hand washing

Multicenter Intervention Program to Increase Adherence to Hand Hygiene Recommendationsand Glove Use and to Reduce the Incidence of Antimicrobial ResistanceAuthor(s): WilliamE.Trick , MD; MichaelO.Vernon , DrPH; SharonF.Welbel , MD;PatriciaDeMarais , MD; MaryK.Hayden , MD; RobertA.Weinstein , MD; ChicagoAntimicrobial Resistance ProjectSource: Infection Control and Hospital Epidemiology, Vol. 28, No. 1 (January 2007), pp. 42-49Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiologyof AmericaStable URL: http://www.jstor.org/stable/10.1086/510809 .Accessed: 05/03/2014 22:27

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaboratingwith JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 140.103.40.237 on Wed, 5 Mar 2014 22:27:22 PMAll use subject to JSTOR Terms and Conditions

infection control and hospital epidemiology january 2007, vol. 28, no. 1

o r i g i n a l a r t i c l e

Multicenter Intervention Program to Increase Adherenceto Hand Hygiene Recommendations and Glove Use and to Reduce

the Incidence of Antimicrobial Resistance

William E. Trick, MD; Michael O. Vernon, DrPH; Sharon F. Welbel, MD; Patricia DeMarais, MD;Mary K. Hayden, MD; Robert A. Weinstein, MD ; for the Chicago Antimicrobial Resistance Project

objective. To determine whether a multimodal intervention could improve adherence to hand hygiene and glove use recommendationsand decrease the incidence of antimicrobial resistance in different types of healthcare facilities.

design. Prospective, observational study performed from October 1, 1999, through December 31, 2002. We monitored adherence tohand hygiene and glove use recommendations and the incidence of antimicrobial-resistant bacteria among isolates from clinical cultures.We evaluated trends in and predictors for adherence and preferential use of alcohol-based hand rubs, using multivariable analyses.

setting. Three intervention hospitals (a 660-bed acute and long-term care hospital, a 120-bed community hospital, and a 600-bedpublic teaching hospital) and a control hospital (a 700-bed university teaching hospital).

intervention. At the intervention hospitals, we introduced or increased the availability of alcohol-based hand rub, initiated an interactiveeducation program, and developed a poster campaign; at the control hospital, we only increased the availability of alcohol-based hand rub.

results. We observed 6,948 hand hygiene opportunities. The frequency of hand hygiene performance or glove use significantly increasedduring the study period at the intervention hospitals but not at the control hospital; the maximum quarterly frequency of hand hygieneperformance or glove use at intervention hospitals (74%, 80%, and 77%) was higher than that at the control hospital (59%). By multivariableanalysis, preferential use of alcohol-based hand rubs rather than soap and water for hand hygiene was more likely among workers atintervention hospitals compared with nonintervention hospitals (adjusted odds ratio, 4.6 [95% confidence interval, 3.3-6.4]) and morelikely among physicians (adjusted odds ratio, 1.4 [95% confidence interval, 1.2-1.8]) than among nurses at intervention hospitals. Asignificantly reduced incidence of antimicrobial-resistant bacteria among isolates from clinical culture was found at a single interventionhospital, which had the greatest increase in the frequency of hand hygiene performance.

conclusions. During a 3-year period, a multimodal intervention program increased adherence to hand hygiene recommendations,especially to the use of alcohol-based hand rubs. In one hospital, a concomitant reduction was found in the incidence of antimicrobial-resistant bacteria among isolates from clinical cultures.

Infect Control Hosp Epidemiol 2007; 28:42-49

From the John H. Stroger Jr. Hospital of Cook County (W.E.T., S.F.W., R.A.W.) and the Rush Medical College (S.F.W., M.K.H., R.A.W.), Chicago, andthe Cook County Department of Public Health (M.O.V.) and Oak Forest Hospital of Cook County (P.D.), Oak Park, Illinois.

Received March 11, 2006; accepted May 1, 2006; electronically published January 3, 2007. 2006 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2007/2801-0008$15.00.

Infection caused by antimicrobial-resistant bacteria is com-mon at many US hospitals.1 Despite evidence that hand an-tisepsis can reduce the frequency of pathogen transmis-sion,2-5 studies have consistently shown that hand hygieneperformance by healthcare workers (HCWs) is woefully in-frequent.6-10 To overcome barriers to hand hygiene reportedby HCWs, the Centers for Disease Control and Prevention(CDC) has promoted the use of alcohol-based hand rubs.11

Compared with soap and water, alcohol-based hand rubs aremore efficient,12 effective,13 and less irritating to skin.14,15

These relative benefits have proved to be compelling in suc-cessful efforts to promote the use of alcohol-based handrub.2,16-18

Because many interventions have not been shown to sus-tain improvements in adherence to hand hygiene recom-mendations when implemented alone, introduction of mul-timodal, bundled interventions has been recommended.2,4,19-23

We designed a multimodal intervention program aimed atincreasing the frequency of hand hygiene performance andglove use among HCWs and decreasing the incidence of re-covery of antimicrobial-resistant bacteria from clinical culture(hereafter, incidence of antimicrobial-resistant bacteria).We implemented the entire intervention program at 3 of 4hospitals that participated in the Chicago Antimicrobial Re-sistance Project and promoted a single interventionuse ofalcohol-based hand rubat the fourth hospital.


intervention to improve hand hygiene adherence and reduce drug resistance 43

methods

Settings

The 4 hospitals comprised a 660-bed acute and long-termcare hospital (hospital A), a 120-bed community hospital(hospital B), a 600-bed public teaching hospital (hospital C),and a 700-bed university teaching hospital (hospital D). FromOctober 1, 1999, through December 31, 2002, we observedhand hygiene performance and glove use by HCWs pro-spectively in 3 units at each hospital. The units representeddifferent intensities in patient care. HCWs were not informedabout the observations. We observed HCWs in 7 intensivecare units (ICUs) (1 in hospital A, 2 in hospital B, 2 in hospitalC, and 2 in hospital D), 3 medical and/or surgical units(hospitals A, B, and C), 1 skilled care unit (hospital A), and1 rehabilitation unit (hospital D). At hospital D, after 4 quar-ters we changed 1 observation unit from the surgical ICU tothe neonatal ICU. Because we did not monitor hand hygieneperformance and glove use in both of these units throughoutthe study period, we performed 2 trend analyses: one analysisincluded data from the surgical ICU and the neonatal ICU,and the other analysis excluded data from both these units.Although the rate of hand hygiene performance or glove useamong HCWs in the neonatal ICU was slightly higher thanthat among HCWs in the surgical ICU, the difference didnot substantially or significantly change the results from thetrend analyses; we report results of the analyses that includeall observations.

Observation of Hand Hygiene Performance and Glove Use

Study personnel, not infection control staff, conducted at least4 hours of observation in each unit per month (ie, 12 hoursper month for each hospital). Observation lasted 60 minutesand was performed during all work shifts. To standardizeobservations, 3 observers were trained by the same person;we required 80% agreement among the observers aboutwhether hand hygiene had occurred for an entire observationperiod before unsupervised observations were allowed. Thetraining included tours of the observation units and discus-sions about each data element. We considered each HCW-patient encounter as a single opportunity for hand hygiene(ie, we did not record all hand hygiene indications duringthese encounters). An encounter included HCW contact withany environmental surface in the patients room. We cate-gorized each opportunity by whether the HCW touched thepatient or an environmental surface only. Because we wereconcerned that multiple observations of a single HCW duringan observation period could influence adherence and violatestatistical assumptions of independence, we allowed only asingle observation of a HCW during an observation period.

In addition to observing the frequency of hand hygieneperformance, we recorded HCW type (ie, physician, nurse,or other), type of contact (ie, patient or environment only),use of gloves during contact and removal of gloves after con-

tact, presence of isolation precautions, and agent used (ie,soap and water or alcohol-based hand rub). To control forincreased awareness of the observer by HCWs being observed,we also recorded the order in which HCWs were observed(eg, first HCW observed, second HCW observed, third HCWobserved) during each observation period. We initiated ob-servations at the study hospitals from October through De-cember 1999, and we began implementing interventions inthe first quarter of 2000.

Educational Intervention and Introductionof Alcohol-Based Hand Rub

The content of our educational program was guided by re-sponses to a survey we distributed during infection controlsessions at the intervention hospitals.24 The survey queriedparticipants about their understanding of pathogen trans-mission, opinions about hand hygiene practices, and per-ceived barriers to hand hygiene (eg, skin irritation and in-adequate provision of hand hygiene infrastructure). The goalof the educational sessions was to increase HCW awarenessof the value of alcohol-based hand rubs; we also stressed theimportance of wearing gloves to reduce hand contamination.The slides from the presentation were adapted by the CDCfor distribution to HCWs or infection control professionalsinterested in improving hand hygiene at their healthcarefacility.25

Starting in March 2000, we delivered 45-minute educa-tional sessions to personnel at hospitals A, B, and C duringtheir annual mandatory infection control education pro-grams. To engage audiences, we used an interactive audience-response system, which allowed participants to respond anon-ymously to questions posed by the speaker and to comparetheir responses to those of the group. Components of thepresentation included review of the CDCs Healthcare Infec-tion Control Practices Advisory Committees hand hygieneguidelines11 and hospital-specific policies, current options forhand antisepsis, data on hospital-specific hand hygiene ad-herence, benefits of using alcohol-based hand rubs (eg, ef-ficiency, efficacy, and improved skin condition), and value ofwearing examination gloves. After the presentation, we dis-tributed pocket-sized bottles of alcohol-based hand rub anda hand hygiene fact sheet. We revised the presentation eachyear; for example, after we noted poor hand hygiene practicesby HCWs who touched an environmental surface only, westressed the importance of hand hygiene in these situations.We continued to emphasize the benefits of alcohol-basedhand rub use and glove use during patient care.

From April through August 2000, we ensured that alcohol-based hand rub (formulated with 62% ethyl alcohol) wasreadily available in all inpatient care areas of all 4 facilities.Wall-mounted hand rub dispensers were installed at or nearthe entrance to each patients room. We solicited HCW inputregarding the choice of alcohol-based hand rub.


44 infection control and hospital epidemiology january 2007, vol. 28, no. 1

table 1. Adherence to Hand Hygiene Recommendations Among HealthcareWorkers (HCWs) at 4 Hospitals

Variable

No. ofopportunities(n p 6,948)

Overalladherence, %

Rate ratio(95% CI)

Hospitala

Control, hospital D 1,742 30 1.0Intervention

Hospital A 1,365 42 1.4 (1.3-1.6)Hospital C 2,263 42 1.4 (1.3-1.5)Hospital B 1,578 49 1.6 (1.5-1.8)

HCWNurse 5,221 42 1.0Physician 1,380 39 0.9 (0.9-1.0)Other 347 20 0.5 (0.4-0.6)

Patient contact during encounterNo, environment only 2,201 26 1.0Unknown 1,492 41 1.6 (1.4-1.7)Yes 3,255 50 1.9 (1.8-2.1)

Time of day7:00 AM to 11:00 AM 1,900 37 1.011:01 AM to 3:00 PM 3,111 41 1.1 (1.0-1.2)3:01 PM to 7:00 PM 1,820 42 1.1 (1.0-1.2)7:01 PM to 6:59 AM 117 38 1.0 (0.8-1.3)

Glove useNo 3,442 28 1.0Yes 3,506 52 1.9 (1.7-2.0)

Contact isolation precautionsNo 6,354 40 1.0Yes 594 45 1.1 (1.0-1.2)

Observation sequenceFirst or second 2,370 38 1.0Third or fourth 2,265 42 1.1 (1.0-1.2)Fifth or later 2,313 41 1.1 (1.0-1.2)

note. Hand hygiene is defined as the use of alcohol-based hand rub or use of soap andwater. CI, confidence interval.a Hospital A is a 660-bed acute and long-term care hospital, hospital B is a 120-bed communityhospital, hospital C is a 600-bed public hospital, and hospital D is a 700-bed university teachinghospital.

Poster Campaign

Between annual educational presentations, we used hospital-wide poster campaigns at the intervention hospitals. The cam-paign featured humorous posters of high-profile hospital ad-ministrative and clinical staff using and encouraging HCWsto use alcohol-based hand rubs. During brief (1015-minute)ward-based academic detailing sessions, a promotional hand-out that contained information on hand care, hand hygiene,benefits of wearing examination gloves, and hospital-specificrates of hand hygiene performance was discussed with HCWs.

Surveillance for Antimicrobial-Resistant Bacteria

From July 1, 1999, through December 31, 2002, we evaluatedthe incidence of recovery of the following antimicrobial-resistant bacteria from clinical culture: methicillin-resistantStaphylococcus aureus (MRSA), vancomycin-resistant entero-

cocci, third-generation cephalosporin-resistant Escherichiacoli or Klebsiella species, fluoroquinolone-resistant Escherichiacoli, and Acinetobacter species or Pseudomonas aeruginosa re-sistant to either imipenem or amikacin. Our analysis excludedoutpatient cultures and cultures performed only for screeningpurposes. We did not record duplicate isolates, defined asisolates with the same resistance phenotype and isolates re-covered from the same patient within a 30-day period. Wecalculated the overall and organism-specific incidence as thenumber of antimicrobial-resistant isolates recovered per 1,000patient-days and report the results aggregated by quarter.

At the intervention hospitals, surveillance for antimicro-bial-resistant isolates was performed by infection control pro-fessionals; they recorded their findings on forms that weresubsequently scanned into a centralized data warehouse.26 Theinfection control professionals categorized isolates as either



figure 1. Trends in frequency of hand hygiene performance after patient or environmental contact or glove use without hand hygieneperformance, by quarter (Q). During the study period, the frequency of hand hygiene performance or glove use without hand hygieneincreased significantly at the intervention hospitals (hospitals A, B, and C; for each hospital) but not at the control hospital (hospitalP ! .001D; ). Analysis of hand hygiene adherence alone revealed significant increases at hospitals A ( ) and B ( ) but notPp .15 Pp .002 Pp .02at hospitals C ( ) and D ( ).Pp .27 Pp .4

community acquired or hospital acquired, using CDC defi-nitions.27 Comparable data from the control hospital werenot collected.

Statistical Analysis

We calculated unit-specific frequencies of hand hygiene per-formance and glove use and aggregated the results by quarter.For each hospital, we determined whether there was a sig-nificant departure from a linear trend of zero by modelinghand hygiene as the dependent variable and quarter as theexplanatory variable. To adjust for unit-specific effects, weincluded terms for each unit in our model. We performedsimilar analyses to assess the trend in hand hygiene practiceaccording to type of contact, agent used, and whether handhygiene was performed or gloves were used.

We evaluated HCW and facility characteristics by meansof bivariable analyses to determine which characteristics wereassociated with performance of hand hygiene or use of gloves.We created indicator variables for each hospital and for hos-pital type (ie, intervention versus control), HCW type, ob-servation sequence (dichotomized as first or second HCWobserved vs other), and type of contact (ie, environment only,patient, or unknown). We identified factors associated withhand hygiene performance or glove use and factors associatedwith preference for alcohol-based hand rub by constructingseparate logistic regression models. We aggregated the inci-dence of infection with antimicrobial-resistant bacteria quar-terly and used linear regression methods to evaluate whetherthe time-dependent trend was significantly different from

zero. Statistical analyses were performed using Stata statisticalsoftware, version 9.0 (Stata).

results

Hand Hygiene Performance and Glove Use

We observed 6,948 potential hand hygiene opportunities dur-ing 1,353 observation sessions. Most observations were ofnurses, followed by physicians and other HCWs (Table 1).Adherence was better at intervention hospitals, after use ofgloves during contact, and after patient contact (Table 1).Small increases in adherence were recorded for contact withpatients under contact isolation precautions, during the mid-dle of the day, and among HCWs observed later during theobservation period (Table 1).

The frequency of hand hygiene performance by quarterincreased significantly during the study period at hospitals Aand B but not at hospital C: at hospital A, the frequencyincreased from 23% to 46% of hand hygiene opportunities( ); at hospital B, from 30% to 50% ( ); andPp .002 Pp .02at hospital C, from 35% to 43% ( ). The frequencyPp .27of hand hygiene performance at the control hospital did notchange during the study period (from 32% to 31% of op-portunities; ) (Figure 1).Pp .40

The increase in the frequency of hand hygiene performanceafter patient contact was greatest at hospital A (from 32% to56% of opportunities; ). Increases in the frequencyPp .06of hand hygiene practice were more significant after contactwith an environmental surface only: at hospital A, the fre-



table 2. Multivariable Analysis of Characteristics Associated With Performance of HandHygiene and Use of Alcohol-Based Hand Rub for Hand Hygiene

Variable

Performance ofhand hygienea

Use of alcohol-basedhand rubb

Adjusted OR(95% CI) P

Adjusted OR(95% CI) P

Hospitalc

Control (hospital D) Reference ReferenceIntervention (hospitals A, B, and C) 1.7 (1.5-1.9) !.001 4.6 (3.3-6.4) !.001

Glove useNo Reference ReferenceYes 2.0 (1.7-2.2) !.001 0.8 (0.6-1.0) .03

Patient contact during encounterNo, environment only Reference ReferenceUnknown 1.6 (1.4-1.9) !.001 1.0 (0.8-1.3) .82Yes 2.1 (1.8-2.4) !.001 0.7 (0.5-0.9) .001

Healthcare workerNurse Reference ReferencePhysician 0.7 (0.8-1.0) .01 1.4 (1.2-1.8) .001Other 0.3 (0.2-0.4) !.001 1.4 (0.8-2.3) .28

note. CI, confidence interval; OR, odds ratio.a A total of 6,948 hand hygiene opportunities were included in the final model. Hand hygiene involvedthe use of either alcohol-based hand rub or soap and water. The model adjusted for the order in whicha healthcare worker was observed during a given observation period: compared with healthcare workerswho were observed first or second, healthcare workers who were observed later were associated withimproved hand hygiene adherence (OR, 1.2 [95% CI, 1.1-1.3]).b A total of 2,803 hand hygiene episodes were included in the final model, which analyzed the choiceof alcohol-based hand rub versus the choice of soap and water for hand hygiene.c Hospital A is a 660-bed acute and long-term care hospital, hospital B is a 120-bed community hospital,hospital C is a 600-bed public hospital, and hospital D is a 700-bed university teaching hospital.

quency increased from 0% to 40% ( ); at hospital B,P ! .001from 0% to 51% ( ); and at hospital C, from 5% toP ! .00138% ( ).Pp .02

To assess overall protection from microbial contaminationof HCWs hands, we tallied the frequency of hand hygienepractice after patient care or glove use during care. Handhygiene performance or glove use increased significantly atall 3 intervention hospitals (quarterly maximum by hospital:74% of opportunities at hospital A, 80% at hospital B, and77% at hospital C) but not at the control hospital (59%)(Figure 1). During the study period, HCWs at interventionhospitals were more likely to protect themselves from mi-crobial contamination by wearing gloves or performing handhygiene (67% of opportunities at hospital A, 74% at hospitalB, and 67% at hospital C), compared with HCWs at the con-trol hospital (50%).

The frequency of alcohol-based hand rub use increased sig-nificantly at hospitals A, C, and D ( ); the increase inP ! .001the frequency of use at the community hospital (hospital B)initially was marked, but use then leveled off ( ) (FigurePp .081). In contrast, the frequency of soap and water use was stableat all facilities except hospital C, where the frequency signifi-cantly decreased during the study period ( ).Pp .005

By multivariable analysis, hand hygiene performance wasmore likely among nurses, at intervention hospitals, afterglove use, after patient contact, and later during the obser-

vation period (Table 2). Choice of alcohol-based hand rubrather than soap and water was more likely at an interventionhospital and among physicians compared with nurses; HCWspreferentially used soap and water after a patients skin wascontacted and after gloves were worn (Table 2). Time of dayand observation sequence had no effect on the choice ofalcohol-based hand rub over soap and water for hand hygiene.

Antimicrobial-Resistant Clinical Isolates

At hospital A, a significant decrease in the overall incidenceof hospital-acquired, antimicrobial-resistant bacteria occurredduring the study period, despite a nonsignificant increase inthe incidence of resistance among community-acquired bac-teria (Figure 2). A decreased incidence of antimicrobial resis-tance was evident for the following organisms: third-generationcephalosporinresistant E. coli ( ), fluoroquinolone-Pp .007resistant E. coli ( ), imipenem-resistant P. aeruginosaP ! .001( ), and MRSA ( ). In contrast, the incidencePp .003 Pp .08of resistance among community-acquired organisms (definedas antimicrobial-resistant organisms recovered from culture ofa specimen obtained within 48 after admission or transfer tohospital A) increased for MRSA ( ) and imipenem-Pp .002resistant P. aeruginosa ( ).Pp .03

No decreases in the incidence of antimicrobial-resistantbacteria were detected at hospitals B and C.



figure 2. Trend in the incidence of antimicrobial-resistant organisms among isolates recovered from clinical culture and trend in thepercentage of hand hygiene opportunities for which healthcare workers (HCWs) used either soap and water or alcohol-based hand rubafter patient contract or used gloves during patient contact in hospital A. During the study period, we observed a significant increase inthe frequency of hand hygiene or glove use ( ), a significant decrease in the incidence of antimicrobial resistance among hospital-P ! .001acquired bacteria ( ), and a nonsignificant increase in the incidence of antimicrobial resistance among community-acquired bacteriaP ! .001( ). See Methods for antimicrobial-resistant organisms for which surveillance was performed. Q, quarter.Pp .12

discussion

We observed a sustained increase in the frequency of handhygiene practice during 3 years at 2 of 3 hospitals that im-plemented a multimodal intervention program. The increasedhand hygiene adherence was primarily the result of increaseduse of alcohol-based hand rubs, especially among physicians.Despite the increased availability of alcohol-based hand rub,no increase in the frequency of hand hygiene practice wasobserved at the hospital that did not implement the inter-vention program. At the hospital that had the greatest increasein the frequency of hand hygiene practice, there was a con-comitant reduction in the incidence of several antimicrobial-resistant bacteria.

Although sustained improvements in hand hygiene havebeen reported by other investigators,2,4,18,21 our study adds totheir findings by documenting improved adherence to handhygiene recommendations at multiple institutions across thespectrum of health care during a 3-year period. Although theoverall increases may appear modest, the following factorsneed to be considered when interpreting our results: we in-cluded any contact with an environmental surface as a handhygiene opportunity, we required that hand hygiene occurbefore contacting any object outside the patients room, andwe did not include the significantly higher frequencies of handhygiene performance recorded by the infection control de-partment at the 2 intervention hospitals that submitted thesedata (these frequencies were 29% and 57% higher than thoserecorded by study personnel; data not shown). In additionto improved compliance with hand hygiene, we documentedfrequent use of examination gloves, which is an adjunct tohand cleansing.

Because isolated interventions, such as education or per-

formance feedback, have only been successful for short pe-riods, we modeled our program on recommendations thatemphasized the importance of a multimodal approach.23 Inaddition, to improve retention of our educational message,we developed a program that facilitated audience partic-ipation through use of handheld keypads; the program wasadapted by the CDC and is available for distribution.25 Ofthe 4 study hospitals, 3 incorporated the multimodal inter-vention; the fourth hospital increased access to alcohol-basedhand rubs without incorporating the other program features,and no increase in the frequency of hand hygiene practiceoccurred.

Use of alcohol-based hand rub supplemented, rather thansupplanted, use of soap and water. Alcohol-based hand rubswere better accepted by physicians than by nurses, which isimportant because historically physicians have been less ad-herent to hand hygiene recommendations. When we surveyedHCWs,24 we found that nurses rather than physicians weremore concerned about skin damage caused by alcohol-basedhand rubs. Despite addressing this concern in our presen-tations, we had less success changing nurses behavior. Inaddition to increased use of alcohol-based hand rubs afterpatient care, overall hand hygiene was improved by increaseduse of gloves during patient care.

Our greatest success at increasing hand hygiene practicewas at the acute and long-term care facility (hospital A) andcommunity hospital (hospital B). Although use of alcohol-based hand rubs increased at the large public teaching hospital(hospital C), the increase in the rate of hand hygiene per-formance was not significant. The relative success at hospitalsA and B may be due to the stability of their work forces:because they had few house staff members or students and



few or no physicians in private practice, it was easier to ed-ucate HCWs with episodic educational sessions. To achievesignificant improvements in hand hygiene adherence at teach-ing hospitals, we would need to ensure good penetration ofour educational efforts to rotating students, house staff mem-bers, and attending physicians.

Coincident with the largest increase in the frequency ofhand hygiene practice, hospital A had significant decreasesin the overall incidence of antimicrobial-resistant bacteriarecovered from clinical culture. This decrease occurred de-spite a stable or increasing incidence of resistance amongcommunity-acquired bacteria. Because of the observationalstudy design, there was no randomized control group; ourfindings must therefore be interpreted with caution. Althoughthe decrease in antimicrobial-resistant organisms at hospitalA may have been the result of reduced hand contamination,other factors may have contributed to the decrease. For ex-ample, there may have been unmeasured changes in thepatient population, and there was a separate intervention(initiated in November 2000) to improve antimicrobial pre-scribing practices at hospital A28; however, the decrease inantimicrobial-resistant organisms preceded the interventionto improve prescribing practices. Because the frequency ofculture performance at hospital A was unchanged during thestudy (data not shown), the decrease in antimicrobial-resis-tant organisms was unlikely to be the result of changes inthe frequency of testing. Our study adds to the seminal workof Pittet et al.2 and to the growing body of evidence thatincreased frequency of hand hygiene performance, especiallyby means of alcohol-based hand rubs, is associated with adecreased incidence of antimicrobial-resistant bacteria2-4,29

and of hospital-acquired infection.5,21 Incorporation of otherinterventions into a multimodal strategy, such as active sur-veillance culture,30,31 environmental decontamination,32 andpatient decolonization,33 could provide an additional level ofcontrol of the spread of antimicrobial resistance.

Theoretical models suggest that, to interrupt transmissionof antimicrobial-resistant organisms in settings where colo-nization pressure is high, hand hygiene needs to occur after60%-80% of patient contacts34; however, the frequent use ofgloves can notably reduce HCW hand carriage of vanco-mycin-resistant enterococci35 and may have contributed tothe decreased incidence of antimicrobial-resistant bacteria.

Our study has several limitations. First, we selected ob-servation units from our study hospitals that represented bothICUs and non-ICUs and allowed for relatively inconspicuousobservations of hand hygiene behavior; we did not matchunits by type of care delivered. Second, we did not controlfor workload by monitoring the ratio of nurses to patients.Third, our observations of hand hygiene performance, gloveuse, and antimicrobial resistance rates suffer from the limi-tations of ecological studies in which associations are notmade at the level of the individual. Fourth, we performed aprospective, observational study because our staffing patternsprecluded randomizing the intervention by unit or HCW

type; however, we included a control hospital that introducedalcohol-based hand rub but did not implement the inter-vention; we also observed hand hygiene behavior over severalyears to evaluate the sustainability of our intervention. Finally,observer bias is also a possible limitation. In fact, we foundthat the rate of hand hygiene practice improved later in theobservation period, especially outside ICUs (data not shown);however, our findings were unchanged when we controlledfor observation sequence.

In conclusion, we found that implementation of a multi-modal intervention program resulted in a sustained increasein adherence to hand hygiene recommendations and use ofalcohol-based hand rub. The provision of an alcohol-basedhand rub in the absence of an aggressive educational inter-vention was insufficient to improve hand hygiene rates. Atone hospital, the increased rate of hand hygiene practice wasassociated with a decreased incidence of antimicrobial-resis-tant bacteria recovered from clinical culture.

Address reprint requests to William E. Trick, MD, Collaborative ResearchUnit, Department of Medicine, Stroger Hospital of Cook County, 1900 WestPolk Street, Suite 1600, Chicago, IL 60612 ([email protected]).

acknowledgments

We thank participants from the Chicago Antimicrobial Resistance Project;Stroger, Oak Forest, and Provident hospitals; Rush University Medical Center;and the CDC. We thank Ellen Holfels for help with editing figures andmanuscript preparation.

This work was funded by the Centers for Disease Control and Prevention(cooperative agreement U50/CCU515853-03).

references

1. Fridkin SK, Steward CD, Edwards JR, et al. Surveillance of antimicrobialuse and antimicrobial resistance in United States hospitals: project ICAREphase 2. Project Intensive Care Antimicrobial Resistance Epidemiology(ICARE) hospitals. Clin Infect Dis 1999; 29:245-252.

2. Pittet D, Hugonnet S, Harbarth S, et al. Effectiveness of a hospital-wideprogramme to improve compliance with hand hygiene: Infection ControlProgramme. Lancet 2000; 356:1307-1312.

3. MacDonald A, Dinah F, MacKenzie D, Wilson A. Performance feedbackof hand hygiene, using alcohol gel as the skin decontaminant, reducesthe number of inpatients newly affected by MRSA and antibiotic costs.J Hosp Infect 2004; 56:56-63.

4. Larson EL, Early E, Cloonan P, Sugrue S, Parides M. An organizationalclimate intervention associated with increased handwashing and de-creased nosocomial infections. Behav Med 2000; 26:14-22.

5. Rosenthal VD, Guzman S, Safdar N. Reduction in nosocomial infectionwith improved hand hygiene in intensive care units of a tertiary carehospital in Argentina. Am J Infect Control 2005; 33:392-397.

6. Bischoff WE, Reynolds TM, Sessler CN, Edmond MB, Wenzel RP. Hand-washing compliance by health care workers: the impact of introducingan accessible, alcohol-based hand antiseptic. Arch Intern Med 2000; 160:1017-1021.

7. Pittet D, Simon A, Hugonnet S, Pessoa-Silva CL, Sauvan V, PernegerTV. Hand hygiene among physicians: performance, beliefs, and percep-tions. Ann Intern Med 2004; 141:1-8.

8. Pittet D, Mourouga P, Perneger TV. Compliance with handwashing ina teaching hospital: Infection Control Programme. Ann Intern Med 1999;130:126-130.



9. Lankford MG, Zembower TR, Trick WE, Hacek DM, Noskin GA, Pe-terson LR. Influence of role models and hospital design on hand hygieneof healthcare workers. Emerg Infect Dis 2003; 9:217-223.

10. Larson E, Kretzer EK. Compliance with handwashing and barrier pre-cautions. J Hosp Infect 1995; 30:S88-S106.

11. Boyce JM, Pittet D. Guideline for Hand Hygiene in Health-Care Settings:recommendations of the Healthcare Infection Control Practices AdvisoryCommittee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene TaskForce. Infect Control Hosp Epidemiol 2002; 23:S3-S40.

12. Voss A, Widmer AF. No time for handwashing!? Handwashing versusalcoholic rub: can we afford 100% compliance? Infect Control Hosp Ep-idemiol 1997; 18:205-208.

13. Trick WE, Vernon MO, Hayes RA, et al. Impact of ring wearing on handcontamination and comparison of hand hygiene agents in a hospital.Clin Infect Dis 2003; 36:1383-1390.

14. Boyce JM, Kelliher S, Vallande N. Skin irritation and dryness associatedwith two hand-hygiene regimens: soap-and-water hand washing versushand antisepsis with an alcohol hand gel. Infect Control Hosp Epidemiol2000; 21:442-448.

15. Larson EL, Aiello AE, Bastyr J, et al. Assessment of two hand hygieneregimens for intensive care unit personnel. Crit Care Med 2001; 29:944-951.

16. Girard R, Amazian K, Fabry J. Better compliance and better tolerancein relation to a well-conducted introduction to rub-in hand disinfection.J Hosp Infect 2001; 47:131-137.

17. Harbarth S, Pittet D, Grady L, et al. Interventional study to evaluate theimpact of an alcohol-based hand gel in improving hand hygiene com-pliance. Pediatr Infect Dis J 2002; 21:489-495.

18. Hugonnet S, Perneger TV, Pittet D. Alcohol-based handrub improvescompliance with hand hygiene in intensive care units. Arch Intern Med2002; 162:1037-1043.

19. Larson EL, Bryan JL, Adler LM, Blane C. A multifaceted approach tochanging handwashing behavior. Am J Infect Control 1997; 25:3-10.

20. Pittet D. Improving adherence to hand hygiene practice: a multidisci-plinary approach. Emerg Infect Dis 2001; 7:234-240.

21. Lam BC, Lee J, Lau YL. Hand hygiene practices in a neonatal intensivecare unit: a multimodal intervention and impact on nosocomial infec-tion. Pediatrics 2004; 114:565-571.

22. Johnson PDR, Martin R, Burrell LJ, et al. Efficacy of an alcohol/chlor-hexidine hand hygiene program in a hospital with high rates of noso-comial methicillin-resistant Staphylococcus aureus (MRSA) infection.Med J Aust 2005; 183:509-514.

23. Kretzer EK, Larson EL. Behavioral interventions to improve infectioncontrol practices. Am J Infect Control 1998; 26:245-253.

24. Wisniewski MF, Kim S, Trick WE, Welbel SF, Weinstein RA. The effectof education on hand hygiene beliefs and practices: a five-year experience.Infect Control Hosp Epidemiol. In press.

25. Materials to promote hand hygiene in your healthcare facility. Availableat: http://www.cdc.gov/handhygiene/. Accessed March 2, 2006.

26. Wisniewski MF, Kieszkowski P, Zagorski BM, Trick WE, Sommers M,Weinstein RA. Development of a data warehouse for hospital infectioncontrol. J Am Med Inform Assoc 2003; 10:454-462.

27. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitionsfor nosocomial infections, 1988. Am J Infect Control 1988; 16:128-140.

28. Schwartz DSL, DeMarais P, Trick W, Weinstein R. Beneficial effect of atraining intervention based on IDSA and SHEA guidelines for antibioticuse in long term care facilities. In: Program and abstracts of the 39thAnnual Meeting of the Infectious Diseases Society of America; 2001; SanFrancisco, CA. Abstract 490.

29. Gordin FM, Schultz ME, Huber RA, Gill JA. Reduction in nosocomialtransmission of drug-resistant bacteria after introduction of an alcohol-based handrub. Infect Control Hosp Epidemiol 2005; 26:650-653.

30. Jernigan JA, Clemence MA, Stott GA, et al. Control of methicillin-re-sistant Staphylococcus aureus at a university hospital: one decade later.Infect Control Hosp Epidemiol 1995; 16:686-696.

31. Ostrowsky BE, Trick WE, Sohn AH, et al. Control of vancomycin-resistant enterococcus in health care facilities in a region. N Engl J Med2001; 344:1427-1433.

32. Hayden MK, Bonten MJ, Blom DW, Lyle EA, van de Vijver D, WeinsteinRA. Reduction in acquisition of vancomycin-resistant enterococcus afterenforcement of routine environmental cleaning measures. Clin Infect Dis2006; 42:1552-1560.

33. Vernon MO, Hayden MK, Trick WE, Hayes RA, Blom DW, WeinsteinRA. Chlorhexidine gluconate to cleanse patients in a medical intensivecare unit: the effectiveness of source control to reduce the bioburden ofvancomycin-resistant enterococci. Arch Intern Med 2006; 166:306-312.

34. Austin DJ, Bonten MJ, Weinstein RA, Slaughter S, Anderson RM. Van-comycin-resistant enterococci in intensive-care hospital settings: trans-mission dynamics, persistence, and the impact of infection control pro-grams. Proc Natl Acad Sci U S A 1999; 96:6908-6913.

35. Tenorio AR, Badri SM, Sahgal NB, et al. Effectiveness of gloves in theprevention of hand carriage of vancomycin-resistant enterococcus speciesby health care workers after patient care. Clin Infect Dis 2001; 32:826-829.


hand washing

Documents