Journal of Hospital Infection (2005) 61, 123–129

www.elsevierhealth.com/journals/jhin

Infection control in neonatal intensive care units

J. Chudleigha,*, M. Fletcherb, D. Gouldc
a
Department of Children’s Nursing, City University, 20 Bartholomew Close, London EC1A 7QN, UK
b
Faculty of Health and Social Care, University of the West of England, Bristol BS16 1QY, UK
c
Department of Applied Biological Sciences, City University, London EC1A 7QN, UK

Received 19 July 2004; accepted 21 February 2005

Available online 18 July 2005

KEYWORDS Summary Healthcare-associated infection is a major problem in acute

Infection control;
Hand hygiene;
Neonatal intensive
care; Knowledge;
Fatigue; Atmosphere;
Environment
hospital settings. Hand decontamination is considered to be the most
effective means of preventing healthcare-associated infection, but is poorly
performed. Few studies have examined technique, which may be important
in neonatal intensive care units (NICUs) where clinical procedures are
intricate and could result in contamination of many areas of the hand,
resulting in cross-infection. This study examined technique in six NICUs.
Eighty-eight nurses were observed. A scoring system was developed so that
technique could be quantified and subjected to statistical testing. The mean
score was 6.29 out of 11 when hands were washed and 3.87 out of 7 when
alcohol hand rub was used, indicating that performance was not optimal.
Scores for technique were not significantly different in each NICU. Senior
nurses achieved higher scores for handwashing (P!0.01), as did nurses
holding positive feelings about the atmosphere in their NICU (PZ0.04).
Junior nurses scored less well on a knowledge questionnaire than senior
nurses (P!0.01). Nurses who had been employed in the neonatal unit for
less than one year also scored less well (P!0.01). Differences in technique
were noted when comparing the beginning and end of long shifts. These
differences were not noted at the beginning and end of standard shifts.
Q 2005 The Hospital Infection Society. Published by Elsevier Ltd. All rights
reserved.

Introduction

Healthcare-associated infection (HCAI) is a major

problem in healthcare settings. It occurs mainly
* Corresponding author. Tel.: C44 207 040 5431; fax: C44 207
040 5717. through cross-infection via hands. Hand decontami-
E-mail address: j.chudleigh@city.ac.uk nation is the most effective means of prevention.1–3
0195-6701/$ - see front matter Q 2005 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jhin.2005.02.017
124
There have been many reports of poor hand
decontamination,2,4 but most studies have only
examined frequency.5–9 Attempts to encourage
hand decontamination after procedures that are
most likely to result in cross-infection do not result
in sustained improvement.10 It has not been
possible to demonstrate an association between
knowledge or positive opinions about infection
control and hand decontamination performance.11
The technique of hand decontamination appears
to have been less well documented than frequency.
One exception is an early study by Taylor12 who
persuaded nurses to wash their hands with their
eyes shut. A dye was used, showing that fingertips,
thumbs and a strip along the palm had not received
contact with soap and water, demonstrating that
the nurses’ technique was suboptimal.
In Gould’s11 study, the technique consisted of
five components for washing with soap or anti-
septic: choice of product; duration; number of
surfaces decontaminated; drying; and disposal of
the paper towel. When alcohol hand rub was used,
decontamination consisted of three components:
choice of product; duration; and number of
surfaces decontaminated. A scoring system was
developed to assess technique. For the sample as a
whole, the mean score was 8.4 out of 12, a finding
which indicated that technique was better than in
earlier studies.
Since the publication of Gould’s11 study, very
little interest has been demonstrated in the
technique of hand decontamination. This is an
oversight in view of the emphasis placed on the use
of alcohol hand rubs, especially for nurses when
workload is high.13
HCAI is a problem in neonatal intensive care units
(NICUs) because of the immature status of the
infants’ immune systems and invasive procedures.14
Studies that investigate clinical practice in this
setting are important to identify variables that
might influence hand decontamination and how it
could be improved. For instance, heavy clinical
workload affects hand decontamination perform-
ance11 and results in increased rates of infection.15
A number of other variables may be influential but
their effects have never been investigated,
although they are known to affect the quality of
nursing care in more general terms. For example,
the atmosphere of the ward impinges on the quality
of care.16 No studies to examine the effects of
fatigue on infection control practice or reported
rates of HCAI exist. However, long (12-h) nursing
shifts are known to influence the quality of nursing
care generally17 and this is thought to be due to
fatigue. Failure to examine the relationship
between shift length and hand decontamination
J. Chudleigh et al.
appears to be an important variable in research
studies that examine clinical practice in settings
where length of shift may vary.
In some NICUs, some nurses work long shifts,
whereas standard (8-h) shifts are usual in other
NICUs. A study to explore hand decontamination in
NICUs thus provided an opportunity to explore the
effect of length of nursing shift as a proxy measure
of fatigue on hand decontamination performance.
Materials and methods
The aim of this study was to examine nurses’ hand
decontamination in NICUs. The influence of knowl-
edge of infection control, seniority, continuing
professional education, opinions of infection con-
trol, availability of resources, ward climate and
shift length were explored. Ninety nurses in six
NICUs participated. Eighty-eight were observed at
the beginning and end of their shifts. Observation
was conducted individually until 10 episodes of
patient contact had been obtained. The same
research assistant collected all the data.
In NICUs, the policy is not to handle infants more
than necessary. Care is planned so that all the
procedures required for a particular neonate take
place successively. Individual babies may not be
handled for some hours, and there is less scope for
initiating contacts between different patients.
Frequency of hand decontamination appears to be
less important than technique, especially as clinical
procedures and equipment are intricate and there is
potential for contamination between the hands and
equipment in close contact with the patient.
Technique
A scoring system was developed so that technique
could be quantified and subjected to statistical
testing. When hands were washed (using chlorhex-
idine or soap), the technique consisted of six
components: choice of product; duration; number
of hand surfaces decontaminated; use of taps;
drying; and disposal of the paper towel. When hand
rub was used, the technique consisted of three
components: choice of product; duration; and
number of hand surfaces decontaminated (Table
I). The relative importance of the individual
components was unknown, so each was scored out
of 15 when hands were washed and nine when
alcohol hand rub was used.12
The policy in Hospitals 1, 3, 4, 5 and 6 stipulated
that chlorhexidine or an alcohol hand rub should be
used before patient contact. After patient contact,
Infection control in NICUs 125

hygiene were adequate, and if they knew where

Table I Scoring system for technique of hand
to seek further information about infection
decontamination
control.
Component No. of points awarded
Soap/ Alcohol hand Ward climate
chlorhexidine rub Climate was assessed using a Likert scale adapted
Choice of product from a previously validated scale.24 High scores
Appropriate 2 2 indicated that the nurses’ perception was
Inappropriate 0 0 favourable.
Duration
O10 s 2 2 Fatigue
!10 s 1 1 The effects of fatigue were assessed by observ-
Surfaces of the hand included ing each nurse at the beginning and end of their
Five 5 5 shifts. Standard (8-h) and long (12-h) shifts were
Four 4 4
observed. Five observations were documented
Three 3 3
Two 2 2
for each nurse at the beginning of the shift and
One 1 1 another five observations were documented at
Tap use the end.
Correct 2 –
Incorrect 0 – Analysis
Hand drying Bivariate statistical analysis and analysis of var-
Thorough 2 – iance (ANOVA) were employed.
Not thorough 1 –
Not at all 0 –
Bin use
Results
Correct 2 –
Incorrect 0 –
Total possible 15 9
Participation
score
Ninety nurses employed in six NICUs participated in
this study. Two nurses were not observed as they
were not involved in direct patient care. Data were
chlorhexidine should be used if hands were likely to collected throughout day and night shifts. No
be contaminated, or an alcohol hand rub if not. The observations were missed. A total of 1760 obser-
policy in Hospital 2 was similar to that in the other vations were made; 880 at the beginning of shifts
hospitals except soap was used instead of and 880 at the end of shifts.
chlorhexidine.
Technique
Other variables
An appropriate product was chosen on 1274 (72.4%)
Knowledge occasions (Table II). There were significant differ-
Knowledge of HCAIs likely to occur in NICUs ences between the NICUs (c2Z186.497, 5 d.f., P!
(methicillin-resistant Staphylococcus aureus18–21 0.01). Nurses in Hospitals 2, 3, 5 and 6 chose an
and Group B haemolytic streptococcus)22,23 was appropriate product more often than expected.
documented on a questionnaire. The maximum
possible score was 92. Duration (Table II)

Seniority and previous nursing experience
Nurses’ seniority was recorded and they were asked
about continuing professional education they had
undertaken.
Opinions of infection control
Opinions were obtained by interview. The nurses
were asked if any products caused soreness or
allergy, whether facilities to undertake hand
The independent samples t-test demonstrated that
duration was significantly longer when hands were
washed (tZK18.384, P!0.01). The mean duration
for the sample was 12.24 s when hands were washed
and 6.26 s with hand rub.
A one-way ANOVA demonstrated significant
differences in the duration of handwashing, with
nurses in Hospital 1 taking significantly less time
than elsewhere (FZ11.089, P%0.05). A one-way
126 J. Chudleigh et al.

ANOVA demonstrated significant differences in

Mean overall hand hygiene

hand rub
duration when an alcohol hand rub was used, with

Alcohol

3.85
4.06
3.79
3.34
4.30
3.61

3.87
nurses in Hospital 3 taking significantly more time
than nurses in the other hospitals (FZ3.83, P!

score
0.01).

Handwash

5.77
6.95
6.17
6.37
6.29
6.21

6.29
Number of surfaces decontaminated
(Table II)
Incorrectly

The average score was 2.96 out of 5 when hands

Use of bin (occasions)

were washed and 2.58 out of 5 when hand rub

10
2
2
2
2
1
1
was used. An independent samples t-test demon-
strated significant differences between scores
when hands were washed compared with
Correctly

1230 occasions when hand rub was used (tZK6.65,

306
210
170
267
132
145

P!0.01).
A one-way ANOVA demonstrated no significant
differences between the NICUs when alcohol hand
thoroughly
Hand drying (occasions)

rub was used. When hands were washed, nurses in

Not

743
183
104
92
184
85
95

Hospitals 4 and 5 decontaminated more surfaces

of their hands than nurses in the other hospitals
(FZ7.25, P!0.05).
Thoroughly

412
125
108
80
85
48
51

Drying (Table II)

Hands were dried thoroughly on 497 occasions

(40.0%). The Chi-square test demonstrated signifi-
Scores for hand surfaces

hand rub
Alcohol

2.58
2.55
3.04
2.31
2.66
2.53

2.58

cant differences between the NICUs. Nurses in

decontaminated

Hospitals 1, 2 and 3 dried their hands thoroughly

more often than nurses in the other hospitals (c2Z
23.96, 5 d.f., P!0.01).
Handwash
Hand hygiene practices in each neonatal intensive care unit

2.75
2.91
2.85
3.20
3.21
2.94

2.96

Disposal (Table II)

The bin was used correctly (without touching) on

Duration of hand hygiene

hand rub
Alcohol

1230 occasions (99.2%). There were no significant

5.65
6.52
8.62
5.82
5.96
6.23

6.26

differences between the NICUs.

(s)

Overall score for technique

Handwash

10.22
13.00
11.44
13.99
12.89
12.44

12.24

Handwashing (Table II)

Four components of technique (duration, num-
ber of surfaces, use of taps and drying) showed
Inappropriate
Choice of hand hygiene pro-

significant positive correlation (P!0.05). It was

possible to summate and take the mean of these
486
190
34
46
123
55
38
duct (occasions)

four scores to give an overall score for tech-

nique. Mean score was 6.29 out of 11. A one-way
ANOVA demonstrated no significant differences
Appropriate

between the scores at each NICU.

1274
210
306
154
177
225
202

Hand rub (Table II)

Two components (duration and number of surfaces)
Table II

showed significant positive correlation (P!0.05). It

Total
Mean
Unit

was possible to summate and take the mean of

1
2
3
4
5
6

these two scores to give an overall score. Mean

Infection control in NICUs
score was 3.87 out of 7. A one-way ANOVA
demonstrated no significant differences between
scores at each NICU.
Variables likely to influence hand
decontamination
Location of data collection
There were no significant differences in overall
scores for hand decontamination between the six
NICUs.
Knowledge of infection control
Knowledge of infection control was poor. The mean
score was 56.3 (61.2%) out of 92 for the 66 (73.3%)
nurses who completed the questionnaire. There
was no relationship between knowledge and hand
decontamination technique.
Seniority and experience
The nurses had worked in neonatal intensive care
for 1–25 years. Most (NZ35, 53%) had been
employed in an NICU for over five years. Twelve
(18.2%) were junior nurses and 54 (81.8%) were
senior nurses. There was no association between
seniority and technique when alcohol hand rub was
used, but when hands were washed, the score for
technique was significantly higher for more senior
nurses (rZ0.370, P!0.01). Nurses who had worked
in an NICU for less than one year scored less well on
the knowledge questionnaire than those who had
been employed for longer (FZ5.18, P!0.01).
Junior nurses also scored significantly less well
(FZ7.73, P!0.01).
Opinions
Most nurses believed that the products available in
their NICU caused sore hands (NZ62, 68.9%) but
less than half complained about the quality of paper
towels (NZ39, 43.3%) and few reported allergies.
There were few reports of shortages of equipment
with 12 nurses (13.3%), eight nurses (8.9%), 10
nurses (11.1%) and 17 nurses (18.9%) reporting
shortages of soap, chlorhexidine, alcohol hand rub
and gloves, respectively. Most nurses (NZ84,
93.3%) knew where to seek help concerning infec-
tion control; 89 (98.9%) regarded infection as a
particular problem in the NICU and 90 (100%)
thought that infection control should be an import-
ant part of nursing care. There was no relationship
between opinions and hand decontamination.
Ward climate
The Likert scale indicated that perceptions of the
atmosphere in the NICU did not differ between
127
nurses in each hospital. However, nurses with
positive feelings about their work place achieved
significantly higher scores when hands were washed
(rZ0.275, P!0.04) although not when using alco-
hol hand rub. There were no significant relation-
ships between the results of the Likert score and
knowledge.
Length of shift
Nurses spent significantly more time decontaminat-
ing their hands at the beginning of long shifts (O
10 s) than at the end (!10 s) (c2Z11.131, 4 d.f.,
PZ0.03), but no significant differences were
recorded for standard nursing shifts. More surfaces
were decontaminated at the beginning of long shifts
when compared with the end of long shifts (UZ
182345.5, P!0.01). No differences were recorded
with standard shifts. Hand drying was better at the
beginning of long shifts than at the end (c2Z
15.229, 1 d.f., P!0.01).
Discussion
As in Gould’s11 study, scores for choice of product
were favourable, but results relating to the dur-
ation of decontamination were less impressive.
Duration was shorter when alcohol hand rub was
used, a finding which also corroborates the much
earlier findings of Ojajarvi et al.5 and which may
reduce the bactericidal effectiveness of alcohol
hand rub. Contact of the surface of the hands with
the decontaminant was incomplete and the inter-
digital space was frequently missed,11,12 meaning
that bacterial counts on these areas could not be
reduced. On the NICUs included in this study,
opportunities for cross-infection are likely to have
been compounded by incomplete drying, which
further increases the risk of transferring bac-
teria.25,26
In the present study, four components of
technique correlated significantly when hands
were washed. The overall score was 6.29 out of 11
compared with 8.4 out of 12 in the study by Gould,11
a finding that indicates less desirable practice in
line with much earlier work.27–29 Another difference
from the study by Gould11 was positive correlation
between two components when alcohol hand rub
was used, allowing the calculation of an overall
score for alcohol hand rub technique. Again, the
score (3.87 out of 7) suggests that technique could
not be regarded as satisfactory. This is an area of
concern in view of the current emphasis on the use
of alcohol hand rub in clinical areas throughout the
UK. 13 Although aspects of technique varied
128
between the six NICUs, scores were not better in
any single hospital.
Knowledge was poor, as in previous studies,11,30
despite increasing emphasis on the importance of
infection control in recent policy documents.31–33
As in earlier studies,10 there was no relationship
between observed hand hygiene and knowledge.
Most of the nurses in this study were senior and
experienced in neonatal intensive care. Scores
were significantly better for more senior nurses
only when hands were washed. However, knowl-
edge was better for more senior nurses. Many
nurses believed that the products available for hand
decontamination caused sore hands, supporting the
findings of other authors.34,35 However, supplies of
products were generally considered to be satisfac-
tory, a pleasing result which contradicts earlier
findings.34 This indicates that the recent publicity
to increase hand hygiene in the UK13 may be having
a positive influence. As in this earlier study,34
infection control was considered to be an important
part of nursing care and attitudes were generally
positive, although in both the previous34 and
present study, there were no relationships between
opinions and hand decontamination performance.
Overall, there were no significant differences in
nurses’ perceptions concerning the climate of their
NICU, a finding that is surprising in view of the
literature which has indicated that such variations
commonly exist.24 Differences emerged in terms of
hand hygiene performance for nurses between the
beginning and end of long (12-h) shifts. At the
beginning of long shifts, nurses spent longer
decontaminating their hands, were more thorough,
and dried their hands more effectively. No differ-
ences could be detected between performance at
the beginning and end of standard (8-h) shifts.
Assuming that nurses are more tired after working
long shifts, this finding is important and merits
further investigation.
References
1. Reybrouck G. The role of hands in the spread of nosocomial
infections. J Hosp Infect 1983;4:103—111.
2. Larson E. A causal link between handwashing and risk of
infection? Examination of the evidence. Infect Control Hosp
Epidemiol 1988;9:28—35.
3. Pittet D, Hugonnet S, Mourouga P, et al. Effectiveness of a
hospital-wide programme to improve compliance with hand
hygiene. Lancet 2000;356:1307—1312.
4. Kretzer EK, Larson EL. Behavioural interventions to improve
infection control practices. Am J Infect Control 1998;26:
245—253.
5. Ojajarvi J, Makela P, Rantasalo I. Failure of hand disinfection
with frequent handwashing—a need for prolonged field
studies. J Hyg 1977;72:109—119.
J. Chudleigh et al.
6. Albert R, Condie F. Handwashing patterns in medical
intensive care units. N Engl J Med 1981;304:1465—1466.
7. Broughall J, Marshman C, Jackson B, et al. An automatic
monitoring system for measuring handwashing frequency in
hospital wards. J Hosp Infect 1984;5:447—453.
8. Mayer JA, Dubbert PM, Miller M, et al. Increasing handwash-
ing in an ICU. Infect Control 1986;7:259—262.
9. Larson E, McGinley KJ, Foglia A, et al. Handwashing
practices and resistance and density of bacterial hand flora
on two pediatric units in Lima, Peru. Am J Infect Control
1992;20:65—72.
10. Naikoba S, Hayward A. The effectiveness of interventions
aimed at increasing handwashing in healthcare workers—a
systematic review. J Hosp Infect 2001;47:173—180.
11. Gould DJ. Nurses’ hand decontamination practice: results of
a local study. J Hosp Infect 1994;28:15—30.
12. Taylor LJ. An evaluation of handwashing techniques. Nurs
Times 1978;74:54—55 [see also p. 108—109].
13. National Patient Safety Agency. Patient safety alert: clean
hands help to save lives. London, UK: National Patient Safety
Agency; 2004. p. 1—4.
14. Albers MJIJ, Mouton JW, Tibboel D. Colonisation and
infection by Serratia species in a paediatric surgical
intensive care unit. J Hosp Infect 2001;48:7—12.
15. Haley RW, Bregman DA. The role of understaffing and
overcrowding in recurrent outbreaks of staphylococcal
infections in a neonatal special care unit. J Infect Dis
1982;145:875—885.
16. Smith P. The relationship between quality of nursing care
and the ward as a learning environment. Developing a
methodology. J Adv Nurs 1987;12:413—420.
17. Facey S. Lessons learnt from personal experience. Nurs
Times 1995;9:30.
18. Haddad Q, Sobayo EI, Basit OBA, et al. Outbreak of
methicillin-resistant Staphylococcus aureus in a neonatal
intensive care unit. J Hosp Infect 1993;23:211—221.
19. Asensio A, Guerrero A, Quereda C, et al. Colonization and
infection with methicillin-resistant Staphylococcus aureus:
associated factors and eradication. Infect Control Hosp
Epidemiol 1996;17:20—28.
20. Coello R, Glynn JR, Gaspar C, et al. Risk factors for
developing clinical infection with methicillin-resistant Sta-
phylococcus aureus (MRSA) amongst hospital patients
initially only colonized with MRSA. J Hosp Infect 1997;37:
39—46.
21. Taylor LR. MRSA. RCN Nurs Update Nurs Stand 1997;11:
4—25.
22. Mahlmeister L. Perinatal group B streptococcal infections:
the nurse’s role in identification and prophylaxis. J Perinat
Neonatal Nurs 1996;10:1—16.
23. Keenan C. Prevention of neonatal group B streptococcal
infection. Am Famil Physician 1998;57:2713—2720.
24. Adams A, Bond S, Arber S. Development and validation of
scales to measure organisational features of acute hospital
wards. Int J Nurs Stud 1995;32:612—627.
25. Ansari SA, Springthorpe VS, Sattar SA, et al. Comparison of
cloth, paper, and warm air drying in eliminating viruses and
bacteria from washed hands. Am J Infect Control 1991;19:
243—249.
26. Mackintosh C, Hoffman P. An extended model for transfer of
micro-organisms via the hands: differences between organ-
isms and the effect of alcohol disinfection. J Hyg 1984;92:
345—355.
27. Sedgewick J. Hand-washing in hospital wards. Nurs Times
1984;80:64—67.
28. Gidley C. Now wash your hands!. Nurs Times 1987;83:40—42.
Infection control in NICUs
29. Larson EL, Eke PI, Laughon BE. Efficacy of alcohol-based
hand rinses under frequent-use conditions. Antimicrob
Agents Chemother 1986;30:542—544.
30. Courteney M. The teaching, learning and use of infection
control knowledge in nursing. NT Res 1998;3:118—131.
31. National Audit Office. The management and control of
hospital-acquired infection in acute NHS trusts in England.
London: National Audit Office; 2000. p. 1—121.
32. Plowman R, Graves N, Griffin M, et al. The socio-economic
burden of hospital acquired infection. London: Public
Health Laboratory Service; 1999. p. 1—16.
129
33. Department of Health. Winning ways: working together to
reduce healthcare associated infection in England. London:
HMSO; 2003. p. 1—32.
34. Gould DJ, Ream E. Nurses’ views of infection control: an
interview study. J Adv Nurs 1994;19:1121—1131.
35. Larson E, Killien M. Factors influencing handwashing beha-
vior of patient care personnel. Am J Infect Control 1982;10:
93—99.