ww.sciencedirect.com

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8

Available online at w

Public Health

journal homepage: www.elsevier .com/puhe

Original Research

Impact of the International Nosocomial InfectionControl Consortium (INICC) Multidimensional HandHygiene Approach in five intensive care unitsin three cities of China

D. Su a, B. Hu b, V.D. Rosenthal c,*, R. Li d, C. Hao a, W. Pan a, L. Tao b,X. Gao b, K. Liu d

a The First Affiliated Hospital Shanxi Medical University, Tai Yuan, Chinab Zhongshan Hospital, Fudan University, Shanghai, Chinac International Nosocomial Infection Control Consortium (INICC), Buenos Aires, Argentinad Beijing Chao-Yang Hospital, Capital Medical University, Beijing, China

a r t i c l e i n f o

Article history:

Received 28 January 2014

Received in revised form

3 February 2015

Accepted 18 February 2015

Available online 25 March 2015

Keywords:

Hand hygiene

Multidimensional approach

Intensive care unit

Infection control

Developing countries

International Nosocomial Infection

Control Consortium

* Corresponding author. International NosocAires 1195, Argentina. Tel.: þ54 11 4861 5826

E-mail address: victor_rosenthal@inicc.orURL: http://www.INICC.org

http://dx.doi.org/10.1016/j.puhe.2015.02.0230033-3506/© 2015 The Royal Society for Publ

a b s t r a c t

Objectives: To evaluate the impact of the International Nosocomial Infection Control

Consortium (INICC) Multidimensional Hand Hygiene (HH) Approach in three hospitals in

three cities of China, and analyze predictors of poor hand hygiene compliance.

Study design: A prospective before-after study from May 2009 to December 2010 in five

intensive care units members of the INICC in China.

Methods: The study was divided into two periods: a 3-month baseline period and a follow-

up period. A Multidimensional HH Approach was implemented, which included the

following elements: 1- administrative support, 2- supplies availability, 3- education and

training, 4- reminders in the workplace, 5- process surveillance and 6- performance

feedback. Observations were done for HH compliance in each ICU, during randomly

selected 30-min periods.

Results: A total of 2079 opportunities for HHwere recorded. Overall HH compliance increased

from 51.5% to 80.1% (95%CI 73.2e87.8; P¼ 0.004). Multivariate analysis indicated that several

variables were significantly associated with poor HH compliance: females vs males (64% vs

55%; 95%CI 0.81e0.94; P¼ 0.0005), nurses vs physicians (64%vs 57%, P¼ 0.004), among others.

Conclusions: Adherence to HH was increased significantly with the INICC multidimensional

approach. Specific programs directed to improve HH in variables found to be predictors of

poor HH compliance should be implemented.

© 2015 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.

omial Infection Control C.g (V.D. Rosenthal).

ic Health. Published by E

onsortium (INICC), Corrientes Ave # 4580, Floor 12, Apt D, Buenos

lsevier Ltd. All rights reserved.

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8980

Introduction

The effectiveness of hand hygiene (HH) before patient contact

to prevent cross infection was demonstrated back in the XIX

Century.1 Since then, in the scientific literature it has been

widely confirmed that improved HH practice reduces health

care-associated infection (HAI) rates.2e4

HAIs pose a serious threat to patient safety, causing patient

mortality and morbidity.5 Traditionally, most studies of HAIs

were from developed countries,6e10 and in limited-resource

countries; this problem had not been systematically

addressed until the International Nosocomial Infection Con-

trol Consortium (INICC) started analyzing and publishing HAI

rates using standardized definitions and methods.11e15

In a recent study by Huang Y et al. conducted at a Chinese

teaching hospital, it was reported that, due to deficiencies in

learning resources and curricula, there is limited knowledge

and practice between Chinese medical students regarding

HAIs.16 In particular, the importance of proper HH procedure

to prevent HAI was significantly underestimated, with only

52.9% of the students considering it as the most important

preventive measure for infection control.16 Additionally, it

was reported that 58.5% of students did not wash their hands

between two different procedures on the same patient, and

78.3% did not follow HH practices before and after touching

wounds when they used gloves.16 The authors thus recom-

mended that surveillance andmonitoring of practicing health

care workers (HCWs) were a priority for effective infection

prevention.16

Successful interventions to improve HH have been

reported from high-income countries7e10,17e19 and limited-

resource countries.3,20e27 From the eighties, investigators

have analyzed the effectiveness of interventions to improve

HH.8e10,17e19 In 1997, Larson et al. explicitly referred to a

multidimensional approach that considered several in-

terventions in a study conducted in the US.10 Likewise,

Rosenthal et al. have implemented programs in Argentina

since 1993 combining administrative support, supplies avail-

ability, education and training, process surveillance and per-

formance feedback, which produced a sustained

improvement in HH compliance,22 with a reduction in HAI

rates.3 In 2002, the US Centres for Disease Control and

Prevention (CDC) published their HH guideline.28 In 2005, the

World Health Organization (WHO) launched the program

‘Clean Care is Safer Care’ to promote HH worldwide,29 and

four years later, in 2009, the WHO published its guidelines

including a combination of previously published data, and a

new formulation for alcohol hand rub (AHR) products, among

several other recommendations.4

Adherence to an appropriate hand hygiene method was

described as a critical health problem that remains unsolved

among the Chinese adult population.30 There are no previous

publications showing HH compliance by HCWs in hospitals

from China. The purpose of this INICC study was to establish

the baseline HH compliance rate by HCWs before patient

contact, analyze risk factors for poor adherence, and imple-

ment and evaluate the impact of an INICC Multidimensional

HH Approach (IMHHA) in five intensive care units (ICUs), of

three hospitals, of three cities of China, which includes the

following elements: (1) Administrative support, (2) Supplies

availability, (3) Education and training, (4) Reminders in the

workplace (5) Process surveillance and (6) Performance feed-

back. The resources on the ICU setting is focused, a patient

care area with the highest HAI rates.31 The ICU was the pri-

ority, therefore, because critically ill patients hospitalized in

ICUs are the most susceptible to acquiring severe and life-

threatening HAIs.31

Methods

Background on INICC

The INICC is an international, non-profit, open, multicentric

HAI surveillance network with a methodology based on the

U.S. CDC/National Healthcare Safety Network (NHSN).32 INICC

is the first research network established to measure and

control HAIs worldwide in hospitals through the analysis of

standardized data collected on a voluntary basis by its mem-

ber hospitals. Gaining new members since its international

inception in 2002, INICC comprises a network of nearly 1000

hospitals in 200 cities of 50 countries in Latin America, Asia,

Africa, Middle East, and Europe, and has become the only

source of aggregate standardized international data on the

epidemiology of HAI internationally.15

Study setting

This study was conducted in five intensive care units of three

hospitals of three cities of China, which were successively

incorporated into the study over a period of 2 years. The types

of ICUs participating in this study were one medical surgical,

one neurosurgical, one respiratory and two surgical ICUs.

Each hospital has an infection control team (ICT)

comprised of at least one infection control practitioner (ICP)

and one physician. The ICP is the HCW in charge of process

surveillance in the ICU, who has at least two years of infection

control experience.

Professional categories of HCWs included nurses, physi-

cians, and ancillary staff (including paramedical technicians,

nurse aides, laboratory team members, radiology team

members, physiotherapists, patient care technicians, para-

medical personnel and patient lift teams).

The study protocol was approved by the institutional

review boards at each hospital.

Study design

A prospective, before-after multicentric study was conducted

from May 2009 to December 2010. The study was divided into

two periods: a baseline and a follow-up period. The baseline

period for HH compliance included episodes documented at

each hospital during their first 3 months of participation, and

the follow-up period included episodes following the fourth

month of participation. Each ICU has a different length of

follow-up period, because they started to participate in the

study at different times; but for all ICUs the length of the

baseline period is exactly the same (3 months). For the

comparison of compliance rates, the ICUswere aligned for the

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8 981

amount of study time, independently of the date at which

they started to participate in the study.

INICC Multidimensional HH Approach (IMHHA)

The IMHHA was implemented at each hospital from the

beginning of their participation in INICC. The approach

includes the following six components: 1- Administrative

support; 2- Supplies availability; 3- Education and training; 4-

Reminders in the workplace; 5- Process surveillance; and 6-

Performance feedback. Although the components are

presented individually, they are interactive elements that

must concur for the effective implementation of any ‘multi-

dimensional’ approach.

1- Administrative support

Hospital administrators of the participating hospitals

agreed and committed to the study, attended infection control

meetings to discuss study findings, and allocated supplies of

HH products.

2- Supplies availability

During the study period, AHRs bottles were available at the

ICUs' entrances, nursing stations and near the site of patient

care (individual patient roomentrances, at bedside tables and/

or at the foot of patient beds). Sinks with water supply, soap

and paper towels were available at the ICUs' entrances,

nursing stations, and common areas of ICUs.

3- Education and training

At the study's ICUs, the ICT members provided 30-min

education sessions to HCWs in each work shift, at the begin-

ning of the study period and at regular times periodically

(everymonth, every 2months, and every 6months, depending

on the ICU) during the follow-up period. Education included

information about indications of HH, and the correct

procedures and technique for HH.

4- Reminders in the workplace

Poster reminders were displayed all around the hospital

settings (i.e. hospital entrance, corridors, ICT office, ICU

entrances, nursing stations, beside each sink, and beside each

AHR bottle). They included simple instructions on HH

performance, in line with the contents of the education and

training program.

5- Process surveillance

Process surveillance of HH practices consisted of the

registrations of potential opportunities for HH,4 and the actual

number of HH episodes, either with water and soap or AHR.

HCWs'HHpracticewas directlymonitored by amember of the

ICT, who had received training sessions by means of a

reporting manual, and who was not an observed HCW.4,11 To

improve data inter-reliability, observers used standardized

monitoring processes, following a protocol and completing

standardized HH surveillance forms that contained a uniform

questionnaire for monitoring HH practices.11 The ICTmember

conducted unobtrusive covert observations on health care

professionals working in the ICU; that is, without interference

from the observer and without being aware that they were

being observed. These observations were done at specific time

periods selected at random, distributed three times during a

week, lasting 30 minutes each and during all work shifts

(morning, afternoon and evening.) For this reason, although

HCWs had been notified that process surveillance would be

conducted some time from the beginning of the study, they

were not aware of the schedule of the monitoring period by

the member of the ICT. The monitoring included HH compli-

ance before patient contact, and before an aseptic task,

because by the time the study in May 2009 was started, the

recommendations of the WHO in ‘Your five moments for HH’

had not been disseminated yet. Potential confounders of HH

included type of ICU, professional category, sex, work shift,

and type of contact.

6- Performance feedback

Every month, the INICC Headquarters team prepares and

sends to each participating ICU a final month-by-month

report on compliance with HH. These charts contain a

running tally of HH compliance by HCWs of the ICUs, and

compliance comparing several variables, such as sex,

HCW professional status, ICU type, contact type, and work

shift. Those charts were reviewed at monthly ICT meetings

and also posted in the ICUs to give performance feedback to

the HCWs of the participating ICUs.11 The performance

feedback process started at the third month of

participation.11

Training of the infection control team for processsurveillance

The ICT members investigators were self-trained with a pro-

cedure manual sent from the INICC Headquarters in Buenos

Aires, specifying how to carry out the HH process surveillance

and how to fill in the INICC forms.11 Internal validation of

forms was performed by investigators at the participating

centre to ensure relevant information was accurately recor-

ded for each case. External validation was performed at INICC

Headquarters with consistency analyses of the database to

ensure matching of data entered. Also, ICT members had

continuous telephone, email and webinar access to a support

team at the INICC Headquarters.

Data collection and processing

Completed INICC process surveillance forms of HH were

sent monthly by ICT members from each participating ICU

to the INICC Headquarters. The team at the INICC Head-

quarters uploaded the data into a database, analyzed and

sent to ICT members of each participating ICU a report of HH

compliance, showing HH compliance by month, by sex, by

HCW profession, by ICU, by work shift, and by type of

contact.11

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8982

Statistical methods

Univariate analysis of variables associated with poor handhygiene, and of impact of hand hygiene approachThe aggregated independent variables (sex of HCWs, profes-

sion of HCWs, type of ICU, type of contact, etc.) of all observed

HH opportunities andHH compliance during all the study, and

comparison of HH compliance during the baseline period and

during the follow-up period were compared using Fisher'sexact test for dichotomous variables. Relative risk (RR) ratios

were calculated for comparisons of analyzed variables asso-

ciated with HH using EPI Info V6. 95% confidence intervals (CI)

were calculated using VCStat (Castiglia). P-values <0.05 by

two-tailed tests were considered significant.

The list of variables and their values is as follows:

1. Hand Hygiene compliance: yes, no (dichotomous variable).

2. Sex: male, female (categorical variable)

3. Type of profession: ancillary staff, nursing staff, physician

(categorical variable).

4. Type of contact: non-invasive, invasive (categorical

variable).

5. Type of intensive care unit: adult, paediatric, newborn

(categorical variable).

6. Work shift: morningwork shift, afternoonwork shift; night

work shift (categorical variable).

7. Period of Participation: months 1e3 of participation,

months 4e12 of participation, second year of participation

(categorical variable).

Multivariate analysis of variables associated with poor handhygieneThe aggregated described independent variables of all

observed HH opportunities and HH compliance during all the

study were compared using logistic regression for dichoto-

mous and continuous variables. Due to the fact that ‘Hand

Hygiene compliance’ is a dichotomous variable and the only

dependent variable, a generalized linearmodel such as logistic

regression is used.The logistic regressionmodelwasapplied to

consider the effect of all the independent variables (sex, typeof

profession, typeofcontact, typeof intensivecareunit andwork

shift) inhandhygiene. The regressionequationof themodel is:

In (P/1 � P) ¼ ß0 þ ß1SE þ ß2TP þ ß3TC þ ß4TI þ ß5WS

where P is the probability of hand hygiene, ß0 is the inter-

cept, ß1, ß2 … bm are partial regression coefficients, SE is the

variable for ‘Sex’, TP is the variable for ‘type of professional’,

TC is the variable for ‘Type of contact’, TI is the variable of

‘type of ICU’, and WS is the variable for ‘Work shift’.

Odds ratio (OR) ratios with 95% CI were calculated for

comparisons of analyzed variables associated with HH using

PASW Statistics 18. P-values <0.05 by two-tailed tests were

considered significant.

Multivariate analysis of impact of INICC hand hygienemultidimensional approachHH opportunities and HH compliance during baseline and

during follow-upwere explored for changes in HH compliance

rates following an ICU joining INICC. The follow-up period

stratified by three-month periods over the first year, and

yearly for second year were looked at. The results of a logistic

regression model is presented to consider change in HH

compliance in INICC participating ICUs over time since the

beginning of the HH surveillance.

For the logistical regression over time, the equation is:

In (P/1 � P) ¼ ß0 þ ß1PP þ ß2IC

where P is the probability of hand hygiene, ß0 is the inter-

cept, ß1, ß2 … ßm are partial regression coefficients, PP is the

variable for ‘Period of Participation’, IC is the variable for

‘Participating ICU’.

Odds ratios are presented, comparing each time period

since the start of the surveillance with the baseline of 3

months. This is a large data set, with 2079 observations and so

the authors were able to adjust for the effect of each ICU on

HH compliance as a categorical variable in the analysis.

Because of the different length of follow-up of each ICU (from

6 months to 16 months), for each time period only ICUs with

follow-up in that time period were included in the baseline

period used for calculating the OR of HH compliance for that

period.

Results

From May 2009 through December 2010, a total 2079 oppor-

tunities were recorded for HH before patient contact, and

before aseptic task in five adult ICUs of three academic

teaching hospitals. The baseline period for each ICU was 3

months and their average follow-up period was 8.5 months

(range 7e20). ICU characteristics are shown in Table 1.

Predictors of poor hand hygiene compliance

619 procedures in males, and 1660 in females; 1635 in

nurses, 397 in physicians, and 47 in ancillary staff were

observed; 1633 were prior to non-invasive patient contacts,

and 443 prior to invasive procedures; 2079 in adult ICUs;

1810 during the morning, and 269 during the afternoon.

Table 2 shows HH distribution among the different ICU

types in the baseline and intervention periods.

Tables 3 and 4 show HH compliance according to each

variable (type of hospital, sex, profession of HCW, type of

procedure, type of unit and work shift), and association with

poor HH, analyzed with univariate andmultivariate statistical

methods.

Components of the INICC Multidimensional Hand HygieneApproach

During the follow-up period, the six components of the

IMHHA were applied simultaneously: 66.6% counted on

administrative support and available supplies for HH and

AHR; 100% educated HCWs (33.3% of them every month,

33.3% every 2 months, and 66.6% every 6 months); 100%

posted reminders (100% of them at ICU entrance, 60% in

Table 1 e Characteristics of the participating intensivecare units (May 2009 through December 2010).

Data ICUs, n (%) ICU HHobservations, n

Type of ICU

Medical surgical 1 (20%) 479

Neurosurgical 1 (20%) 465

Respiratory 1 (20%) 413

Surgical 2 (40%) 722

All ICUs 5 (100%) 2079

ICU, intensive care unit; HH, hand hygiene.

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8 983

common ICU areas); process surveillance was conducted by

100%; 100% provided performance feedback (100% of them

every month).

Impact of the INICC Multidimensional Hand HygieneApproach on hand hygiene compliance

Table 5 and Fig. 1 show the results of a logistic regression

model to consider change in HH compliance in INICC

participating ICUs over the whole study period. Overall HH

compliance increased from 51.5% to 80.1% (95% CI 73.2e87.8;

P ¼ 0.004). Multivariate analysis indicated that the following

variables were significantly associated with poor HH

compliance: females vs males (64% vs 55%; 95% CI

Table 2 e Distribution of hand hygiene compliance by type of

ICUs (n) Baseline period(HH compliance/HH observations)

Medical Surgical 1 37% (69/188)

Neurosurgical 1 52% (95/184)

Respiratory 1 55% (91/165)

Surgical 2 64% (111/174)

All 5 51% (366/711)

ICU, intensive care units; CI, confidence interval; RR, relative risk; HH, ha

Table 3 e Hand hygiene compliance by type of variable. Univa

Variable % (# HH/# opportunities

Type of Hospital Academic 62% (1287/2079)

Sex Female 64% (1055/1660)

Male 55% (232/419)

HCW Nurses 64% (1039/1635)

Physicians 57% (225/397)

Ancillary Staff 49% (23/47)

Procedure Non-invasive 62% (1010/1633)

Invasive 62% (275/443)

Work Shift Morning 64% (1153/1810)

Afternoon 50% (134/269)

HCW, health care worker; ICU, intensive care unit; AS, ancillary staff; F, fe

Nb, newborn; M, morning work shift; A, afternoon work shift; N, night w

0.81e0.94; P ¼ 0.0005), and nurses vs physicians (64% vs 57%,

P ¼ 0.004).

Discussion

This is the first study that has showed an improvement in

HH compliance in China due to the implementation of the

IMHHA. The impact of the IMHHA in a diverse ICU popula-

tion from three cities of China, showing that the six mea-

sures of the IMHHA implemented in each ICU were followed

by very substantial improvements in HH practices were

analyzed.

Baseline HH compliance (51.5%) of HCWs at the study ICUs

was similar to that shown in previous studies, whose HH

compliance rates ranged from 9% to 75%.4

The results of the multivariate analysis showed that there

was higher compliance in females, as also identified in in-

dividuals unrelated to health care, such as the findings of

Guinan et al. showing higher compliance by female stu-

dents.33 In a recent study by Tao et al. conducted to describe

patterns of hand washing behaviour among Chinese adults,

female gender was a factor significantly associated with

appropriate hand hygiene practices.30 The influence exerted

by gender in HH compliance should be regarded a strong

factor that contributes to female HCWs' higher internal

motivation for HH and health care, because of an

intensive care unit.

Intervention period(HH compliance/HH observations)

RR (95% CI) P-value

43% (124/291) 1.16

(0.96e1.40)

0.1121

63% (177/281) 1.22

(1.01e1.5)

0.0373

73% (181/248) 1.32

(1.01e1.6)

0.0058

80% (439/548) 1.26

(1.05e1.5)

0.0101

67% (921/1368) 1.31

(1.19e1.43)

0.0001

nd hygiene.

riate analysis.

) Comparison RR 95% CI P. Value

e e e e

F vs M 0.87 0.81e0.94 0.0005

Ns vs Ph 0.9 0.83e0.97 0.004

Ns vs AS 0.8 0.7e0.83 0.0001

Ph vs AS 0.86 0.73e1.0 0.08

NI vs I 0.996 0.86e1.16 0.96

M vs A 0.78 0.73e0.84 0.0001

male; M, male; Ni, non-invasive; I, invasive; Ad, adult; Pe, Paediatric;

ork shift; NS, nursing staff; Ph, physicians; AS, ancillary staff.

Table 4 e Hand hygiene compliance by type of variable.Logistic regression, multivariate analysis.

Variable Adjusted OR 95% CI P-value

Sex (baseline: female) 1.0

Male 0.76 0.6e0.98 0.04

Type of professional

(baseline: nurses)

1.0

Physicians 0.82 0.63e1.1 0.13

Ancillary staff 0.57 0.32e1.0 0.06

Type of contact

(baseline: invasive)

1.0

Non-invasive 0.98 0.8e1.22 0.9

Type of ICU

(baseline: adult)

1.0

Work shift

(baseline: morning)

1.0

Afternoon 0.54 0.42e0.7 0.0001

ICU, intensive care unit; OR, odds ratio; CI, confidence interval.

Table 5 e Hand hygiene improvement by year of participation

Years since joining INICC HHobservations

Number of ICUsincluded

Nu

First 3 months (baseline) 711 5

Months 4e12 1228 5

Second year 140 1

INICC, International Nosocomial Infection Control Consortium; HH, hand

ratio.

Fig. 1 e Hand hygiene improvem

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8984

unquestioned predisposition for females to be the predomi-

nant gender in family health, nursing and related pro-

fessions.34 As stated in the scientific literature, this factor

needs to be fully acknowledged as a useful parameter for

further research and for the design of HH programs.34 In

consonance with this, compliance was higher among nurses,

as also shown in a study by Rosenthal et al., in 2005, in which

compliance was lower among physicians and ancillary staff

compared to nurses.3

This approach included administrative support. In 2003

Rosenthal et al. showed that higher HH adherence was asso-

ciated to administrative support.22 Supplies availability were

also included. In 2000, Bischoff et al.35 showed that easily

accessible dispensers of AHR revealed the more dispensers

per bed, the higher HH compliance. Education and training

were also included, which were other basic independent in-

terventions identified to foster adequate HH performance. As

shown in 1990 by Dubbert et al.,36 educational intervention

with routine classes improved HH compliance by 97% over

.

mber of hospitalsincluded

HH %(95% CI)

Adjusted OR(95% CI)

P-value

3 51.5% (47.7e55.2)

3 65.8% (63.1e68.4) 1.65 (1.36e2.0) 0.0001

1 80.1% (73.2e87.8) 2.44 (1.33e4.5) 0.004

hygiene; ICU, intensive care units; CI, confidence interval; OR, odds

ent by year of participation.

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8 985

four weeks. Likewise, but within the context of limited-

resource countries, Rosenthal et al.22 showed HCWs' educa-tion improved HH adherence, and that compliance increased

further if performance feedback was also implemented. Re-

minders at workplace were also included. In 1989, Conly et al.9

showed the importance of reminders to raise HCWs' aware-

ness of the relation between correct HH performance and HAI

reduction.

2079 opportunities for HH are measured. Every month, the

ICT team provided performance feedback to HCWs of each

ICU. This is a most motivating aspect of the IMHHA for HCWs.

Knowing the outcome of their efforts reflected by the mea-

surement of their practices and HAI incidence can be a most

rewarding or conscious-raising factor to ensure the IMHHA'seffectiveness. From 1998 in Argentina3,22 and 2002 inter-

nationally,11e15 INICC has introduced outcome and process

surveillance and feedback on outcomes and performance,

combined with training and education, as ameans to improve

quality in health care to a new level.3,22 Multidimensional

interventions that focus on standardization of surveillance

has resulted in a sustained improvement in HH.37

Through the last decade, INICC has undertaken a global

effort in America, Asia, Africa, the Middle East, and Europe to

respond to the burden of HAIs, and has achieved extremely

successful results, by increasing HH compliance,

and consequently reducing the rates of HAI andmortality.15,38

This study has some limitations. First, the participating

hospitals are not representative of China. Second, the INICC

did not measure ‘Your five moments for HH’ as advised by

WHO in May 2009, because the IMHHA started in 1998 in

Argentina,3 and in 2002 internationally;11 that is, several

years before the recommendation of WHO was published

and disseminated. This limitation results in limited

comparability. As from 2009, INICC included WHO's rec-

ommendations in its process surveillance forms and man-

uals.4 Third, it should be noted that this is an observational,

before-after, methodology, which implies with less strength

of evidence than other study designs. Direct observation of

adherence typically involves difficulty in assuring inter-

observer reliability, especially given the broad scope of

this research in terms of facilities. Finally, the quality of HH

technique is hard to capture and there might have been a

Hawthorne effect. Nevertheless, as reported in the

scientific literature, the Hawthorne effect may be a useful

tool for sustaining and improving hand hygiene

compliance.39

Conclusions

As demonstrated, the implementation of the IMHHAda

multimodal intervention focused on six inexpensive, sim-

ple, but effective elementsdin China, a limited-resource

country, improved HH compliance in five intensive care

units of three hospitals of three cities of China. Specific

programs directed to improve HH in variables found to be

predictors of poor HH compliance should be implemented. It

is INICC's primary objective to foster infection control

practices, by freely facilitating elemental and inexpensive

resourceful tools to tackle this problem effectively and

systematically, leading to greater and steady adherence to

infection control programs and guidelines, such as HH

compliance, and to the correlated reduction of HAIs and

their consequences, such as mortality and extra cost.

Therefore, the worldwide implementation of the

IMHHA was proposed as a promising path to improving

patient care.

Author statements

Acknowledgements

The authors thank the many health care professionals at

each member hospital who assisted with the conduct of

surveillance in their hospital, including the surveillance

nurses, clinical microbiology laboratory personnel, and the

physicians and nurses providing care for the patients during

the study; without their cooperation and generous assistance

this INICC would not be possible; Mariano Vilar and D�ebora

L�opez Burgardt, who work at INICC headquarters in Buenos

Aires, for their hard work and commitment to achieve INICC

goals; the INICC Country Coordinators and Secretaries (Altaf

Ahmed, Carlos A. �Alvarez-Moreno, Anucha Apisarnthanarak,

Bijie Hu, Namita Jaggi, Hakan Leblebicioglu, Montri Luxsu-

wong, Eduardo A. Medeiros, Yatin Mehta, and Lul Raka,); and

the INICC Advisory Board (Carla J. Alvarado, Nicholas Graves,

William R. Jarvis, Patricia Lynch, Dennis Maki, Gerald

McDonnell, Toshihiro Mitsuda, Cat Murphy, Russell N.

Olmsted, Didier Pittet, William Rutala, Syed Sattar, and Wing

Hong Seto), who have so generously supported this unique

international infection control network.

Ethical approval

Every hospital's Institutional Review Board agreed to the study

protocol, and participants' confidentiality was protected by

codifying the recorded information, making it only identifi-

able to the infection control team.

Funding

The funding for the activities carried out at INICC headquar-

ters were provided by the corresponding author, Victor D.

Rosenthal, and Foundation to Fight against Nosocomial

Infections.

Competing interests

All authors report no conflicts of interest related to this article.

Author contributions

V.D.R. was responsible for study conception and design;

software development; data assembly, analysis, and inter-

pretation; epidemiologic analysis; statistical analysis;

administrative, technical, and logistical support; and drafting

of the manuscript. All authors were involved in provision of

study patients, collection of data, critical revision of the

manuscript for important intellectual content, and final

approval of the manuscript.

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8986

Appendix

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8 987

r e f e r e n c e s

1. Raju TN. Ignac Semmelweis and the etiology of fetal andneonatal sepsis. J Perinatol 1999;19:307e10.

2. Simmons B, Bryant J, Neiman K, Spencer L, Arheart K. Therole of handwashing in prevention of endemic intensive careunit infections. Infect Control Hosp Epidemiol 1990;11:589e94.

3. Rosenthal VD, Guzman S, Safdar N. Reduction in nosocomialinfection with improved hand hygiene in intensive care unitsof a tertiary care hospital in Argentina. Am J Infect Control2005;33:392e7.

4. Pittet D, Allegranzi B, Boyce J. The World Health Organizationguidelines on hand hygiene in health care and theirconsensus recommendations. Infect Control Hosp Epidemiol2009;30:611e22 [the official journal of the Society of HospitalEpidemiologists of America].

5. Jarvis WR. Selected aspects of the socioeconomic impact ofnosocomial infections: morbidity, mortality, cost, andprevention. Infect Control Hosp Epidemiol 1996;17:552e7.

6. Safdar N, Crnich CJ, Maki DG. Nosocomial infections in theintensive care unit associated with invasive medical devices.Curr Infect Dis Rep 2001;3:487e95.

7. Lam BC, Lee J, Lau YL. Hand hygiene practices in a neonatalintensive care unit: a multimodal intervention and impact onnosocomial infection. Pediatrics 2004;114:e565e71.

8. Preston GA, Larson EL, Stamm WE. The effect of privateisolation rooms on patient care practices, colonization andinfection in an intensive care unit. Am J Med 1981;70:641e5.

9. Conly JM, Hill S, Ross J, Lertzman J, Louie TJ. Handwashingpractices in an intensive care unit: the effects of aneducational program and its relationship to infection rates.Am J Infect Control 1989;17:330e9.

10. Larson EL, Bryan JL, Adler LM, Blane C. A multifacetedapproach to changing handwashing behavior. Am J InfectControl 1997;25:3e10.

11. Rosenthal VD, Maki DG, Graves N. The InternationalNosocomial Infection Control Consortium (INICC): goals andobjectives, description of surveillance methods, andoperational activities. Am J Infect Control 2008;36:e1e12.

12. Rosenthal VD, Maki DG, Salomao R, Moreno CA, Mehta Y,Higuera F, Cuellar LE, Arikan OA, Abouqal R, Leblebicioglu H.Device-associated nosocomial infections in 55 intensive careunits of 8 developing countries. Ann Intern Med2006;145:582e91.

13. Rosenthal VD, Maki DG, Mehta A, Alvarez-Moreno C,Leblebicioglu H, Higuera F, Cuellar LE, Madani N, Mitrev Z,Duenas L, Navoa-Ng JA, Garcell HG, Raka L, Hidalgo RF,Medeiros EA, Kanj SS, Abubakar S, Nercelles P, Pratesi RD.International Nosocomial Infection Control Consortiumreport, data summary for 2002e2007, issued January 2008. AmJ Infect Control 2008;36:627e37.

14. Rosenthal VD, Maki DG, Jamulitrat S, Medeiros EA, Todi SK,Gomez DY, Leblebicioglu H, Abu Khader I, MirandaNovales MG, Berba R, Ramirez Wong FM, Barkat A, Pino OR,Duenas L, Mitrev Z, Bijie H, Gurskis V, Kanj SS, Mapp T,Hidalgo RF, Ben Jaballah N, Raka L, Gikas A, Ahmed A, Thule TA, Guzman Siritt ME. International Nosocomial InfectionControl Consortium (INICC) report, data summary for2003e2008, issued June 2009. Am J Infect Control 2010;38. 95-104 e2.

15. Rosenthal VD, Bijie H, Maki DG, Mehta Y, Apisarnthanarak A,Medeiros EA, Leblebicioglu H, Fisher D, Alvarez-Moreno C,Khader IA, Del Rocio Gonzalez Martinez M, Cuellar LE, Navoa-Ng JA, Abouqal R, Guanche Garcell H, Mitrev Z, PirezGarcia MC, Hamdi A, Duenas L, Cancel E, Gurskis V,Rasslan O, Ahmed A, Kanj SS, Ugalde OC, Mapp T, Raka L,Yuet Meng C, Thu le TA, Ghazal S, Gikas A, Narvaez LP,

Mejia N, Hadjieva N, Gamar Elanbya MO, Guzman Siritt ME,Jayatilleke K. International Nosocomial Infection ControlConsortium (INICC) report, data summary of 36 countries, for2004e2009. Am J Infect Control 2012;40:396e407.

16. Huang Y, Xie W, Zeng J, Law F, Ba-Thein W. Limitedknowledge and practice of Chinese medical studentsregarding health-care associated infections. J Infect Dev Ctries2013;7:144e51.

17. Mayer JA, Dubbert PM, Miller M, Burkett PA, Chapman SW.Increasing handwashing in an intensive care unit. InfectControl 1986;7:259e62. IC.

18. Graham M. Frequency and duration of handwashing in anintensive care unit. Am J Infect Control 1990;18:77e81.

19. Dorsey ST, Cydulka RK, Emerman CL. Is handwashingteachable?: failure to improve handwashing behavior in anurban emergency department. Acad Emerg Med 1996;3:360e5[official journal of the Society for Academic EmergencyMedicine].

20. Rosenthal VD, Pawar M, Leblebicioglu H, Navoa-Ng JA,Villamil-Gomez W, Armas-Ruiz A, Cuellar LE, Medeiros EA,Mitrev Z, Gikas A, Yang Y, Ahmed A, Kanj SS, Duenas L,Gurskis V, Mapp T, Guanche-Garcell H, ernandez-Hidalgo R,Kubler A. mpact of the International Nosocomial InfectionControl Consortium (INICC) multidimensional hand hygieneapproach over 13 years in 51 cities of 19 limited-resourcecountries from Latin America, Asia, the Middle East, andEurope. Infect Control Hosp Epidemiol 2013;34:415e23.

21. Allegranzi B, Sax H, Bengaly L, Richet H, Minta DK, ChraitiMN, Sokona FM, Gayet-Ageron A, Bonnabry P, Pittet D.Successful implementation of the World Health Organizationhand hygiene improvement strategy in a referral hospital inMali, Africa. Infect Control Hospital Epidemiol Off J Soc HospitalEpidemiol Am 2010;31:133e41.

22. Rosenthal VD, McCormick RD, Guzman S, Villamayor C,Orellano PW. Effect of education and performance feedbackon handwashing: the benefit of administrative support inArgentinean hospitals. Am J Infect Control 2003;31:85e92.

23. Miranda-Novales MG, Sobreyra-Oropeza M, Rosenthal VD,Higuera F, Armas-Ruiz A, P�erez-Serrato I, Torres-Hern�andez H,Zamudio-Lugo I, Flores-Ruiz EM, Campuzano R, Mena-Brito J,S�anchez-L�opez M, Ch�avez-G�omez A, Rivera-Morales J, Valero-Rodrı́guez JE. Impact of the International Nosocomial InfectionControl Consortium (INICC) Multidimensional Hand HygieneApproach during 3 years in 6 hospitals in 3 cities in Mexico. JPatient Saf; 2014 [in press].

24. Medeiros EA, Grinberg G, Rosenthal VD, Bicudo Angelieri D,Buchner Ferreira I, Bauer Cechinel R, Zanandrea BB,Rohnkohl C, Regalin M, Spessatto JL, Scopel Pasini R, Ferla S.Impact of the International Nosocomial Infection ControlConsortium (INICC) Multidimensional Hand HygieneApproach in 3 cities in Brazil. Am J Infect Control 2014;43:10e5.

25. Leblebicioglu H, Koksal I, Rosenthal VD, Akan OA, OzgultekinA, Kendirli T, Erben N, Nevzat Yalcin A, Ulusoy S, Sirmatel F,Ozdemir D, Alp E, Yıldızdas‚ D, Esen S, Ulger F. Impact of theInternational Nosocomial Infection Control Consortium(INICC) multidimensional hand hygiene approach, over 8years, in 11 cities of Turkey. J Infect Prev 1757177414560249,first published on December 10, 2014 as http://dx.doi.org/10.1177/1757177414560249.

26. Chakravarthy M, Myatra SN, Rosenthal VD, Udwadia FE,Gokul BN,Divatia JV, PoojaryA,SukanyaR,KelkarR,KoppikarG,Pushparaj L, BiswasS, Bhandarkar L, Raut S, Jadhav S, Sampat S,ChavanN,BahiruneS,DurgadS.The impactof the InternationalNosocomial Infection Control Consortium (INICC) multicenter,multidimensional handhygiene approach in two cities of India.J Infect Public Health 2014;8:177e86.

27. Barahona-Guzman N, Rodriguez-Calderon ME, Rosenthal VD,Olarte N, Villamil-Gomez W, Rojas C, Rodriguez-Ferrer M,

p u b l i c h e a l t h 1 2 9 ( 2 0 1 5 ) 9 7 9e9 8 8988

Sarmiento-Villa G, Lagares-Guzman A, Valderrama A,Menco A, Arrieta P, Dajud-Cassas LE, Mendoza M, Sabogal A,Carvajal Y, Silva E. Impact of the International NosocomialInfection Control Consortium (INICC) Multidimensional HandHygiene Approach in three cities of Colombia. Int J Infect Dis2013;19:67e73.

28. Boyce JM, Pittet D. Guideline for hand Hygiene in health-caresettings. Recommendations of the Healthcare InfectionControl Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Society for HealthcareEpidemiology of America/Association for Professionals inInfection Control/Infectious Diseases Society of America.MMWR Recomm Rep 2002;51:1e45. quiz CE1-4, Morbidity andmortality weekly report Recommendations and reports/Centers for Disease Control.

29. WHO launches global patient safety challenge; issuesguidelines on hand hygiene in health care. Indian J Med Sci2005;59:461e3.

30. Tao SY, Cheng YL, Lu Y, Hu YH, Chen DF. Handwashingbehaviour among Chinese adults: a cross-sectional study infive provinces. Public Health 2013;127:620e8.

31. Andrus ML, Horan TC, Gaynes RP. Surveillance of healthcare-associated infections. In: Jarvis WR, editor. Bennett andBrachman's hospital infections. 5th ed. Philadelphia: LipppincottWilliams & Wilkins; 2007. p. 73e89.

32. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillancedefinition of health care-associated infection and criteria forspecific types of infections in the acute care setting. Am JInfect Control 2008;36:309e32.

33. Guinan ME, McGuckin-Guinan M, Sevareid A. Who washeshands after using the bathroom? Am J Infect Control1997;25:424e5.

34. Buvinic M, Medici A, Fernandez E, Torres AC. Genderdifferentials in health; 2006.

35. Bischoff WE, Reynolds TM, Sessler CN, Edmond MB,Wenzel RP. Handwashing compliance by health care workers:the impact of introducing an accessible, alcohol-based handantiseptic. Arch Intern Med 2000;160:1017e21.

36. Dubbert PM, Dolce J, Richter W, Miller M, Chapman SW.Increasing ICU staff handwashing: effects of educationand group feedback. Infect Control Hosp Epidemiol1990;11:191e3.

37. Scheithauer S, Eitner F, Hafner H, Floege J, Lemmen SW. Long-term sustainability of hand hygiene improvements in thehemodialysis setting. Infection;41:675e680.

38. Rosenthal VD, Maki DG, Rodrigues C, Alvarez-Moreno C,Leblebicioglu H, Sobreyra-Oropeza M, Berba R, Madani N,Medeiros EA, Cuellar LE, Mitrev Z, Duenas L, Guanche-Garcell H, Mapp T, Kanj SS, Fernandez-Hidalgo R. Impact ofInternational Nosocomial Infection Control Consortium(INICC) strategy on central line-associated bloodstreaminfection rates in the intensive care units of 15 developingcountries. Infect Control Hosp Epidemiol 2010;31:1264e72.

39. Kohli E, Ptak J, Smith R, Taylor E, Talbot EA, Kirkland KB.Variability in the Hawthorne effect with regard to handhygiene performance in high- and low-performing inpatientcare units. Infect Control Hosp Epidemiol 2009;30:222e5.