

educational

programmes

Critical P1: 62.3

P2: 73.3,

P<0.001

BSI

P1: 1.2/1000 AmD

P2: 1.0/1000 AmD,

P¼0.085

HAI

P1: 3.7/1000 AmD

P2: 3.1/1000 AmD,

P¼0.002

UTI

P1: 1.5/1000 AmD

P2: 1.2/1000 AmD, P<0.009

RTI

P1: 0.53/1000 AmD

P2: 0.35/1000 AmD,

P¼0.025

(continued on next page)

V . M o u a jo u e t a l.

/ J o u rn a l o f H o sp ita

l In fe ctio

n 1 1 9 (2 0 2 2 ) 3 3 e 4 8

4 1

Table III (continued )

Year, first author,

country [ref]

Setting Study design Intervention Assessment of bias

Hand hygiene compliance

Hospital- acquired infection

2016

Shabot et al.,

USA [19]

Several

hospitals

Beforeeafter Educational

programme

Critical P1: 58.1

P2: 84.4, P<0.01

P3: 94.7, P<0.01

P4: 95.6, P<0.01

CLABSI

P1: 0.83/1000 CLD

P2: 0.63/1000 CLD

P3: 0.42/1000 CLD

P4: 0.58/1000 CLD

P1, P4; P<0.05

2015

Fox et al.,

USA [20]

ICU Beforeeafter Patient hand

hygiene

protocol

Critical P1: 35.0

P2: 66.0

CLABSI

P1: 1.1/1000 CD

P2: 0.50/1000 CD,

P¼0.64

CAUTI

P1: 9.1/1000 CD

P2: 5.6/1000 CD,

P¼0.24 2014

Derde et al.,

Netherlands [21]

ICU Interrupted

time series

Universal

chlorhexidine

body-washing

combined with

hand hygiene

improvement

Moderate P1: 51.5

P2: 69.0

P3: 76.7

MRSA

P1: 4.0/100 days at

risk

P2: 3.4/100 days at

risk

P3: 1.5/100 days at

risk

VRE

P1: 0.40/100 DAR

P2: 0.74/100 DAR

P3: 1.02/100 DAR

HRE

P1: 9.77/100 DAR

P2: 9.55/100 DAR

P3: 5.91/100 DAR

2014

Jayaraman et al.,

USA [22]

ICU Beforeeafter NA Critical Surgical ICU

P1: 100

P2: 97.6, P¼0.93 Trauma ICU

P1: 90.0

P2: 96.75, P¼0.14

MRSA

Surg. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD,

P¼0.04 Trau. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD,

P¼0.04

VRE

Surg. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD,

P¼0.04 Trau. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD,

P¼0.04

CDI

Surg. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD, P¼0.04 Trau. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD, P¼0.04

Any resistant pathogen

Surg. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD, P¼0.04 Trau. ICU

P1: 6.7/1000 PD

P2: 2.7/1000 PD, P¼0.04

2014

Johnson et al.,

USA [23]

Hospital wide Beforeeafter Education e

quality

improvement

project

Critical P1: 58.0

P2: 96.0

P3: 98.0

CLABSI

P1: 4.1/1000 DD

P2: 1.1/1000 DD

P3: 0.4/1000 DD

2013

DiDiodato,

Canada [24]

Hospital wide Cross-sectional Education and

training

programme

Serious P1: 54.3

P2: 67.5

P3: 73.4

P4: 79.6

CDI

P1: 0.32/1000 PD

P2: 0.41/1000 PD

P3: 0.39/1000 PD

P4: 0.37/1000 PD

MRSA bacteraemia

P1: 0.02/1000 PD

P2: 0.01/1000 PD

P3: 0.01/1000 PD

P4: 0.01/1000 PD

2013

Morioka et al.,

Japan [25]

ICU Beforeeafter Contact

precautions

Low P1: 50.0

P2: 75.0

MRSA

P1: 3.5/1000 PD

P2: 1.3/1000 PD,

P<0.01

2013

Mukerji et al.,

Canada [26]

ICU Beforeeafter HH promotional

programme e

education

Moderate P1: 76.0

P2: 67.0

P3: 76.0

HAI

P1: 4.0%

P2: 6.0%

P3: 4.0%

2013

Salama et al.,

Kuwait [27]

ICU Beforeeafter Education

programme

Low P1: 43.0

P2: 61.4, P<0.01

CDI

P1: 0.2/1000 PD

P2: 0.0/1000 PD,

P¼0.2

CAUTI

P1: 5.5/1000 CD

P2: 5.9/1000 CD,

P>0.4

CLABSI

P1: 18.6/1000 CD

P2: 3.4/1000 CD, P<0.01

MRSA

P1: 0.9/1000 PD

P2: 0.0/1000 PD, P<0.01

AB

P1: 5.4/1000 CD

V . M o u a jo u e t a l.

/ J o u rn a l o f H o sp ita

l In fe ctio

n 1 1 9 (2 0 2 2 ) 3 3 e 4 8

4 2

E. coli

P1: 1.1/1000 PD

P2: 0.7/1000 PD,

P>0.2

P2: 3.5/1000 CD,

P<0.02

LTRI

P1: 17.6/1000 VD

P2: 5.2/1000 VD, P<0.01

HAI

P1: 37.2/1000 PD

P2: 15.1/1000 PD, P<0.01

KP

P1: 2.7/1000 PD

P2: 0.7/1000 PD,

P<0.01

PA

P1: 2.7/1000 PD

P2: 0/1000 PD,

P<0.01

2012

Kirkland et al., USA

[28]

Hospital wide Interrupted

time series

Education and

improvement

programme

Critical P1: 71.5

P2: 89.0

þþþ

HAI

P1: 4.8/1000 PD

P2: 3.3/1000 PD

2012

Monistrol et al.,

Spain [29]

Hospital wide Beforeeafter Hand hygiene

education

programme

Low P1: 54.3

P2: 75.8,

P¼0.005 P3: 75.8

CAUTI

P1: 5.5/1000 CD

P2: 3.5/1000 CD,

P¼0.2

UTI

P1: 3.3/1000 HD

P2: .2.2/1000 HD,

P¼0.2

MRSA colonization

P1: 0.92/1000 HD

P2: 0.25/1000 HD, P¼0.2 P3: 0.15/1000 HD

HAI

P1: 6.93/1000 HD

P2: 6.96/1000 HD, P¼0.9

RTI

P1: 0.51/1000 HD

P2: 0.89/1000 HD,

P¼0.5

CDI

P1: 0.13/1000 HD

P2: 0.63/1000 HD,

P¼0.2 2011

Harne et al., USA

[30]

Single ward Quasi-

experimental

Education

paired with

positive

reinforcement

behavioural

interventions

Low Group 1

P1: 75.0

P2: 66.6

Group 2

P1: 60.8

P2: 66.2

Group 3

P1: 63.3

P2: 62.4

HAI group 1

P1: 4.70/1000 PD

P2: 2.66/1000 PD

HAI group 2

P1: 5.80/1000 PD

P2: 4.80 1000 PD

HAI group 3

P1: 3.10/1000 PD

P2: 2.31/1000 PD

2011

Waisbourd -Zinman

et al.,

Israel [31]

Hospital wide Prospective

cohort

None reported Serious P1: 33.7

P2: 49.0, P<0.01

Rotavirus

P1: 20.3%

P2: 12.7%, P<0.01

2011

Yeung et al., Hong

Kong [32]

Long-term care

facility

Randomized

controlled trial

Alcohol-based

antiseptic hand

rub

High Treatment

P1: 25.8

P2: 33.0

Control

P1: 25.8

P2: 30.0, P¼0.01

Serious infection

treatment

P1: 1.42/1000 RD

P2: 0.65/1000 RD,

P¼0.002 Control

P1: 0.49/1000 RD

P2: 1.05/1000 RD,

P¼0.04

UTI treatment

P1: 0.27/1000 RD

P2: 0.16/1000 RD,

P¼0.30 Control

P1: 0.9/1000 RD

P2: 0.27/1000 RD,

P¼0.06

BSI treatment

P1: 0.18/1000 RD

P2: 0.06/1000 RD, P¼0.12 BSI control

P1: 0.0/1000 RD

P2: 0.12/1000 RD, P¼0.05

2010

Helder et al.,

Netherlands [33]

ICU Interrupted

time series

Hand hygiene

programme

Serious P1: 68.8

P2: 86.9, P<0.01

HAI

P1: 17.3/1000 PD

P2: 13.5/1000 PD,

P¼0.03 2008

Ebnother et al.,

Switzerland [34]

Hospital wide Beforeeafter Infection

control

programme,

alcohol-based

hand rub

Critical P1: 59.0

P2: 79.0

HAI

P1: 11.7%c

P2: 6.8%

2008

Huang et al., Taiwan

[35]

Nursing home Beforeeafter Hand hygiene

training

programme

Serious P1: 9.3

P2: 30.4, P<0.01

HAI

P1: 2.04/1000 BD

P2: 1.52/1000 BD,

P<0.01

(continued on next page)

V . M o u a jo u e t a l.

/ J o u rn a l o f H o sp ita

l In fe ctio

n 1 1 9 (2 0 2 2 ) 3 3 e 4 8

4 3

Table III (continued )

Year, first author,

country [ref]

Setting Study design Intervention Assessment of bias

Hand hygiene compliance

Hospital- acquired infection

2008

Rupp et al., USA [36]

ICU Prospective,

controlled,

crossover trial

Alcohol-based

hand gel

Serious Unit 1

P1: 38.0

P2: 52.0

P3: 69.0

Unit 2

P1: 37.0

P2: 32.0

P3: 68.0

CLABSI unit 1

P1: 1.51/1000 DD

P2: 3.49/1000 DD

P3: 1.49/1000 DD

CLABSI unit 2

P1: 3.01/1000 DD

P2: 3.49/1000 DD

P3: 3.22/1000 DD

CAUTI unit 1

P1: 1.10/1000 DD

P2: 1.98/1000 DD

P3: 3.87/1000 DD

CAUTI unit 2

P1: 2.99/1000 DD

P2: 4.57/1000 DD

P3: 3.37/1000 DD

2007

Muto et al., USA [37]

Hospital wide Beforeeafter CD infection

control bundle

Critical P1: 79.3

P2: 85.4

P3: 73.3

P4: 67.3

CDI

P1: 0.81/1000 PD

P2: 0.83/1000 PD

P3: 0.72/1000 PD

P4: 0.89/1000 PD

2007

Pessoa-Silva et al.,

Switzerland [38]

Single unit Beforeeafter Hand hygiene

promotion

Low P1: 42.0

P2: 45.0

P3: 55.0, P¼NA

HAI

P1: 11.1/1000 PD

P2: 7.9/1000 PD

P3: 8.2/1000 PD,

P¼NA 2006

Girou et al., France

[39]

Multiple units Prospective

cohort

HH promotion

campaign

Serious P1: 51.7

P2: 67.0, P<0.01

MRSA colonization

P1: 16.1%b

P2: 20.5%, P¼0.18 2006

Hayden et al.,

USA [40]

ICU Beforeeafter Routine

environmental

cleaning

measures

Low P1: 40.0

P2: 57.0

P3: 29.0

P4: 43.0

VRE colonization

P1: 33.47/1000 PD

P2: 16.84/1000 PD

P3: 12.09/1000 PD

P4: 10.4/1000 PD

2004

Barnes et al.,

USA [41]

ICU Beforeeafter Waterless

alcohol-based

skin degermers

and lotion, staff

education,

elimination of

artificial nails

Critical P1: 53.0

P2: 79.0

CLABSI

P1: 3.9/1000 CLD

P2: 2.5/1000 CLD

MRO

P1: 3.4/1000 INPT

ADM

P2: 1.4/1000 INPT

ADM

2004

Swoboda et al.,

USA [42]

Single unit Quasi-

experimental

Electronic

monitoring and

computerized

voice prompts

Moderate P1: 19.1

P2: 27.3

P3: 24.4, P<0.05

HAI þþþ P1: 61.88/1000 PD

P2: 55.39/1000 PD

P3: 36.83/1000 PD

BSI

P1: 9.28/1000 PD

P2: 9.35/1000 PD

P3: 5.52/1000 PD

Lower RTI

P1: 3.10/1000 PD

P2: 1.44/1000 PD

P3: 0.0/1000 PD

Upper RTI

P1: 3.10/1000 PD

P2: 2.90/1000 PD

P3: 1.84/1000 PD

Catheter

colonization

P1: 9.28/1000 PD

P2: 7.91/1000 PD

P3: 7.36/1000 PD

UTI

P1: 15.47/1000

PDP2: 14.39/1000

PD

P3: 12.81/1000 PD

2004

Won et al.,

Taiwan [43]

ICU Beforeeafter Multi-modal

campaign for

hand hygiene

promotion

Low P1: 43.0

P2: 80.0

HAI

P1: 15.13/1000 PD

P2: 10.69/1000 PD,

P¼0.003

BSI

P1: 4.47/1000 PD

P2: 3.90/1000 PD,

P¼0.558

RTI

P1: 3.35/1000 PD

P2: 1.06/1000 PD, P¼0.002

UTI

P1: 1.01/1000 PD

P2: 0.47/1000 PD, P¼0.185

V . M o u a jo u e t a l.

/ J o u rn a l o f H o sp ita

l In fe ctio

n 1 1 9 (2 0 2 2 ) 3 3 e 4 8

4 4

N o n -s p e c ifi e d

P 1 : 4 .1 4 / 1 0 0 0 P D

P 2 : 2 .7 8 / 1 0 0 0 P D ,

P ¼0

.0 5 5

2 0 0 2

S h a re k e t a l. ,

U S A [4 4 ]

IC U

Q u a si -

e x p e ri m e n ta l

E d u c a ti o n a l a n d

b e h a v io u ra l

in te rv e n ti o n

L o w

P 1 : 4 7 .4

P 2 : 8 5 .4 , P < 0 .0 0 1

C O N S

P 1 : 6 .1 / 1 0 0 0 P D

P 2 : 3 .2 / 1 0 0 0 P D ,

P ¼0

.0 0 5

E . co li a n d

st re p to c o c c u s

P 1 : 0 .1 9 / 1 0 0 0 P D

P 2 : 0 .3 7 / 1 0 0 0 P D ,

P ¼0

.9 2

2 0 0 0

P it te t e t a l. ,

S w it z e rl a n d [4 5 ]

H o sp it a l w id e

B e fo re e a ft e r

H a n d h y g ie n e

p ro m o ti o n

c a m p a ig n

S e ri o u s

P 1 : 4 7 .6

P 2 : 6 6 .2 , P < 0 .0 0 1

M R S A b a c te ra e m ia

P 1 : 0 .7 4 / 1 0 ,0 0 0 P D

P 2 : 0 .2 4 / 1 0 ,0 0 0 P D ,

P < 0 .0 0 1

M R S A tr a n sm

is si o n

P 1 : 2 .1 6 / 1 0 ,0 0 0 P D

P 2 : 0 .9 3 / 1 0 ,0 0 0 P D ,

P < 0 .0 0 1

H A I

P 1 : 1 6 .9 % a

P 2 : 9 .9 % , P ¼0

.0 4

1 9 9 0

S im

m o n s e t a l. ,

U S A [4 6 ]

IC U

B e fo re e a ft e r

E d u c a ti o n

c a m p a ig n

S e ri o u s

P 1 : 2 2 .0

P 2 : 2 9 .9 , P ¼0

.0 7 1

H A I

P 1 : 2 .9 / 1 0 0 IC U -d a y s

P 2 : 2 .8 / 1 0 0 IC U -d a y s

þþ þ,

m a n u a ll y c a lc u la te d ; A m D , a d m is si o n -d a y s;

A O B , a ss e ss m e n t o f b ia s;

B D , b e d -d a y s;

B S I, b lo o d st re a m

in fe c ti o n ; C A U T I, c a th e te r- a ss o c ia te d u ri n a ry

tr a c t in fe c ti o n ; C D , c a th e te r- d a y s;

C D I, C lo st ri d iu m

d if fi ci le

in fe c ti o n ; C L A B S I, c e n tr a l- li n e -a ss o c ia te d b lo o d st re a m

in fe c ti o n ; C L D , c e n tr a l- li n e -d a y s;

C v c D , c e n tr a l- v e n o u s- c a th e te r- d a y s;

D A R , d a y s a t ri sk ; D D , d e v ic e -d a y s;

E . co li , E sc h e ri ch

ia co li ; H D , h o sp it a l- d a y s;

H H , h a n d h y g ie n e ; H R E , h ig h ly

re si st a n t e n te ro c o c c i;

IC U , in te n si v e c a re

u n it ; IN P T A D M , in p a ti e n t a d m is si o n ; Ip D , in p a ti e n t- d a y s;

K P , K le b -

si e ll a p n e u m o n ia e ; M D R O , m u lt i- d ru g -r e si st a n t o rg a n is m s;

M R O , m u lt i- re si st a n t o rg a n is m s;

M R S A , m e ti c il li n -r e si st a n t S ta p h y lo co cc u s a u re u s;

P D , p a ti e n t- d a y s;

R D , re si d e n t- d a y s;

R T I, re s-

p ir a to ry

tr a c t in fe c ti o n ; U T I, u ri n a ry

tr a c t in fe c ti o n ; V R E , v a n c o m y c in -r e si st a n t e n te ro c o c c i.

a P re v a le n c e , c a se s o f n o so c o m ia l in fe c ti o n / 1 0 0 a d m is si o n s.

b P a ti e n ts

w it h M R S A c a rr ia g e / p a ti e n ts

sc re e n e d .

c P re v a le n c e ra te

o f n o so c o m ia l in fe c ti o n s in

th e in st it u ti o n .

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e48 45

HHC rate observed can be correlated with certain types of HAI such as device-associated HAIs.

In terms of HHC assessment, direct observation has been established as the gold standard for HHC monitoring despite its evident flaws (i.e. small number of HHC opportunities that can be measured, Hawthorne effect, etc.), as it is considered to be the only system that enables the recording of detailed infor- mation on HH events. Alternatively, electronic monitoring has the advantage of being able to survey a large number of opportunities, giving information on when alcohol-based hand rub is being dispensed more frequently [53], hence giving a more accurate picture of HHC on a given ward at a given time. However, as highlighted by Cawthorne and Cooke as electronic monitoring does not provide granular data on HH performance (i.e. how, how long, when) [54], it does not overcome the bias induced by artificially increasing HHC due to the presence of an evaluation system (be it an observer or an electronic monitor), and it does not inform on HH technique, thus making it impossible to provide feedback on proper HH technique, which is equally important. It has been reported that electronic monitoring gives lower HHC rates compared with direct observation [55]. There is still a need for strong evidence indicating that electronic monitoring is better than direct observation, and its correlation with HAI incidence rates [53,54].

Finally, most studies reported overall HAIs without speci- fying either the type of HAI or specific pathogens. This is a problem as the pathophysiology of various HAIs differs and the causal pathway may vary, making HH more or less vital in the prevention of these HAIs. Moreover, as different hospitals perform surveillance for different pathogens and different types of HAI, an ‘overall HAI’ rate is not insightful. This not only hinders interstudy comparability but fails to provide data that could be used to detect if optimal HHC rates differ by type of pathogen.

This systematic review had several limitations. First, most studies did not report raw data (i.e. sample size, HH oppor- tunities, HAI numerator and denominator, hospital size) which hindered the ability to perform robust analysis. The authors can only comment on the trends rather than on specific stat- istical differences. In addition, most studies were beforeeafter designs with no comparison group. A reduction in HAI incidence rates could be the result of natural variability over the obser- vation period. Most studies failed to report how observers were trained, what IPC measures were in place in their facility, and if adjustment for risk factors was made when calculating HAI incidence or performing statistical testing. Furthermore, the Hawthorne effect, whereby rates of HHC are artificially infla- ted when HCWs are aware they are being observed, has been reported to be a cause of bias when HHC observations are measured by direct observation [7], so the high HHC rates observed may not reflect the actual HHC rates. There could also be an observation bias in length of time of HHC observa- tion, where high HHC rates observed within a short period may not be sustained after the observed period, falsely classifying HCWs as being highly compliant when they are not and thus misclassifying the exposure. This review was also restricted to high-income countries, to serve as a comparison for environ- ments similar to that of Canada. Its application to low- or middle-income countries should be interpreted with caution. Finally, sample size was not assessed in this review as very few studies reported the sample size of HCWs on whom HHC was

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e4846

evaluated, so this information was not extracted systemati- cally. Unfortunately, these limitations are very common in studies evaluating HHC, as it speaks to the difficulty in con- ducting properly designed studies on this topic. McLaws pro- vided a detailed explanation of the most common methodological flaws in studies evaluating the relationship between HH and HAIs [49]. The difficulty in designing robust studies in this field mainly lies in the lack of benchmarks. There are no guidelines on sampling strategies (for example the recommended sample size of HCW and the number of HH opportunities necessary to obtain a representative HHC rate for a given ward), no standardised method of training direct observers and assessing inter-rate reliability, no recommended observation period required to successfully infer a reduction in HAI rates to an increase in HHC rates, no recommended method of accounting for concomitant IPC strategies and there are no recommended types of HAIs (shown to be susceptible to HHC) that should be studied. Without a standardized approach to the research methodology on this topic, different studies will most likely continue to give different results, hence hindering the necessary evidence-based research.

In conclusion, most studies reported an HHC rate between 60% and 70%, and very high HHC rates remain difficult to reach. HHC and HAI followed a negative relationship up to HHC of approximately 60%, above which the incremental benefits of higher HHC rates had little effect on HAI incidence rates. However, the available evidence was mostly of low quality, and the lack of raw data and the heterogeneity of reported data prevented statistical analysis and the pooling of data, so only visual trends could be discussed; this calls for cautious inter- pretation. More evidence is needed to support the idea that very high HHC rates can curb HAI incidence rates significantly beyond what is currently observed.

Acknowledgements

The authors wish to thank Genevieve Gore for assistance in the literature search strategy.

Conflict of interest statement None declared.

Funding source This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. CQ is supported by an external salary award from the Fonds de recherche du Québec e Santé.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jhin.2021.09.016.

References

[1] World Health Organization. WHO guidelines on hand hygiene in health care e First Global Patient Safety Challenge e clean care is safer care. Geneva: WHO; 2009.

[2] World Health Organization. Prevention of hospital-acquired infections e a practical guide. 2nd ed. Geneva: WHO; 2002.

[3] Haque M, Sartelli M, McKimm J, Abu Bakar M. Health care- associated infections e an overview. Infect Drug Resist 2018;11:2321e33.

[4] Bolton P, McCulloch TJ. The evidence supporting WHO recom- mendations on the promotion of hand hygiene: a critique. BMC Res Notes 2018;11:899.

[5] Shekelle PG, Wachter RM, Pronovost PJ, Schoelles K, McDonald K, Dy S, et al. Making Health Care Safer II: an updated critical analysis of the evidence for patient safety practices. In: Com- parative effectiveness review no. 211. Bethesda, MD: Agency for Healthcare Research and Quality; 2018.

[6] Erasmus V, Daha TJ, Brug H, Richardus JH, Behrendt MD, Vos MC, et al. Systematic review of studies on compliance with hand hygiene guidelines in hospital care. Infect Control Hosp Epidemiol 2010;31:283e94.

[7] Kingston L, O’Connell NH, Dunne CP. Hand hygiene-related clin- ical trials reported since 2010: a systematic review. J Hosp Infect 2016;92:309e20.

[8] Larson E. State-of-the-science e 2004: time for a ‘no excuses/no tolerance’ (NET) strategy. Am J Infect Control 2005;33:548e57.

[9] Beggs CB, Shepherd SJ, Kerr KG. How does healthcare worker hand hygiene behaviour impact upon the transmission of MRSA between patients? An analysis using a Monte Carlo model. BMC Infect Dis 2009;9:64.

[10] Sterne JA, Hernán MA, Reeves BC, Savović J, Berkman ND, Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016;355:i4919.

[11] Sterne JAC, Savovi�c J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898.

[12] Boyce JM, Laughman JA, Ader MH, Wagner PT, Parker AE, Arbogast JW. Impact of an automated hand hygiene monitoring system and additional promotional activities on hand hygiene performance rates and healthcare-associated infections. Infect Control Hosp Epidemiol 2019;40:741e7.

[13] von Lengerke T, Ebadi E, Schock B, Krauth C, Lange K, Stahmeyer JT, et al. Impact of psychologically tailored hand hygiene interventions on nosocomial infections with multidrug- resistant organisms: results of the cluster-randomized controlled trial PSYGIENE. Antimicrob Resist Infect Control 2019;8:56.

[14] Ragusa R, Giorgianni G, Lupo L, Sciacca A, Rametta S, La Verde M, et al. Healthcare-associated Clostridium difficile infection: role of correct hand hygiene in cross-infection control. J Prev Med Hyg 2018;59:E145e52.

[15] Al Kuwaiti A. Impact of a multicomponent hand hygiene inter- vention strategy in reducing infection rates at a university hos- pital in Saudi Arabia. Interv Med Appl Sci 2017;9:137e43.

[16] Musu M, Finco G, Mura P, Landoni G, Piazza MF, Messina M, et al. Controlling catheter-related bloodstream infections through a multi-centre educational programme for intensive care units. J Hosp Infect 2017;97:275e81.

[17] Rupp ME, Fitzgerald T, Hayes K, Van Schooneveld T, Hewlett A, Clevenger R, et al. Effect of cessation of contact isolation for endemic methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci. Infect Control Hosp Epidemiol 2017;38:1005e7.

[18] Chen JK, Wu KS, Lee SS, Lin HS, Tsai HC, Li CH, et al. Impact of implementation of the World Health Organization multimodal hand hygiene improvement strategy in a teaching hospital in Taiwan. Am J Infect Control 2016;44:222e7.

[19] Shabot MM, Chassin MR, France AC, Inurria J, Kendrick J, Schmaltz SP. Using the Targeted Solutions Tool� to improve hand hygiene compliance is associated with decreased health care- associated infections. Jt Comm J Qual Patient Saf 2016;42:6e17.

[20] Fox C, Wavra T, Drake DA, Mulligan D, Bennett YP, Nelson C, et al. Use of a patient hand hygiene protocol to reduce hospital- acquired infections and improve nurses’ hand washing. Am J Crit Care 2015;24:216e24.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e48 47

[21] Derde LPG, Cooper BS, Goossens H, Malhotra-Kumar S, Willems RJL, Gniadkowski M, et al. Interventions to reduce col- onisation and transmission of antimicrobial-resistant bacteria in intensive care units: an interrupted time series study and cluster randomised trial. Lancet Infect Dis 2014;14:31e9.

[22] Jayaraman SP, Klompas M, Bascom M, Liu X, Piszcz R, Rogers SO, Jr, et al. Hand-hygiene compliance does not predict rates of resistant infections in critically ill surgical patients. Surg Infect 2014;15:533e9.

[23] Johnson L, Grueber S, Schlotzhauer C, Phillips E, Bullock P, Basnett J, et al. A multifactorial action plan improves hand hygiene adherence and significantly reduces central line- associated bloodstream infections. Am J Infect Control 2014;42:1146e51.

[24] DiDiodato G. Has improved hand hygiene compliance reduced the risk of hospital-acquired infections among hospitalized patients in Ontario? Analysis of publicly reported patient safety data from 2008 to 2011. Infect Control Hosp Epidemiol 2013;34:605e10.

[25] Morioka I, Yahata M, Shibata A, Miwa A, Yokota T, Jikimoto T, et al. Impact of pre-emptive contact precautions for outborn neonates on the incidence of healthcare-associated meticillin- resistant Staphylococcus aureus transmission in a Japanese neo- natal intensive care unit. J Hosp Infect 2013;84:66e70.

[26] Mukerji A, Narciso J, Moore C, McGeer A, Kelly E, Shah V. An observational study of the hand hygiene initiative: a comparison of preintervention and postintervention outcomes. BMJ Open 2013;3:e003018.

[27] Salama MF, Jamal WY, Mousa HA, Al-Abdulghani KA, Rotimi VO. The effect of hand hygiene compliance on hospital-acquired infections in an ICU setting in a Kuwaiti teaching hospital. J Infect Public Health 2013;6:27e34.

[28] Kirkland KB, Homa KA, Lasky RA, Ptak JA, Taylor EA, Splaine ME. Impact of a hospital-wide hand hygiene initiative on healthcare- associated infections: results of an interrupted time series. BMJ Qual Saf 2012;21:1019e26.

[29] Monistrol O, Calbo E, Riera M, Nicolás C, Font R, Freixas N, et al. Impact of a hand hygiene educational programme on hospital- acquired infections in medical wards. Clin Microbiol Infect 2012;18:1212e8.

[30] Harne-Britner S, Allen M, Fowler KA. Improving hand hygiene adherence among nursing staff. J Nurs Care Qual 2011;26:39e48.

[31] Waisbourd-Zinman O, Ben-Ziony S, Solter E, Chodick G, Ashkenazi S, Livni G. The percentage of nosocomial-related out of total hospitalizations for rotavirus gastroenteritis and its associ- ation with hand hygiene compliance. Am J Infect Control 2011;39:166e8.

[32] Yeung WK, Tam WS, Wong TW. Clustered randomized controlled trial of a hand hygiene intervention involving pocket-sized con- tainers of alcohol-based hand rub for the control of infections in long-term care facilities. Infect Control Hosp Epidemiol 2011;32:67e76.

[33] Helder OK, Brug J, Looman CW, van Goudoever JB, Kornelisse RF. The impact of an education program on hand hygiene compliance and nosocomial infection incidence in an urban neonatal inten- sive care unit: an intervention study with before and after com- parison. Int J Nurs Stud 2010;47:1245e52.

[34] Ebnöther C, Tanner B, Schmid F, La Rocca V, Heinzer I, Bregenzer T. Impact of an infection control program on the prevalence of nosocomial infections at a tertiary care center in Switzerland. Infect Control Hosp Epidemiol 2008;29:38e43.

[35] Huang TT, Wu SC. Evaluation of a training programme on knowledge and compliance of nurse assistants’ hand hygiene in nursing homes. J Hosp Infect 2008;68:164e70.

[36] Rupp ME, Fitzgerald T, Puumala S, Anderson JR, Craig R, Iwen PC, et al. Prospective, controlled, cross-over trial of alcohol-based hand gel in critical care units. Infect Control Hosp Epidemiol 2008;29:8e15.

[37] Muto CA, Blank MK, Marsh JW, Vergis EN, O’Leary MM, Shutt KA, et al. Control of an outbreak of infection with the hypervirulent Clostridium difficile BI strain in a university hospital using a comprehensive ‘bundle’ approach. Clin Infect Dis 2007;45:1266e73.

[38] Pessoa-Silva CL, Hugonnet S, Pfister R, Touveneau S, Dharan S, Posfay-Barbe K, et al. Reduction of health care associated infection risk in neonates by successful hand hygiene promotion. Pediatrics 2007;120:e382e90.

[39] Girou E, Legrand P, Soing-Altrach S, Lemire A, Poulain C, Allaire A, et al. Association between hand hygiene compliance and methicillin-resistant Staphylococcus aureus prevalence in a French rehabilitation hospital. Infect Control Hosp Epidemiol 2006;27:1128e30.

[40] Hayden MK, Bonten MJ, Blom DW, Lyle EA, van de Vijver DA, Weinstein RA. Reduction in acquisition of vancomycin-resistant enterococcus after enforcement of routine environmental cleaning measures. Clin Infect Dis 2006;42:1552e60.

[41] Barnes S, Barron D, Becker L, Canola T, Salemi C. Kaiser Perma- nente national hand hygiene program. Perm J 2004;8:45e50.

[42] Swoboda SM, Earsing K, Strauss K, Lane S, Lipsett PA. Electronic monitoring and voice prompts improve hand hygiene and decrease nosocomial infections in an intermediate care unit. Crit Care Med 2004;32:358e63.

[43] Won SP, Chou HC, Hsieh WS, Chen CY, Huang SM, Tsou KI, et al. Handwashing program for the prevention of nosocomial infections in a neonatal intensive care unit. Infect Control Hosp Epidemiol 2004;25:742e6.

[44] Sharek PJ, Benitz WE, Abel NJ, Freeburn MJ, Mayer ML, Bergman DA. Effect of an evidence-based hand washing policy on hand washing rates and false-positive coagulase negative staph- ylococcus blood and cerebrospinal fluid culture rates in a level III NICU. J Perinatol 2002;22:137e43.

[45] Pittet D, Hugonnet S, Harbarth S, Mourouga P, Sauvan V, Touveneau S, et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet 2000;356:1307e12.

[46] Simmons B, Bryant J, Neiman K, Spencer L, Arheart K. The role of handwashing in prevention of endemic intensive care unit infections. Infect Control Hosp Epidemiol 1990;11:589e94.

[47] Mahida N. Hand hygiene compliance: are we kidding ourselves? J Hosp Infect 2016;92:307e8.

[48] Lambe KA, Lydon S, Madden C, Vellinga A, Hehir A, Walsh M, et al. Hand hygiene compliance in the ICU: a systematic review. Crit Care Med 2019;47:1251e7.

[49] McLaws ML. The relationship between hand hygiene and health care-associated infection: it’s complicated. Infect Drug Resist 2015;8:7e18.

[50] Sickbert-Bennett EE, DiBiase LM, Willis TMS, Wolak ES, Weber DJ, Rutala WA. Reduction of healthcare-associated infections by exceeding high compliance with hand hygiene practices. Emerg Infect Dis 2016;22:1628e30.

[51] Scheithauer S, Lemmen SW. How can compliance with hand hygiene be improved in specialized areas of a university hospital? J Hosp Infect 2013;83:S17e22.

[52] Boyce JM. Current issues in hand hygiene. Am J Infect Control 2019;47(Suppl.):A46e52.

[53] Ward MA, Schweizer ML, Polgreen PM, Gupta K, Reisinger HS, Perencevich EN. Automated and electronically assisted hand hygiene monitoring systems: a systematic review. Am J Infect Control 2014;42:472e8.

V. Mouajou et al. / Journal of Hospital Infection 119 (2022) 33e4848

[54] Cawthorne KR, Cooke RPD. A survey of commercially available electronic hand hygiene monitoring systems and their impact on reducing healthcare-associated infections. J Hosp Infect 2021;111:40e6.

[55] Leis JA, Powis JE, McGeer A, Ricciuto DR, Agnihotri T, Coyle N, et al. Introduction of group electronic monitoring of hand hygiene

on inpatient units: a multicenter cluster randomized quality improvement study. Clin Infect Dis 2020;71:e680e5.

[56] Chen Y-C, Sheng W-H, Wang J-T, Chang S-C, Lin H-C, Tein K-L, et al. Effectiveness and limitations of hand hygiene promotion on decreasing healthcare-associated infections. PLoS One 2011;6:e27163.