Jmmal of Advanced Nursmg, 1991,16,1216-1225

Nurses' hands as vectors of

hospital-acquired infection: a review
Dmah Gould BSc MPhil RGN DipN CertEd
Lecturer tn Nurstng, Department of Nursirtg Studies, Kmg's College London, Untverstty of
London, Cornwall House Annex, Waterloo Road, London SEl 8TX, England

Accepted for pubbcahon J Apnl 1991

GOULD D (1991) Journal of Advanced Nursing 16,1216-1225

Nurses' hands as vectors of hospital-acquired infection: a review

Hospital-acquired lnfechons (HAl) are notonous for the manner m which they
comphcate the course of the onginal illness, increase costs of hospital stay and
delay recovery This review will bnefly outlme the problems presented by HAl
m developed countnes and present evidence that Staphyhcoccus aureus and gram
negative bacilh, the mam causative agents, reach susceptible patients via the
contact rather than airbome route, predommantly on the hands of hospital staff
Good hand hygiene could help reduce the economic burden and patient distress
caused by MAI, but there is evidence that it is infrequently and poorly performed
by nurses, the health care stafiF most frequently in contmuous contact with
patients Possible reasons are explored in an attempt to ldenhfy strategies to
improve hand hygiene

INTRODUCTION
Hospital-acquired mfechons (HAl), defined as mfedions
which are neither present or mcubatmg before hospital
admission (Scheckler 1978), are notonous for their
economic burden on the health service and dishess
brought to pahents In consequence, the epidemiology of
HAl has attracted considerable research attention, though
prease current costs are difKicult to establish and complicated to calculate if all relevant fadors are considered, yet
rapidly become outdated (Dixon 1978, Rubenstein et al
1982)

contnbutes diredly to morbidity and to mortahty (Cross

et al 1980), especiaUy among the most debihtated pahents
(Bntt et al 1978), who are most likely to be lmmunocompromized (Zunmerh 1985) HAl most conunonly
results from bactena which are present on the skm
NORMAL SKIN FLORA: APPLYING

KNOWLEDGE TO THE DEVELOPMENT OF

HOSPITAL-ACQUIRED INFECTIONS

Reybrouck (1983) affirms that knowledge of normal skm

flora IS of value in the prevention of HAl because it provides the basis for our understandmg of the sigmficance
and incidence of chfferent types of bactena camed on skm,
Scale of the problem
espeaally hands, and suggests how this informahon can
In Bntam and the USA, HAl cxxurs with greatest frequency be applied to develop effedive hand hygiene pohaes
among surgical pahents The site mostfi-equentlyaffeded This rational approach is not without problans, as the
IS the unnary hact, particularly among cathetenzed demarcahon between what conshtutes normal and abnorpahents, foUowed by wounds and the lower respiratory mal IS often hazy (Schaechter 1989) Extensive studies by
had (see Meers et al 1981, Scheckler 1978) A rehospec- Noble & SommerviUe (1974) determined the composition
hve study of 16 hterature reports between 1933 and 1973 of normal skinflorato be coagulase negahve Staphylococa
revealed that hospital stay was prolonged between 13 to and Comeybactenum Greater chscrepancy reigns over
26 3 days as a result of HAl (Brachman et al. 1980) HAl numbers and location A sensihve biopsy method described
1216

Hospital-acqmred infection

by Selwyn & Elhs (1972) suggests considerable vanation m

bactenal counts on different parts of the skm, with a much
higher density m moist than dry areas
Healthy mtad skm resists bactenal invasion, both
because it exhibits a degree of self-disinfection towards
contaminants and because under normal areumstanees its
deheately balaneed eeosystem of eommensal orgamsms
help to keep foreign speaes at bay However, eontanunation ean sometimes result m longer-term eolomzahon by
exhaneous baetena, espeeiaUy when the skm is unusually
moist or damaged (Ojajarvi 1979), a factor of eonsiderable
lmportanee when eonsidenng prevention of eross infeehon
Transfer of baetena from one part of the body to another
site, normally free of orgamsms, may result m lnfeehon the
damp penneal skm of patients with spmal eord mjunes may,
for example, become eolomzed with coliforms (Sanderson
& Weissler 1990) Migrahon mto the bladder via a unnary
catheter results in badenuna (Sanderson & Rawal 1987)
Also relevant is the tendency of hospital patients to aequire
a skm flora chfferent to that of the general population
(Montgommene & Morrow 1978) whieh may eontanunate
their immediate environment providmg opportumhes for
eross uifeehon Its eonshtuent baetena may be more antibiotie resistant than the skm flora of healthy adult members
of the general population (Larson et al 1986)
Considerable attenhon has been paid to the normal
flora of the hands Nearly 50 years ago, Pnee (1938)
dishnguished between 'resident' and 'hansient' baetena
through quantitative laboratory handwashmg studies
Those organisms whieh eould eventually be removed by
repeated and thorough handwashes were categonzed as
hansient, thought at the time to represent contammants
which under normal arcumstances would probably che
withm 24 hours of moculation The remaimng resident
badena, regarded as the true skm flora, persisted deep
m the duds of sweat glands and subungal spaces The
existence of hansient and resident hand flora on the basis of
whether or not they can be removed by stnd hand hygiene
has smce been venfied by Hann (1973) and Gross et al
(1979), but it has become apparent that contaminants,
espeaaUy gram negahve speaes, may be camed for weeks
or months (Cooke rf a/ 1981, Larson 1981) Handcamage
among approxunately 20-30% of hospital staff has been
reported (Bruun & Solberg 1973, Adams & Mame 1982),
but isolahon rates of up to 80% have been menhoned m
relahon to neonatal and bums umts (Kmttle et al 1975)
Weanng nngs mcreased camage rate of underlymg bactena
dunng field stucbes (Hoftnan et al 1985), but laboratory
expenments have yet to dononstrate any mcreased nsk of
cross utfedion firom bactena beneath nngs Qacobson et al
1985)

Reybrouck (1983) emphasizes that isolation of the same

bactenal strains from patients and the hands of hospital
staff reported m numerous studies does not conshtute
absolute proof of eross infedion, but it is highly suggestive, espeeiaUy today with sensitive methods of serotypmg
baetena, and evidenee of sueeessful conhol of outbreaks
once a stnct handwashmg regune has been implemented
More defimte evidence of cause and effed may never
become available when mvestigatmg cross mfechon,
especiaUy by the contad route (Stamm et al 1981)
Today, the agents responsible for most HAl are
Staphylococcus aureus and gram negahve rods, with other
pathogens sometimes responsible for outbreaks In aU cases
there is suffiaent evidence to demonstrate that spread is
pnmanly via the contact route, with hands, the part of the
body m most continuous contad with pahent and environment, probably playmg a major role The role of airbome
spread and the manimate environment as a source of HAl
wiU now be explored, eondudmg that these are of mueh
less significance

Staphylococcal infections
During the 1950s, Staphylococa were recognized as
responsible for mcreasmg rates of HAl (Goodall 1952,
McDermott 1956) and their abihty to develop anhbiohc
resistance was an emergmg problem (Colebrook 1955)
Early work focused on air dissemmation via skm scales as a
possible route of spread, particularly from some members
of staff identifiable as 'heavy chspersers' Expenments with
medical students demonshated that approxunately 14% of
the male population acted as persistent permeal earners of
Staph aureus Dispersal of free badena mto the cur eould
oeeur dunng exerase m a speaal ehamber (Ridley 1959),
but under nonnal areumstanees these would probably
attaeh themselves to dothes Nasal eamage was reported
as more widespread, but field and laboratory stuches
mchcated that dissemination usuaUy occurred not chrectly
through the air m droplets, but by an mdired route m
which nasal secretions were first found to contammate skm,
dothmg and probably hands (Hare & Thomas 1956) It was
suggested that limited hansfer might occur via fnciion or
air cunrents, but later mvestigations mvolvmg the use of a
sht sampler to deted airbome transmission dunng a major
staphyloccKcal outbreak revealed that, if this occurs at aU, it
operates over only very short distances (Peaeoek et al
1980)
These observahons eonfirm the results of mgemous
field studies which today would probably be prohibited cm
ethical grcHinds (Mortimer et al 1966) These were conducted m a neonatal umt to hace spread of Sta|>hyloccx:a
1217

D Gould

and Streptococa from babies already colonized between survive significantly longer than 'non-outbreak' bactena
nurses and other infants There was very bttk spread bom when arhfiaally moculated onto the hands of laboratory
nurses to infants who were m proximity but not touchmg volunteers
When contamination via the airbome route was prevented,
Persuasive evidence for the hands as vectors of hospitala 43% transmission rate occurred from coloruzed to pre- acquired gram negahve sepsis is provided by Casewell &
viously uncolonized babies, providing the nurse did not Philbps (1977) who demonstrated that 16% of staff m an
wash hands between contacts Anhseptic handwashmg mtenstve care unit had Klebsiella hand contammation of
reduced transmission rate to 14%
the same serotypes as those colomzmg pahents LaboraAirbome spread of Staphylococa remams a major prob- tory expenments showed that badena remained viable up
lem m theatre, espeaally where orthopaedic prostheses are to 150 minutes following arhficial moculahon onto the
implanted Oalovaara & Puranen 1989) and also m bums hands suffiaent time for cross mfechon to occur dunng
uruts where pahents have lost large areas of skm, the normal nursmg duhes Qothmg, ward air and dust samples
body's chief defense against mfechon (Ayltffe & Lowbury were seldom contammated, supporhng work reviewed
1982) In the ward, hazards of airbome spread, mduding earlier by Noble et al (1976) which conduded that,
over short distances from skm scales on dothes, appiear to although some individuals disperse gram negahve bactena
have been over-eshmated (Mackintosh 1982) Babb et al heavily, there is no evidence to support airbome spread
(1983) demonstrated that even when dothes are heavily Contmued woric over a 4-year penod demonstrated that
contaminated by Staphylococa released m large numbers 24% of 2315 cnhcally ill pahents became colonized with
from a heavily discharging wound, this does not appear to Klebsiella, almost always with the same capsular strains
represent a sigruficant threat to other pattents on the same (Casewell & Philbps 1978) Possession of a mucus covermg, not carnage of an anhbiohc-resistant plasmid, was
ward
apparently
the influential fador m bactenal survival on
Unfortunately, the mtroduchon of new synthehc
finger
tips
(Casewell
& Desai 1983) Outbreaks of gram
penidlbns which brought dramahc improvements m the
treatment of Staphylococcal infections throughout the negative mfechon have been traced to nurse earners and
1960s allowed complacency to develop (Cafferkey et al arrested when culpnts were removed from patient contact
1985), while promotmg multiply resistant Staphylococcal (Burke effl/ 1971)
strams Consequently, the 1970s and 1980s have been
punctuated by repeated epidemics of methialbn resistant
Staphyhcoccns auretts (MRSA) throughout the world Pob-Hand carriage of pathogenic organisms
aes for controi vary accordmg to available faabhes and From time to time, nurses' hands must inevitably become
circumstance (SpKer 1984), but exhaushve mveshgahon of contammated with pathogenic organisms, espeaally as
outbreaks mdicates that hands ofiFer the duef means of there is evidence that bedpan washers and disinfedion
spread, with cnhcally ill pahents who become colonized or procedures do not adequately destroy all entenc pathogens
infeded operatmg as reservou-s (Thompson et al 1982) (Cune et al 1978, Block ei al 1990) Survival on hands is
Nurses can become earners, conhnbuhng to nsks of cross possible for some hours (Samandirfal 1983) and the hands
mfechon (Shanson 1985), a factor which may cause anxiety of pahents may also become contammated, mcreasmg nsks
m their professional and personal bves (Tuffriell 1988)
of cross mfechon (Lawrence 1983, Pntchard & Hathaway
1988)
A study by Black et al (1981) is one of the few
Gram negative bacteria
expenmental studies designed to show a causal link
Unbke Staphylococa, gram negahve bacteria do not between handwashmg and nsk of mfechon Followmg the
generally resist dessicahrai and controi can be effected to a mtroduchon of a stnct handwashmg programme in a day
large extent by providmg an environment that is dean and care centre, the inodotce of dianhoea among children in
dry (Maurer 1985) They tend to colonize pahents who are the study centre was significantly and consistently lower
immunocom{n-omized and, once agam, mfection is more than m control centres over a 35-week penod Larson
bkely to follow colonizahon (Moody el a/ 1972) Initial (1988), remaricing on the pauaty of prospechve dmical
studies by Lowbury (1969) suggested that most gram tnals to test a causal bnk between hand hygiene and HAI,
negahve badena dry out and die rapidly when inoculated attributes their absence to pioneers of the nud-mneteith
onto human skin, a ccmclusirat since substanhated by century (Seminelweis, Lister iuid Nightingale) who d^^eded
Cooke et al (1981) who established Hiat ^^ectes and shams sudi dramahc reduduHts m moibidity and mortality from
previously responsd>ie for hospAt^ outbreaks were able to mfechcm by implementmg hygiene into health care that
1218

Hospttal-acqmred mfechon

anhsepsis has too long been recognized as important and

benefiaal for research withholding it to be considered
viable on ethical grounds Most evidence comes mdirectly
from shidies already reviewed here However, if evidence
for a direct link bebveen hand hygiene and HAI is lacking,
it IS provided by work with respiratory pathogens which
appear to depend to a large extent on spread by the contact
route (Gwaltney et al 1978) Prevenhon is achieved when
hand hygiene compliance is good (Ledair et al 1987)

HANDS AND INANIMATE ENVIRONMENT

The surrounding environment has little bearing on rates of
HAI (McGowan 1981, Bauer rfa/ 1990) This is confirmed
by Maki et al (1982) m a 'natural' expenment, possible
when a hospitai moved from old to new, more spaaous
premises where facilities (including improved ventilahon
mtended to reduce airbome spread) had been upgraded
Extensive microbiological surveillance before and after the
move revealed that despite greater environmental contammahon m the older building, rate of HAI remamed
unchanged
A few authors have apparently lncnminated the
environment in HAI, but m all cases a link between
environment and susceptible patient must logically exist
Bentham (1979), descnbmg an outbreak of Klebsiella,
suggested that the floor around a leakmg bedpan macerator had acted as a reservoir, but acknowledges that the
route from floor to pahent was probably via nurses' hands,
unwashed after removmg overshoes Similarly, Carter

(1990) demonstrated high counts of aerobic baalli on the

floor of an mtensive care unit and on nurses' hands
Transfer could never be venfied absolutely, but is
suggested to have occurred in the same way

Links between faulty hand hygiene, equipment and

HAI
Invasive devices bypassing the body's nahiral bamers to
micro-organisms vastly mcrease nsks of HAI (Tafuro &
Rishicaa 1984) Mulhall (1990) pomts out that although
doctors are usually responsible for sitmg mtravenous
cannulae, catheters and endotracheal tubes, nurses look
after them, providing care whidi, though rouhne, is complicated Rates of mfechon related to parhcular typ>es of
equipment show considerable vanahon accordmg to the

6nchngs of an extensive multicentre madence study

(Nystrom ei al 1983), although there is little doubt that
high dependency pahents undergomg more procedures
are at greatest nsk (Daschner 1985) There is also some

evidence that nsk of sepsis is mcreased when new techniques are mtroduced with which staff have limited
expenence
A prospective survey by Dumas et al (I97I) drew
attenhon to high levels of contamination assoaated with
intravenous volume control sets, lmked to poor maintenance (leakage, dirty injection ports) and breaches m
asepsis, espeaally handwashmg Later prospechve studies
recorded lower infechon rates explamed through new, less
easily contammated designs of equipment and the simultaneous development of stnct protocols for asepsis
(Buxtoneffl/ 1979, Shmozafa rf / I983,Leroyrfa/ 1989)
Where asephc technique broke down, mfection was more
likely to supervene This evidence lends weight to HAI
bemg dependent mainly on the contact route for spread,
with hands, which manipulate equipment, playmg a vital
role
Handwashmg performance
Over the years, the results of microbiology and field
studies have mdicated repeatedly that scrupulous hand
hygiene remams the smgle most important factor favourmg reduction of HAI (Lowbury et al 1970, Larson 1981,
Larson 1989), a suggestion which should be welcomed, as
hand hygiene is relahvely uncomplicated and mexpensive
Its aim IS to remove all non-resident micro-orgarusms to
below the level necessary to conshtute an lnfechve dose
before transfer can occur to a susceptible patient
Although a quick, perfunctory handwash with soap and
water followed by bnsk drying has been rejwrted m one
study to remove transient bactena (Sprunt et al 1973), field
and laboratory studies have reached agreement on the
supenonty of skm disinfectants (eg chlorhexidine and
povidone-iodine), providing the handwash is long enough
for them to exert efifect Some of these agents exert a
culmmahve effect if used repeatedly, which soap and water
does not, but it is important to recognize that any agent
suffiaently gentle for applicahon to human stan will not
destroy or remove all existmg bactena
The evaluation of handwashmg is a ccHnplicated task
compnsmg not only choice of appropnate agent, but also
frequency, durahon, appropnateness (whether hands are
washed whai they should be) and perfomiance of technique (Larson & Lusk 1985) Research has consistently
shown that all aspects may be faulty
Albert & Comke (1981) surveymg frequency of handwashmg m an ITU over 1014-hour penods, observed that
staff washed their hands less than half the hme foUowmg
pahent contact Their cntenon iac 'contact' was strict, as
it mvolved minimal touchmg (eg pulse takmg), but is

1219

D Gould

probably justified as CaseweU & PhiUips (1977) demonshated that sudi adivities can result m hansfer of 10^
CFU (colony forming units) to nurses' hands This may
be suffiaent to conshtute an lnfechve dose or result m
colonization if transferred to a very debihtated pahent the
study was undertaken m ITU
Albert & Condie (1981) chd not attempt to document
appropnatoiess This issue has been addressed by Taylor
(1978), who estabhshed dunng 129 observations of handwashmg episcxles that nurses did not chstmguish between
dean and chrty situahons, a findmg later corroborated by
BroughaU et al (1984) Taylor attnbuted this to nurses'
apparent behrf that, unless visibly soiled, hands caimot
spread mfechon, although herresearchwas not designed to
test this Handwashmg durahon is often bnef (Quraishi et
al 1984), averagmg 8 8 seconds accordmg to Graham
(1990), compared to 10 seconds recommended by CDC
A possible cnhasm of many stuches is that presence of
an observer may have mfluenced normal behaviour, even
though staff were not told the real purpose unhl after data
coUechon was complete m most cases However, Leonard
(1986) and Larson et al (1986a) have both commented
on the encnmous vanahon m handwashmg frequency
between different nurses, suggestmg that for a task as
routme as handwashmg there is httle evidence of Hawthorn
Effed This problem was overcome altogether by BroughaU
et al (1984), who recorded handwashmg frequency by a
momtonng system attached to soap dispensers found m
tnals to operate with 93% accuracy Nurses washed hands
on an average of 5-10 times per shift, but claimed to do so
more often when asked to rate frequency by the researdi
team, afindmgsubstanhated by Larson et al (1986b)
In most stuches, authors have attempted to rate only a
few of the fadors suggested by Larson & Lusk (1985),
perhaps because such dose and detailed observahon is time
consummg and difficult to orgaruze, especaaUy when the
researcfi design donands that staff should be kept unaware
of the true purpose of the study Performance of techruque
has been examined least of aU, notably m one of the
smaUest scale stuches (Taylor 1978) Quahty of handwashmg tended to be pcwr, with some surfaces omitted
repeatedly
In recent years, health care professionals have become
ccHTcerrred not cmly with preventing HAl but also protedmg themselves against blcx>d bome pathogens (HTV, HBV)
by weanng gloves when handhng blood and bcxly fluids
This has led to confiision about the need to wash hands
after gloves have been ronoved, as some audK>rs claim this

may not always be necessary Oackson & Lynch 1984) Tliis

VKW IS erroneous g b \ ^ can become pundured m use
(KoriKiwicz et al 1989), allow passa^ of virus paitides
1220

even when mtad (Komeiwicz 1989), split under pressure

(Dalgleish & Malkovsky 1988) and promote mulhphcahon
of skm bactena by creahng warm, moist condihons
(McGinley et al 1988) They must be changed between
every pahent as they carmot be washedfi^eeof pathogens
(Doebbelmgrffl/ 1988)

EXPLORING REASONS FOR POOR HAND

HYGIENE
The need to reduce HAl has been recognized durmg the
development of quality assurance programmes m view of
the dear relevance to pahent safety and tangible economic
retum coupled with the relahvely measurable nature of
mfechon rates (Shaw 1986) However, CadwaUader (1989),
chsappomted after the implementahon of a new mfechon
conhol policy, conduded that the experhse of microbiologists and infechon conhol nurses wdl be of limited
benefit m the absense of commitment fi-om nurses who
must implement their suggeshons Lack of motivation
and accountabihty for HAl on an mdividual basis may
be contnbutory fadors (Nursing Times News 1991) A
queshonnaire study by Larson & KiUien (1982) sought to
identify fadors which mfluenced staff to wash or not wash
hands Inchviduals were aware of the need to reduce HAl
but were deterred through the possibility of developmg
sore, dry skm The authors judged that future compliance
might be secured by closer examination of deterrent factors A study in the Far East ldenhfied tachcs employed by
mfechon conhol nurses to secure comphance and asked
chnical nurses to idenhfy which approaches they found
most helpful (Seto et al 1990) Speaalist and ward nurses
found trust based on professional resped mutuaUy more
benefiaal than coeraon or threats from senior staff In the
UK, mfedion conhol nurses do not occupy hne managenal
posihons m the nursmg hierarchy and it is chfficult to
imagme coeraon havmg much impact m hospitals m our
scKaety
Lack of resources may be an issue related to motivation
Observmg that nurses tended to wash hands more often at
a sink posihoned near the nurses' stahon, BroughaU et al
(1984) proposed that more sinks placed nearer to the
pahent care areas might mcrease comphance A study by
Kaplan & McGuckhn (1986) found supporting evidence,
but Preston et al (1981), documenhng handwashmg and
infechon rates before and after the upgrachng of an ITU,
chd not
Evot when facihties are good staff may not wash hands
because they have developed sore, cby skm, itself undesirable as this uKreases bienal aAotazakion (Ojajarvi
1981) Nurses are weU aware of the nsks (see Larson &

Hospttal-acquiral mfechan

Killien 1982) A queshonnaire study by Newsom et al

(1988) estabbshed that choice of hand scrub preparahon
depended mainly on skm tolerance This problem is not
insurmountable as manufacturers are now paymg mcreased
attenhon to product acceptabibty Recent tnals have
demonstrated that cleansing with disposable alcoholic
wipes mcorporatmg emollients Qones et al 1986, Butz et al
1990), antimicrobial gel (Newman & Seitz 1990) or an
emulsion to replace soap and water (Kolan et al 1989) can
reduce crackmg, drymg and erythema while effectively
removmg transient bactena
Related to availabibty and acceptability of resources is
the issue of bemg too busy to use them Throughout the
bterature, there are numerous suggeshons that at very
busy times hand hygiene is more likely to break down
(Lowbury et al 1970, Noone et al 1983), although Taylor
(1978), in a small-scale observahon study, could not relate
levels of ward achvity to handwashmg Haley & Bregman
(1982), employmg a mulhvanate statishcal model, conelated under-staffing and overcrowdmg m a neonatal nursery
to cross infection culmmatmg m a staphylococcal outbreak
In contrast to subjects m the study by Broughall etal (1984),
these nurses and dodors recognized and were concemed
about defects m hand hygiene when busy

Local policy
Local pobcy may influence handwashmg and glove weanng
specifically m relahon to catheter care (Crow et al 1988),
though m this study medical speciality, diagnosis and
reason for cathetenzahon did not Similarly, Ho-Yen et al
(1984), employing a queshonnaire to evaluate nurses'
knowledge of hepahtis B, could fmd no difference between
nurses employed m different dmical settmgs, a result
surpnsmg as the nsks of seroconversion parallel degree of
exposure to blood (Pantebckef/ 1981), a fact which might
have been reflected m staff educational opportimities
Inevitably, poor hand hygiene has been attnbuted to
lack of knowledge, a view endorsed by Sedgwick (1984),
who pomts out that apart from teaching in relation to
aseptic technique, nurses receive bttle guidance Possibly
this IS because handwashmg is regarded as a 'social' rather
than a 'technical' or 'professional' achvity The impact of
theoretical lnstruchon on dimcal performance of asepsis
appears to be an under-researched area (Feldman 1969)
Although providmg more acceptable altemahves to
soap and water results m sbghtly improved compliance
when evaluated over short penods of time (Graham 1990),
there is limited mdicahon that 'educahonal' campaigns
have efifechve long-term benefit Williams & Buckles'
(1988) longitudinal quasi-expenmental study measured

knowledge and athtudes to liAI before and after staff were

exposed to a senes of pamphlets, posters and videos m a
test hospitai compared to a control where no intervenhon
had occurred Handwashmg frequency detected by electronic monitors attached to soap dispensers showed an
mcreased frequency of handwashmg matched by increased
knowledge, but 6 months later these effects were no longer
apparent
Mayer et al (1986) and Conly et al (1989) successfiilly
mcreased handwashmg practice m high dependency uruts,
but reported a decbne m compliance with mcreasmg time
since implementation of the educahonal and enforcement
campaigns Initial success was attnbuted m these studies to
providing staff with feedback on rates of handwashmg,
perhaps mcreasmg their sense of accountabibty T o p up'
campaigns are probably needed for reinforcement with
staff tumover Becker et al (1990), reporhng on sharps
injury and lack of compbance with sharps disposal pobcy,
attnbutes the disappomtmg effects of conhnumg educahon
to lack of speaficity teachmg is usually the same for all
staff, regardless of dimcal settmg or length of expenence
This research team concluded that before improvements
m practice and motivation can be expected efforts are
necessary to estabbsh knowledge and bebefs already held
by mdividual members of stiff, followed by educahon
more tailored to particular need
Role models
On a more posihve note, good role models may mcrease
hand hygiene compbance (Larson 1983) and there is
evidence that the mtroduction of mfechon control baison
nurses dmical nurses who have had additional trairung
m mfechon control may enhance awareness of nsks and
influence prevenhon strategies (Chmg & Seto 1990)

CONCLUSION
This review has demonstrated that mcreasmg rates of HAJ
are due chiefly to spread by the contact route and that
disseminahon must occur to a considerable extent on
nurses' hands Hand hygiene, the most important means
of preventmg HAI, is often poorly performed, somehmes through lack of knowledge and also because even
when nurses have the requisite knowledge of appbed
microbiology, mohvation is poor
Poor facilities and equipment, bemg too busy and lack
of encouragement from suitable rok models may be
influential, but their contnbution is presently unknown
More time should be spent documentmg preasely what
nurses know zhouk HM and how they perform all aspects

1221

D GouM
of hand hygiene before posihve attempts are made to
provide them with informahon they presoitiy lack and
encouragement to perfonn more efiechvely
In the present dunate of educationaireformat basic and
postbasic level, the prevenhon of HAI through nursmg
prachce should be regarded as an important dudlenge

Butz A M , Laughon B E . Gullette D L & Larson E L (1990)

Alcdiol-unpregnated wipes as an altemahve m hand hygiene
Amencan Joumal of Infechon Control 18, 70-76
BuxtonAE,HighsmithAX & Gamer J A (1979)Contaminahon
of lnha venous infusion fluid effects of changmg administrahon
sets Annals of Intemai Maitarte 90, 764-768
Cadwailader H (1989) Settmg the seal on standards Nurstng
Ttmes 85(37), 71-72
Caffa-key
M T , Colonan D , McGrath B, Keane C T , Hone R,
References
Pomeroy H et al (1985) Methiallm resistant Staphylococct4s
Adams BG &MarneTJ (1982) Hand carnage of gram negahve
auretis m Dublin Lancet it, 705708
rods may not be transient Joumal of Hygiene 89, 2331
Carter R (1990) Rihial and nsk. Nursmg Ttmes 86(13), 63-64
Albert R & Condie F (1981) Handwashmg pattems in medical Casewell M W & Desai N (1983) Survival of mulhply-resistant
intensive care units New Englarui Joumal of Medtane 304,
Klebstella aerogenes and other gram-negahve baalli on finger
1465-1466
hps Jottmal of Hospttal Infechon 4, 350-360
Aydiflfe GA. J & Lowbury E J L (1982) Airbome infechon in Casewell M W & Phillips I (1977) Hands as a route of
h o ^ t a l Journal of Hospttal Infechon 3,117140
hansmission for Kkbstella species Bnttsh Medtcal Journal 2,
Babb JiL, Davies J G & Aylifie G A J (1983) Contaimnahon of
1315-1317
protective dothmg eu\d nurses' unifonns in an isolation ward
Casewell M W & Phillips I (1978) Epidemiological pattems of
Joumal of Hospttal Infectton 4,149-157
Klehstella colonisahon and mfection m an mtensive care ward
Bauer T M , O f n e r E , Just H M , Just H &DaschnerFD (1990)
Joumal ofHygtene 80,295-300
An epidemiological study assessing the relative importance of Chmg T Y & Seto W i i (1990) Evaluatmg the effiaency of
airbome and direct contact h'ansnussion of micro-organisms in
the infechon control liaison nurse in the hospital Joumal of
a medical mtensive care unit Jottmal of Hospttal Infechon 1 5 ,
Advanced Nurstng 15,1128-1131
301-309
Colebrook L (1955) Infechon acquired m hospital Lartcet ii,
Becker M H , J a n z N K , Band J,BartleyJ, Snyder M B fcGaynes
885-891
R.P (1990) Non-compliance with universal precauhons
Conly J N , Hills S, Ross J, Lertzman J & Louie TJ (1989)
policy why do physicians and nurses recap needles? Amencan
Handwashmg prachces m an mtensive care umt the effects of
Joumal of Infechcm Control 18,232-239
an educahonal programme and its relationship to mfechon
Bentham AJ (1979) An tnveshgahon tnto a aoss-tnfechm probrates Amencan Joumal of Infechon Control 17,330-339
lem tn a cardio-thoraac umt Proceedtngs of the Tenth Annual
Cooke E M , Edmondson A S & Starkey W (1981) The ability of
Sympostum of The Infechcm Control Nurses' Assoaahon ICNA
strams of Klebsulla aerogena to survive on the hands Joumal of
London
hAaitcal Mtcrohohgy 14,443-450
Black R.E, Dykes A C , Kern EA. & Wells S P (1981) HandQ o w R., Mulhall A & Chapman R (1988) Indwelbng catheterwashmg to prevent diarrhoea in day centres Amencan Joumal
lzahon and reiated nursmg prachce Jottmal of Advanced
of Eptdemtology 113,445-451
Nmmg 13,489-495
Blodc C , Baron O , Bogokowski B, Amit P & Rubenstem E
Cune
D , Speller D C E, Simpson R.A., Stq>hens M & Cooke D I
(1990) An m-use evaluahon of decontanunahon of poly(1978) A hospitai epidenuc caused by gentanuan-resistant
prolene versus steel bedpans Joumal of Hospttal Infectton 16,
Klebstella
aerogenes Joumal of Hygterte 80,115123
331-338
Daigleish
A
G
& Malkovsky M (1988) Surgical gloves as a
BrachmanPS,DanBB,HaleyR.W,HootonTM,GamerJ.S &
mechanical
bamer
against human unmunodefiaency viruses
Allen J R. (1980) Nosocomial surgical infechons madence and
Bnttsh
Joumal
of
Surgery
75,171-171
cost Surgtcal Chmcs of North Amenca 60,1525
Daschner
F
D
(1985)
Useful
and useless hygienic techniques m
Bntt MR., Sdvaupner CJ & Matsumiya S (1978) Seventy of
mtensive
care
units
Intenstve
Care Medtane 11,280-283
underiymg disease as a predictor of nosoccxnial mfechon
Dixon R E (1978) The effect of infechons on hospital care Annals
Jourrtal of the Amencan Medtcal Assoaahon 239,10471051
of Intemai Medtane 89(2), 749-753
Broughall J M , Marshman C & Jackson B (1984) An automahc
monitormg system for measunng handwashmg heqaency m Doebbelmg B N , Pfaller M A , Houston AiC & Wenzel R P
(1988) Removal of nosoccnnial pathogens from the contamihospitai wards Joumal of Hospttal Infectton S, 4A7A53
nated glove Annak of Intemai l\4eitane 109,394-398
Bruun J N & Solberg C O (1973) Hand carnage of gram
negahve baaib and Staphylococcus aums Bnttsh Medical Dianas R J, Wamer J F fc DaltOTi H P (1971) Sephcaonia from
Joumal 2,580-552
mtravenous infusions New Et^kmd Joumal of Medtane 284,
257-260
B u r k e J P , I n g a l l D , K l e m J C G e z o n H M fcFinlandM (1971)
Protet4S mtmlnbs aied^xts m a hos{Mtal niawiy h-aced to a FddnuBi H (1%9) ljexasp% t r a n ^ from pn^raimned instnichuman earner New Et^Umd Joumal of f^kd$cme 284,115121
hcoi to diracal performaice Nurmng Researdi 18(1), 51-54
1222

Hospttal-aafutred mfectum
Goodali J W D (1952) Qoss-mfection in hospital wards its Komeiwicz D M , Laughton D , Butz A. & Larson B (1989)
mcid^ice and prevention Lancet i, 807-812
Integnty of vmyl and latex prex:edure gloves Nttrstng Research
Graham M (1990) Frequency and durahon of handwashmg m an
38(3), 144-146
mtensive care unit Amencan Joumal of Infection Control 18, Larson E (1981) Persistent camage of gram-negahve bactena on
77-81
hands Amencan Joumal of Infectton Control 9,112-119
Gross A , Cutnght D E , McGinley D & D'Allessandro S M
Larson E (1983) Influence of a role mexiel on handwashmg
(1979) Effect of surgical scrub on miaobial population under
behaviour Amencan Joumal of Infectton Control B., 146
fingemails Amencan Joumal of Surgery 13S,A6i-A67
Larson E (1986) Physiologic and microbiologic dianges m skm
Gross P A , Neu H C , Aswakopee N , Antwerpen C &
related to frequent handwashmg Infectton Control 7,25-63
Aswakopee N Deaths from noseKomial mfection Expenence Larson E (1988) A causal hnk between handwashmg and nsk of
in a Umversity hospital and commumty Amencan Joumal of
infection? Examination of the evidence Infectton Control and
Medtane 68,219-223
Hospttal Eptdemwlogy 9, 2835
Gwalhiey J M , Moskalski PB & Hendley J O (1978) Hand Larson E (1989) Wash your hands even when you use gloves
to hand transmission of rhmovirus colds Annals of Intemal
Amencan Joumal of Nursing 7,935939
Mediane 88, 463-467
Larson E & Killien M (1982) Factors mfluencmg handwashmg
Haley R W & Bregman D A. (1982) The role of understaffing and
behaviour of patient care personnel Amencan Joumal of
overcrowding m recurrent outbreaks of staphylococcal infecInfection Control 10, 93-99
tions m a neonatal special care unit Joumal of Infectious DiseasesLarson E & Lusk E (1985) Evaluating handwashmg techmque
145, 875-885
Joumal of Advanced Nursing 10, 547552
Hann J B (1973) The source of 'resident' flora The Hand 5,
Larson E, McGinley A R, Foglia G H , Talbot G M & Leyden
247-252
J J (1986a) Composition and antumcrobic resistance of skm
Hare R & Thomas C G A (1956) The transmission of
flora m hospitalized and healthy adults Joumal of Clmical
Staphylococcus aureus Bnttsh Medical Joumal 2,840844
Microbtology 23, 604-608
Hoffman P N , Cooke E M , McCarviUe M R & Enunerson A M
Urson E, McGinley K J, Grove G L, Leyden J L & Talbot G H
(1985) Micro-orgamsms isolated from skin under weddmg
(1986b) Physiologic, microbiologic and seasonal effects of
nngs wom by hospital staff Bnttsh Medtcal Joumal 290,
handwashmg on the skm of health care persormel Arrtencan
206-207
Joumal of Infection Control 14, 5159
Ho-Yen D O , &ossan M N & Walker E (1984) Nursing
Lawrence M (1983) Pahent hand hygiene a edmical enquiry
knowledge of hepahtis B mfection Joumal of Advanced Nursing
Nursing Times 79(22), 24-25
10,169-172
Ledair J M , Freeman J & Sullivan B F (1987) Prevenhon of
Jackson M M & Lynch P (1984) Too much or too httle?
nosocomial respiratory syncytial virus infeeions through
Amencan Joumal of Nursing 84,208-211
compliance with glove and gown isolation precautions New
Jacobsen G , Thick JE, Mccune JH & Farrell LD (1985)
England Joumal of Mediane 317, 329-334
Handwashmg nngweanng and number of micro-organisms
Leonard M (1986) Handlmg infection Nurstng Ttmes 82(38),
Nursing Research 34(3), 186-188
81-84
Jalovaara P & Puranen J (1989) Air bactenal and partide counts
Leroy O , BiUiau V , Beuscart C, Sanhe C, Chidiac C, Ranace C
in total hip replacement operations usmg non-woven and
et al (1989) Nosocomial infection assoaated with long-term
cotton gowns and drap>es Joumal of Hospttal Infection 4,
artenal cannulation Intensive Care Medtane 15, 241-245
Lowbury EJ, Thom B T , Lilly H A , Babb JR & Whittall K
Jones M V , Rowe G B, Jackson B & Pntchard N J (1986) The
(1970) Sources of mfechon with Pseudomonas aerugtnosa m
use of alcoholic paper wipes for routine hand deansmg results
pahents with hacheostomy Joumal of Medtcal Mtcrobtohgy 3,
of tnals in two hospitals Joumal of Hospttal Infection 6,
39-55
268-274
Lowbury E J L (1969) Gram-negahve baaUi on the skm. Bnhs>i
Kaplan L M & McGuddm M S (1986) Increasmg handwashmg
Joumal of Dermatology 1 (Siq>pl.), 55-61
comphance with more accessible sinks Infection Control
McDermott W (1956) The problem of Staphylococcal mfection
408-410
Bntish Medical Joumal 2, 838-840
Kmttle M A , Eitzman D V & Baer H (1975) The role of
hand contamination m the epidemiology of gram negative McGinley KJ, Larson E & Leyden J J (1988) Composihon and
density of micro-flora m the subungual space of the hand
noseKormal mfections Joumal of Paedtatncs 86,433-437
Joumal of Clinical Mtcrobtohgy 26,950-953
KolanPJ,OjajarviJ,LauharantaJ &MakelaP (1989)Cleansing
of hands wiHi enulsion a solution to skm problems of McGowan JE (1981) Environmental factors m nosoccnial
mfection a selective (oais Remew of Infecttous Disease 3,
hospital staff? Joumal of Hospital Infection 13,377-386
760-769
Komeiwicz D M (1989) Leakage of vims through stressed vmyl
Maontosh
C A (1982) A testmg tune for gowns Joumal of
and latex exammahon gloves Fifth International Conference
Hospttal
Infectton
3,5-8
on AIDS Montreal, June
1223

D Gould
M a b D G , Alvardo C, Hassemer C c Zilz MA (1982) Relahon
of the manimate environment to endemic nosocomiai
infection. New England Joumal of Medtane 307,1562-1566
Maurer I (1985) Hospttal Hygtate 3nl edn Edward Amold,
London
Mayer jA., Dubbert PM, MiUer M , Buricett P A & Chapman
5 W (1986) Increasing handwashmg in an intensive care unit
Infechon Conh-ol 7,259-262
Meers P D , Ayliffe G-A J, Emmerson A M , Leigh DA., MayonWhite R L , Macimtosh CA et al (1981) Report of the
NahtHuil Survey of Infection m Hospitals Jottmal of Hospital
Infechon 2, 23-28
Montgommene JZ & Morrow J W (1978) Pseudomonas
colonization m patients with spinal cord mjury Aimnam
Joumal of Epidennology 108,328-336
Moody M R., Young V M & K e n t o n D M (1972) Pseudomonas
aerugtnosa in a cenhe for cancer research Distnbuhon of
mtraspeaes ty|>es from human and environmental sources
Journal of Infechote Dtsease 125, 95-101
Mortimer EA., Wolinsky E, Gonzaga AJ & Rammelkamp C H
(1966) Role of airbome transmission of StaphylcKoccal
mfechons Brthsh Medical Joumal 1,319-322
Mulhall A B (1990) The contnbution of the basic saences to
nursmg practice research Joumal of Advanced Nursmg 13,
1354-1357
Newman J L & Seitz J C (1990) Intermittent use of an anhmicrobial hand gel for reducing soap-mduced lmtahon of
health care personneL Amencan Joumal of Infechon Control 18,
194-200
Newsom S W B, Rowland C & Wells F C (1988) What is in the
surgeon's glove? Joumal of Hospital Infechon 11 (suppl A),
244-259
Noble S W B, H a b b n a J D F , Van FurthR, Smith I fedeRaay
C (1976) Quanhtahve studies on the dis(>ersal of skm bactena
mto the air Joumal of Medtcal Microbiology 9,53-61
Noble S W B & Sommerville D A (1974) Mtcrobtology of the
Human Sktn Saunders, London
N o o n e M R , Pitt T L , B e d d e r M , Hewlett A M
ftRogersKB
(1983) Pseudomonas aerugtnosa coloruzahon in an intensive
therapy unit role of cross mfechon and host factors Bnhsh
Medtcal Journal 1,341-344
Nursing Times News Report (1991) Nurstt^ Ttmes 87(1), 10
Nyshxim B, Larson O S , Dankert J, Daschner E, Meers P D
6 Rotter M (1983) Bacteraemia m surgical patiotts with
mtravenous devices a European mulhcentre madence study
Joumal of Hospttal Infechon 4,338-349
Ojajarvi J (1980) Eflfiechvenessofhand washmg and disinfechon
methodb in removmg transioit bactena after pahent nursmg
Joumal of Hygiene 85,193-203
Ojajarvi J (1981) TTie importance of soap elechcm for rouhne
hand hygiene m hospital Jourral of Hygiene 85,275-283
Pantehck B S , Steer A C & Lewis H D (I98I) Hepahhs B
infections m hospital personnel dunng an e i ^ t year penod
pohaes fen- screoimg mid pregnaiKy m high n ^ areas
Amencan Joumal of Medicme 70,924-926

1224

Peacodc JE, Marsik FJ & Wenaal R P (1980) Methialhnresistant ^aphylococa4S awretis mtroduchon and spread within
a hospital Annals oflntermil Medtane 93,526-532
Preston G A , Larson EL & Stamm W E (1981) The effect of
pnvate isolation rooms on patient care practices coloruzahon
and mfechons on an intensive care unit Amencan Joumal of
Medtane 70,641-645
Pnce PB (1938) The dassification of transient and resident
microbes Joumal irflnfechous Disease 63,301-308
Pntchard V & Hathaway C (1988) Pahent handwashmg
prachce Nursir^ Ttmes 84(36), 68-72
Quraishi Z A , McGucklm M S & Blais F X (1984) Durahon of
handwashmg m mtensive care units a descriptive study
Amencan Joumal of Infectum Control 11,8387
Reybrouck G (1983) The role of hands m the spread of
nosocomiai mfechons Joumal of Hospttal Infedion 4,103111
Ridley M (1959) Penneal camage of Stepfc-oreus Bnttsh Medtcal
Joumal 1, 270-273
Rubenstein E, Green M , Nolan M , Ami P , Bemstem L &
Rubenstem A (1982) The effects of nosocomiai infections on
length and costs of hospital stay Joumal of Anttmtcrohtal
Chemotherapy 9 (si^pl.) 130
Samandi A R , Huq M I & Ahmed Q S (1983) Detechon of
rotavirus m handwashmgs of attendants of children with
diarrhoea Bnttsh Medtcal Joumal 1,188
Sanderson P J & Rawal P (1987) Contamination of the environment of spmal cord mjured patients by organisms causmg
unnary tract infection Joumal of Hospttal Infechon 10,173-178
Sanderson P J & Weissler S (1990) The relahon of eolomzahon
of the penneum to bactenuna and environmental contamination m spinally-mjured pahents Joumal (^Hospital Infechon
15,229-234
Sdiaechter M (1989) Ncjrmal miaobial flora In Mechanisms of
Mtcrobtol Dtsease (Sdiaechter M , Medoff G & Sdhlessinger D
eds), WiUiams and Wilkms, Baltimore
Schedder W E (1978) NoscKomial mfechons m a community
hospital 1972 through 1976 Archtves of Intemal Maitane 138,
1792-1794
Sedgwick J (1984) Handwashmg m hospital wards Ntastng
Ttmes 80(20), 64-^7
Selwyn S & Elhs H (1972) Skin bactena and skm dismfechon
reconsidered Bnttsh Joumal of Mediane 1,136-140
Seto W H , Chmg T Y , Chu Y B , Ng S H & Ong S G (1990)
Evaluahon of stafif compliance with 'mfluencmg' tachcs m
relation to infechon control pohcy implonentahon Journal of
Hospital Mfechon 15,157-166
Shanson D C (1985) Control of a hospital outbreak of methicillmresistant Staphyhcoccus aureus mfechons value of an isolahon
unit Joumal of Hospital Infechon 6,285-292
Shaw C D (1986) lntroduang Quahty Assurance Kmg's Fund
Paper No 64 King's Fund Centre, London
ShinozakiT,DefflieR5,Mazu2anJE,HamelAJ ScHazeltonD
(1983) Bactenal cont^nmahon of artoial lines A prospective
study Jowrtml of the Amencan M^iml ^oaahm
249,

223-225

Hospttal acqtttred mfechon

Spicer WJ (1984) Three shategies in the conhoi of Staphyle>cex:a mdueimg Methiallin-resistant Staphylococcus aureus
Joumal of Hospttal Infectton 5 (suppL A), 45-49
Sprunt K, Redman W &LeidyG (1973) Antibactenal effechveness of routme handwasbng Amencan Joumal of Pediatncs 5 2,
264-271
Stamm W E, Weinstem R A & Dixon R.E (1981) Companson of
endemic and epidemic nosocomial infections Amencan Joumal
of Medtane 70,393
Tafuro P & Ristucaa P (1984) Recognihon and control of
hospital-acquired infection in mtensive care units Heart and
Lung 13(5), 486-495

Taylor L J (1978) An evaluation of handwashmg tedmiques

Nursing Ttmes Part I 74(2), 54-55, Part n 74(3), 108-109
Thompson R L, Cabezudo I & Wenzel R P (1982) Epidemiology
of nosocomial infections caused by methiallm-resistant
Staphylococcus aureus Annals of Intemal Medtarte 9 7,309-316
Tuffnell C (1988) MRSA. lsolatuig pahents and nurses New
Zealand Nursing Joumal 81(8), 21-23
Williams E & Buckles A (1988) A lack of mohvahon Nursing
Times 84(22), 60-71
Zimmerh W (1985) Illness and latrogemc factors compremused
and host defense mechanisms Intensive Care Mediane 1 1 ,
174-178

1225