Journal of Hospital Infection (2002) 50: 276±280
doi:10.1053/jhin.2002.1202, available online at http://www.idealibrary.com on
Hand contamination before and after
different hand hygiene techniques:
a randomized clinical trial
J.-C. Lucet*, M.-P. Rigaud*, F. Mentrey, N. Kassisz, C. Deblangy*
A. Andremontzx and E. Bouvetx
*Infection Control Unit, yDepartment of Biostatistics and Epidemiology, zBacteriology Laboratory and
x Infection Control Committee, Bichat-Claude Bernard Hospital, Paris, France
Summary: The efficacy of alcohol-based handrubs (ABH) for hand hygiene (HH) compared with handwashing (HW) remains to be established in the clinical setting. Factors associated with severe hand contamination before HH techniques were medical ward, physician and not wearing gloves. Forty-three
healthcare workers [HCW, 26 nurses (N), nine nurse assistants (NA) and eight physicians (P)] each performed
six HH techniques in random order, immediately after a patient care activity: HW with non-antiseptic soap for
10 (US10) and 30 (US30) s; HW with antiseptic (polyvidone iodine- or chlorhexidine-based) soap for 10
(AS10), 30 (AS30) or 60 (AS60) s; and ABH (Sterillium, Bode Chemie, Germany). The fingertips of the
dominant hand were pressed on to agar for culture before and after each HH technique. Five hundred and
sixteen specimens were obtained. Log10-transformed bacterial count reductions after HH were 0.74, 0.51,
1.13, 1.14, 1.21 and 1.40 for US10, US30, AS10, AS30, AS60 and ABH, respectively; both AS and ABH were
significantly better than US. Qualitative assessment showed that 11 of the 256 pre-HH specimens (4.3%) had
pathogenic bacteria, and that two of these 11 remained positive after HH (US in both instances).
& 2002 The Hospital Infection Society
Introduction
Handwashing is recognized as a basic measure
for preventing nosocomial infections.1 However,
compliance with handwashing in hospital environments is generally less than 50%.2 Furthermore,
correct handwashing technique, particularly in
respect of duration, is often not practiced. Reasons
for insufficient compliance have recently been
Received 17 August 2001; revised manuscript accepted
5 February 2002; published online 18 March 2002.
Author for correspondence: Dr J. C. Lucet, Unite d'HygieÁne et de
Lutte contre l'Infection Nosocomiale, GH Bichat-Claude
Bernard, F-75877 Paris Cedex 18, France.
Fax: ‡33 (0)1 40 25 88 11;
E-mail: jean-christophe.lucet@bch.ap-hop-paris.fr
0195±6701/02/040276 1 05 $35.00/0
studied.3 Problems include insufficient facilities,
lack of training and information, a lack of time and
a high patient care load, as well as cutaneous intolerance of soaps. Alcohol-based hand disinfectant
solutions are used increasingly as an alternative to
handwashing4,5 and may enable many obstacles in
the way of handwashing to be overcome.
The microbiological efficiency of alcohol-based
hand disinfectants has been demonstrated in vitro5 in
reducing the bacterial count on artificially contaminated hands. In a clinical situation, several
studies have shown that handrubbing with alcoholbased hand disinfectants is more efficient than
with unmedicated soap.6±8
Experimental conditions, however, especially the
duration of handwashing, have not always been
described in these studies. Furthermore, there are no
& 2002 The Hospital Infection Society
Hand hygiene and hand contamination
comparative data on the efficiency of handrubbing
with an alcohol-based compound and handwashing
with antiseptic agents in clinical environments. We
conducted a randomized clinical trial to compare the
bacterial efficiency of various hand hygiene techniques, including handrubbing with an alcohol-based
compound, and handwashing with antiseptic agents
and with unmedicated soap. Factors predisposing to
hand contamination before hand hygiene were also
assessed.
Methods
The study was conducted in seven wards of the
Bichat-Claude Bernard hospital over a two-month
period (May and June 2000): the two medical ICUs,
the surgical ICU, the cardiac surgery ICU, the surgical recovery unit, and two medical units (infectious
diseases unit and internal medicine unit). Five to
seven volunteers from each unit, all healthcare
workers (HCWs), were asked to participate. The
volunteers were chosen so as to have at least one
doctor, one nurse assistant and two nurses in each
service.
Each volunteer performed six hand hygiene
techniques in random order immediately after
a healthcare procedure: handwashing with unmedicated soap for 10 and 30 s, handwashing with antiseptic soap for 10, 30 or 60 s, or handrubbing with an
alcohol-based disinfectant. The duration of handwashing was selected because it was the criterion
recommended in France (30 s for handwashing with
unmedicated soap, 60 s for antiseptic handwashing),
or because it was the criterion usually observed
in clinical environments (10 s). The unmedicated
soap in use was available in 1 L disposable bags.
The antiseptic soap used in the study was either
Hibiscrub1 (containing 4% chlorhexidine gluconate, Zeneca Astra, Rueil Malmaison, France) or
Betadine1 (containing 10% polividone iodine,
Asta Medica, MeÂrignac, France). The alcoholbased solution was Sterillium1 (Bode Chemie,
Hamburg, Germany), containing 1-propanol, 30%;
2-propanol, 45%; mecetronium ethyl-sulphate,
0.2%, and emollients.
The hand hygiene technique was standardized
in terms of volume of product used, duration of
application, method for drying hands with a towel
and absence of hand recontamination after drying.
The six hand hygiene techniques were typically
performed over one week by each volunteer. In
the event of two hand hygiene techniques being
277
performed on the same day by a volunteer, at least
6 h separated each experiment.
For each volunteer, a record was made of age, sex
and job. The following information was recorded for
each hand hygiene technique: hospital ward in which
the procedure was performed, presence of hand
irritation, type of healthcare procedure before hand
hygiene, and whether gloves were worn during this
procedure. Hand irritation was visually assessed
using a simplified score with four categories: absence
of hand irritation, moderate lesions, severe lesions
without hand cracking, presence of hand cracking or
skin breaks. The same infection control observer
(MPR) observed the state of hands, and collected
microbiological specimens.
Microbiological techniques
After a procedure, all five fingertips on the dominant
hand of the volunteer were pressed on a trypticasesoy agar (BioMeÂrieux, Marcy l'Etoile, France) for
15 s. In order to obtain identical conditions for each
sample, the agar was applied on to the fingers by the
observer with identical pressure. The hand hygiene
technique was then performed, and a second imprint
of the fingertips was obtained 1 min later.
Plates were incubated at 37 C under aerobic
conditions, and colony-forming units (CFUs) were
counted after 48 h. The maximum count was
300 CFUs; beyond this figure, it was considered that
there was a confluence, and a bacterial count of
350 CFUs was arbitrarily assigned to the specimen.
Potential pathogenic bacteria from transient flora
(i.e. Staphylococcus aureus, enterobacteriaceae,
aerobic Gram-negative bacteria, enterococci and
fungi) were identified using standard microbiological
techniques. Residual antiseptic activity was not
inactivated in the culture media.
Statistical analysis
Bacterial counts on hands were compared using the
log10-transformed bacterial count. In order to take
into consideration the log10-transformed bacterial
count in specimens without bacterial growth, each
bacterial count was increased by one bacterium.
Firstly, predictors of hand contamination before
the hand hygiene techniques were estimated using
a linear mixed-effects model, in order to take into
account dependency among repeated observations
from a given volunteer (SAS PROC MIXED).
A univariate analysis was performed to test the
278
J.-C. Lucet et al.
influence of several variables. A multivariate analysis
was then performed, including variables found to be
associated with hand contamination at the level of
P ˆ 0.1. A similar procedure was used to compare
hand contamination in each group of hand hygiene
techniques
Secondly, a linear model was used to compare the
differences in log10 bacterial counts after hand hygiene techniques. The model was adjusted for the
volunteer effect because of the crossover design of
the study (SAS PROC GLM). A global test was performed, and comparisons between groups were done
using Tukey's method to adjust for multiple comparisons. Several other covariates were also tested.
Statistical analysis was conducted using SAS
software (version 6.12), with a type I error (P value)
set at 0.05.
Results
A total of 43 healthcare workers took part in the
study: 26 nurses, nine nurse assistants and eight
physicians. There were 29 women and 14 men, with
a mean age of 35.5. Thirty-three HCWs were from
ICUs and 10 from medical wards.
All the HCWs performed each of the six hand
hygiene procedures. Therefore, a total of 516 specimens was obtained, 258 before and 258 after hand
hygiene. The mean SD number of CFUs (median)
before hand hygiene techniques was 81 109 (38).
Hand contamination before hand hygiene
Factors significantly associated with marked hand
contamination before hand hygiene are presented in
Table I. Hand contamination was highest in HCWs
working in a medical ward and in physicians. Hand
contamination was not significantly different after
contact with patients, after contact with patients'
environments, or after contact with body fluid secretions or waste. It was significantly higher in HCWs
without direct patient contact (mainly undertaking
housekeeping activities or rest breaks). Wearing
gloves during the procedure (N ˆ 81) significantly
reduced hand contamination. Hand contamination
was no different in HCWs with hand irritation.
In the multivariate analysis, factors associated
with marked bacterial hand contamination before
hand hygiene were (Table I): working in a medical
ward (P ˆ 0.0004), physician (P ˆ 0.02), and not
wearing gloves during the procedure (P ˆ 0.002).
Hand contamination after hand hygiene
Bacterial counts before hand hygiene procedures
were not significantly different (Table II), with an
Table I Factors associated with bacterial contamination of the hands of 43 healthcare workers before hand hygiene
Variables
Type of unit
± ICU (N ˆ 198)
± Not an ICU (N ˆ 60)
Sex
± Male (N ˆ 84)
± Female N ˆ 174)
Job title
± Nurse (N ˆ 156)
± Nurse assistant N ˆ 54)
± Physician (N ˆ 48)
Type of care
± Contact with patient (N ˆ 85)
± Contact with patient's environment (N ˆ 135)
± Contact with bodily fluid secretions or waste (N ˆ 10)
± No care (N ˆ 28)
Gloves worn during procedure
± Yes (N ˆ 81)
± No (N ˆ 177)
Skin irritation
± None (N ˆ 161)
± Moderate (N ˆ 86)
± Severe (N ˆ 11)
*Univariate analysis; ymultivariate analysis.
Log10 CFU (mean SD)
P*
Py
1.36 0.72
1.92 0.48
0.0021
0.0004
1.65 0.63
1.40 0.74
0.18
NS
1.43 0.71
1.37 0.76
1.86 0.52
0.10
0.02
1.31 0.82
1.57 0.60
1.42 0.40
1.86 0.70
0.001
NS
1.17 ‡ 0.82
1.65 0.60
0.04
0.002
1.48 0.71
1.53 0.71
1.45 0.73
0.36
NS
Hand hygiene and hand contamination
279
Table II Bacterial contamination of the hands before and after various hand hygiene techniques in 43 healthcare workers
Bacterial contamination
Handwashing with unmedicated soap, 10 s
Handwashing with unmedicated soap, 30 s
Handwashing with antiseptic soap, 10 s
Handwashing with antiseptic soap, 30 s
Handwashing with antiseptic soap, 60 s
Alcohol-based handrub
Before hand hygiene
(Log10 CFU, mean SD)
After hand hygiene
(Log10 CFU, mean SD)
1.49 0.66
1.40 0.70
1.60 0.70
1.46 0.64
1.48 0.83
1.53 0.74
0.75 0.56
0.89 0.54
0.47 0.49
0.33 0.45
0.28 0.48
0.13 0.22
average (median) of between 60 (27) and 101 (64)
CFUs among the six groups. There was a significant
effect of the hand hygiene techniques on the reduction in bacterial counts (P < 10ÿ4). For the pairwise
comparisons, the following results were found
(Table II).
Bacterial reduction after handwashing with
antiseptic soap (either 10, 30 or 60 s) or handrubbing
with the alcohol-based disinfectant was significantly
greater than that obtained after handwashing with
unmedicated soap (either 10 or 30 s). There was no
significant difference between handwashing with
antiseptic soap (either 10, 30 or 60 s) and handrubbing with the alcohol-based disinfectant.
No statistically significant difference was found
between handwashing with unmedicated soap for
10 or 30 s, although there was a trend towards greater
reduction after handwashing with unmedicated soap
for 10 s, compared with handwashing with unmedicated soap for 30 s.
A slight increase in bacterial reduction was
observed after longer handwashing with antiseptic
soap (10, 30 and 60 s), but the trend was not statistically significant. Each HCW used the same antiseptic soap in all instances for the three antiseptic
handwashing procedures: Hibiscrub in 27 HCWs,
and Betadine in 16 HCWs. Bacterial reduction was
similar in the two groups (data not shown).
Three-hundred and eighty-eight specimens were
culture-positive; 241 (94%) before and 147 (57%)
after hand hygiene. The predominant bacteria were
skin commensals: Corynebacterium sp., coagulasenegative staphylococci, Micrococcus sp. or Bacillus sp.
Bacteria from transient flora grew from 11 (4.6%)
culture-positive specimens before hand hygiene:
Staphylococcus aureus (N ˆ 3), Acinetobacter
baumannii (N ˆ 2), Pseudomonas aeruginosa (N ˆ 2),
fungi (N ˆ 2) and enterococci (N ˆ 2). The
specimens obtained after hand hygiene were still
culture-positive in two of these 11 cases, both after
handwashing with unmedicated soap (10 s in one
instance, 30 s in the other).
Discussion
This study had two aims: to assess the factors
associated with hand contamination after care, and to
measure the microbiological efficiency of different
hand hygiene techniques.
In our study, the average and median levels of
hand contamination before hand hygiene were 81 and
39 CFUs, respectively. These figures are comparable
with those of another study using the same hand
culture method, where hand contamination was
assessed after care to determine the predictive factors
of hand contamination.8
Several factors were found to be associated with
marked hand contamination. Contamination was
higher in the medical than the intensive care units.
This result is surprising since it was suggested
that compliance with handwashing was greater
in the medical unit than in intensive care.3 However,
the greater number of interventions warranting
handwashing in intensive care probably increases
handwashing in these wards, and explains this.
Hand contamination was similar for nurses and
nurse assistants, and was greater for doctors after
adjusting for other variables. These results have
been found elsewhere9, and may be explained
by undertaking fewer procedures requiring handwashings and being less compliant with the
requirement.
Hand contamination was similar after contact
with the patient, after contact with the patient
environment and after contact with body fluids or
treatment waste. It was significantly higher if there
had been no direct patient contact, probably because
in this case the hands had not been washed. As
observed in other studies, wearing gloves reduced
hand contamination.8,10
280
The other aim of the study was to assess the
efficiency of different forms of hand hygiene. A nonsignificant tendency was observed towards greater
efficiency of handwashing with an unmedicated soap
for 10 s than for 30 s. It has been shown that
a prolonged duration of handwashing with nonantiseptic soap leads to greater reduction of bacterial
counts.5 Our unusual result may be explained by the
mobilization of bacteria from the deeper layers of the
epidermis after prolonged handwashing.11 Handwashing with an antiseptic soap for 10, 30 or 60 s
produced similar reductions in the bacterial count on
hands. In an attempt to mimic clinical conditions of
use in our study, antiseptics were not inactivated on
agar for the culture of fingertips. It was hoped this
would reproduce any carry-over effect due to the
persistence of the antibacterial activity of antiseptic
soap on hands.
Several clinical studies have compared the efficiency of handwashing with an unmedicated soap
with use of an alcohol-based solution.6±8 Our results
confirm the greater efficiency of the alcohol-based
product over handwashing with an unmedicated
soap. However, to our knowledge, no clinical study
has assessed the efficiency of antiseptic handwashing
and that of alcoholic hand disinfection. The latter
was at least as efficient as antiseptic handwashing. It
is possible that a study on a larger population, and
using a more sensitive hand bacteria collection
method (e.g. the `glove juice' method) would
demonstrate a significant difference between the
two groups. Our results suggest that alcohol-based
handrubbing could replace handwashing with either
unmedicated or antiseptic soap for hand hygiene.
The qualitative assessment of bacteria on fingertips showed that only 4.6% of hands were contaminated with bacteria from transient flora before
hand hygiene. After hand hygiene, bacteria from the
transient flora were cultured in two instances, both
after handwashing with unmedicated soap. This
confirms that handwashing with unmedicated soap
does not reliably remove pathogenic bacteria from
hands.8,12,13
In conclusion, this prospective randomized study
in a clinical environment shows that alcohol-based
hand rubbing and handwashing with antiseptic soap
are more effective than handwashing with unmedicated soap, in reducing the bacterial count on hands.
The need to reduce carriage of pathogenic bacteria to
a minimum suggests that the use of alcohol-based
hand disinfectants should be preferred to unmedicated soaps, and possibly to antiseptic soaps.
J.-C. Lucet et al.
Acknowledgements
We thank the staff of the Bichat-Claude Bernard who
volunteered to participate in the study. This study
was supported by a grant from Rivadis (Thouars,
France) and Bode Chemie (Hamburg, Germany).
References
1. Larson E. Skin hygiene and infection prevention: more
of the same or different approaches? Clin Infect Dis
1999; 29: 1287±1294.
2. Pittet D. Improving compliance with hand hygiene
in hospitals. Infect Control Hosp Epidemiol 2000; 21:
381±386.
3. Pittet D, Mourouga P, Perneger TV. Compliance with
handwashing in a teaching hospital. Infection Control
Program. Ann Intern Med 1999; 130: 126±130.
4. Pittet D, Hugonnet S, Harbarth S et al. Effectiveness
of a hospital-wide programme to improve compliance
with hand hygiene. Lancet 2000; 356: 1307±1312.
5. Rotter ML. Hand washing and hand disinfection. In:
Mayhall CG, ed. Hospital Epidemiology and Infection
Control. Baltimore: Williams & Wilkins; 1996:
1052±1068.
6. Larson EL, Eke PI, Laughon BE. Efficacy of alcoholbased hand rinses under frequent-use conditions.
Antimicrob Agents Chemother 1986; 30: 542±544.
7. Zaragoza M, Salles M, Gomez J, Bayas JM, Trilla A.
Handwashing with soap or alcoholic solutions?
A randomized clinical trial of its effectiveness.
Am J Infect Control 1999; 27: 258±261.
8. Pittet D, Dharan S, Touveneau S, Sauvan V, Perneger
TV. Bacterial contamination of the hands of hospital
staff during routine patient care. Arch Intern Med
1999; 159: 821±826.
9. Daschner FD. The transmission of infections in
hospitals by staff carriers, methods of prevention and
control. Infect Control 1985; 6: 97±99.
10. Tenorio AR, Badri SM, Sahgal NB et al. Effectiveness
of gloves in the prevention of hand carriage of
vancomycin-resistant enterococcus species by health
care workers after patient care. Clin Infect Dis 2001; 32:
826±829.
11. Meers PD, Yeo GA. Shedding of bacteria and skin
squames after handwashing. J Hyg (Cambr.) 1978; 81:
99±105.
12. Ayliffe GA, Bridges K, Lilly HA, Lowbury EJ,
Varney J, Wilkins MD. Comparison of two methods
for assessing the removal of total organisms and
pathogens from the skin. J Hyg (Cambr.) 1975; 75:
259±274.
13. Ehrenkranz NJ, Alfonso BC. Failure of bland soap
handwash to prevent hand transfer of patient bacteria
to urethral catheters. Infect Control Hosp Epidemiol
1991; 12: 654±662.