A close look at alcohol gel as an

antimicrobial sanitizing agent

Daryl S. Paulson, PhDa
Eleanor J. Fendler, PhDb
Michael J. Dolan, BSb
Ronald A. Williams, MSb
Bozeman, Montana, and Cuyahoga Falls, Ohio

Background: Hand transmission of microbes by health care workers is a primary cause of nosocomial infections in both long-term and acute care
facilities. Compliance with effective handwashing and hand sanitization regimens can break this cycle.
Methods: We investigated the antimicrobial efficacy and irritation potential of 5 handwash product regimens: a nonantimicrobial lotion soap, an
antimicrobial lotion soap, an alcohol gel sanitizer, a nonantimicrobial lotion soap with an alcohol gel sanitizer, and an antimicrobial lotion soap with
an alcohol gel sanitizer. The regimens were evaluated by using a Healthcare Personnel Handwash procedure, and irritation was assessed by using
expert hand evaluation after 25 consecutive washes.
Results: The Healthcare Personnel Handwash data showed that the mean log reductions from baseline were greatest for the lotion soaps with alcohol
gel sanitizer, less for the alcohol and the antimicrobial soap alone, and least for the bland soap. All of the product regimens showed a low potential for
skin irritation.
Conclusion: In terms of both microorganism reduction and skin irritation, the most effective product regimens were the use of alcohol gel sanitizer
in combination with either an antimicrobial or a plain lotion soap. (AJIC Am J Infect Control 1999;27:332-8)

The potential for health care personnel to be a vector
in transmitting hospital-acquired disease continues to
be significant. In addition, the number of situations in
which health care personnel have been implicated in
contributing to hospital illness continues to increase,
and various types of hospital-acquired diseases may be
spread through a health care facility via this mode of
transmission.1,2
The microorganisms that normally colonize the hand
surfaces pose little threat of infectious disease trans-
mission from health care personnel to patients who are
not immunocompromised. Instead, the threat comes
primarily from “transient” pathogenic microorganisms,
which contaminate the hand surfaces. In the health
care industry, contamination usually occurs from con-
tact with excretions or infectious exudates and is most
commonly transmitted through hand contact.1,2 The
most significant transient bacteria include strains of
From the BioScience Laboratories, Inc, Bozeman,a and GOJO
Industries, Inc, Cuyahoga Falls.b
Supported in part by GOJO Industries, Inc.
Reprint requests: Eleanor J. Fendler, PhD, GOJO Industries, Inc,
3783 State Road, Cuyahoga Falls, OH 44223.
Copyright © 1999 by the Association for Professionals in
Infection Control and Epidemiology, Inc.
0196-6553/99/$8.00 + 0 17/46/96194
susceptible and antibiotic-resistant Staphylococcus
aureus, Enterococcus faecium, Enterococcus faecalis,
Escherichia coli, and species of Pseudomonas,
Streptococcus, Clostridium, and Proteus.
Nosocomial disease outbreaks result in staggering
financial costs and necessitate extreme regulatory
processes to ensure safety in hospitals and other health
care service industries.3-6 The most effective way to
break the contamination cycle between health care per-
sonnel and patients is for personnel to perform a thor-
ough handwash by using an effective antimicrobial
hand cleanser.1,2,7,8
There is some confusion as to the actual antimicro-
bial effectiveness of commercially available products.
The purpose of this study was to aid infection control
professionals in selecting the most appropriate hand-
cleansing products and procedures to meet their need.
Before evaluating any hand cleanser, it is critical that
a valid experimental design, including appropriate test-
ing and sampling methods be developed.9
Three hand-sampling methods commonly used to
evaluate hand cleanser effectiveness are the “swab,”
“finger press,” and glove juice techniques. In brief, the
swab sampling method consists of swabbing the palmar
hand surfaces and between the fingers or on fingertips
with a premoistened cotton swab and culturing on an
agar plate. The finger press method is performed by
having the test subjects press their palmar surfaces or
fingertip pads onto an agar plate. However, the third

332
AJIC
Volume 27, Number 4
method (glove juice) is the only method specified by
regulatory agencies in the United States for evaluating
hand antiseptic products.10 This method is performed
by having the subject don latex surgical gloves, instill-
ing a microbial stripping solution into the glove, and
then massaging the hands and fingers through the glove
for 1 minute.11
The major problem with the first 2 hand-sampling
methods is lack of reliability. Neither method is as accu-
rate and precise in measuring known quantities of
marker microorganisms as is the glove juice method.12
In this context, accuracy refers to the ability of the sam-
pling method to estimate true microbial population.
Precision is the ability of the sampling method to accu-
rately estimate the true microbial population repeatedly.
In a study comparing these 3 hand-sampling meth-
ods at hand contamination levels of 102 through 108, the
glove juice method was the only technique that accu-
rately measured all of the contamination levels. Both
the swab and finger press methods routinely underesti-
mated the actual contamination populations and were
also imprecise in repeated estimations of known micro-
bial population numbers.12 Hence, the method of
choice is the glove juice sampling procedure.
There has been some argument concerning the need
to sample the entire hand when mainly the fingers are
involved in inadvertent contamination, particularly
with fecal matter after defecation. Several researchers
have advocated use of a fingertip wash and have shown
that results are valid and precise.13,14 However, when
trying to ascertain the reliability of measuring a known
number of microorganisms on the fingertips, analysis
of the finger press method described earlier would sug-
gest that the sampling variability would be unaccept-
able. One hundred-fold accuracy and precision errors
were not uncommon.12 In addition, regarding fecal con-
tamination, once the fingers have been contaminated,
the contaminants are quickly disseminated over the pal-
mar surfaces via finger and hand movement, so whole-
hand sampling is not unreasonable. Hence, it was con-
cluded that the entire hand surface should be sampled.
Another problem in evaluating hand disinfectants is
that to evaluate a product’s antimicrobial efficacy, the
number of transient microorganisms existing on the
hands must be know and not confused with the normal
or resident microbial populations. Recall that the
microorganisms encountered on the hand surfaces fall
into 2 categories. The first consists of transient
microorganisms “picked up” from the environment by
health care personnel. These microorganisms common-
ly reside on the hand surfaces only temporarily. The
second category (resident bacteria) consists of the
microorganisms that reside permanently on the hand
surfaces (ie, the normal skin flora).
Paulson et al 333
Because it is nearly impossible to quantify the micro-
bial populations of either of these categories reliably, it is
simpler and more effective to use a marker microorgan-
ism as the transient microorganism. In this study, Serratia
marcescens was used as a marker because it produces a
characteristic red colony that is easily distinguishable
from normal microbial populations of the hands, as well
as from other contaminating microorganisms. The use of
S marcescens provides a clear indication of the effective-
ness of the handwash without confounding the test by
mixing transient and normal microorganisms.
The use of S marcescens is appropriate in several other
ways. First, it is not normally a pathogenic or disease-
producing microorganism, so there is little risk of infec-
tion to human volunteer subjects. Second, it is resistant
to mechanical removal from the hands in much the same
way as are potential pathogens such as E coli,
Staphylococcus species, and Enterococcus species. Third,
because it is seeded onto the hands at a known popula-
tion and the baseline levels for all subjects will be equiv-
alent, the antimicrobial properties of the handwash
products can be compared easily and directly.
In evaluating handwash products designed mainly to
remove transient microorganisms, 2 parameters of
interest are the immediate and persistent antimicrobial
effects. The immediate antimicrobial effect is a mea-
sure of both the mechanical removal (if a handwash is
used) and the immediate inactivation of the microor-
ganisms by the antimicrobial compound (eg, alcohol,
triclosan, parachlorometaxylenol [PCMX]), if one is
used in the product. The persistent effect is a measure
of how well the product’s antimicrobial component pre-
vents microbial recolonization of the skin surfaces,
either by inhibition or lethality.
In brief, nonantimicrobial soaps are useful
degerming agents because of their ability to enhance
mechanical removal of transient microorganisms dur-
ing a warm water handwash. Antimicrobial soaps, such
as surgical scrubs and health care personnel hand-
washes containing chlorhexidine digluconate, triclosan,
or PCMX, have immediate antimicrobial effects as a
result of both mechanical degerming during the wash
and the microbicidal activity of the antimicrobial com-
pound. They also have persistent antimicrobial activity
as a result of specific retention of the antimicrobial
compound on the hands. However, an important
caveat: even if a hand-cleansing product is effective in
removing transient microorganisms and preventing
regrowth, it also must be mild to the skin after many
consecutive applications. If not, health care personnel
are unlikely to comply with handwashing regimens.
Alcohol and alcohol gels in which alcohol levels exceed
50% have the most pronounced immediate antimicrobial
effects. However, they lack persistent antimicrobial prop-

334 Paulson et al
Table 1. Hand-cleansing product configurations
Configuration Regimen Active ingredient
1 Nonantimicrobial lotion soap No active
2 Antimicrobial lotion soap 0.6% PCMX
3 Alcohol gel sanitizer 62% Ethanol
4 Nonantimicrobial lotion soap No active and
and alcohol gel sanitizer 62% ethanol
5 Antimicrobial lotion soap and 0.6% PCMX
alcohol gel sanitizer and 62%
ethanol
erties. Alcohol gels, which are used without a water
wash, have excellent immediate antimicrobial effects.
APIC guidelines15 recommend the use of an alcohol-
based handrub for hand antisepsis in cases where the
hands are not visibly soiled. In the case of heavy soilage
or contamination with blood or other organic material,
hand-cleansing with a detergent-based product is recom-
mended before use of the alcohol gel handrub.
This study was designed to compare the efficacy of dis-
infection among 5 handwash products produced by
GOJO Industries, Inc, (Cuyahoga Falls, Ohio): a nonan-
timicrobial lotion soap (DermaPro Lotion Skin
Cleanser); an antimicrobial lotion soap (DermaPro
Antimicrobial Lotion Soap with 0.6% PCMX; an alcohol
gel sanitizer (Instant Antiseptic Hand Sanitizer with 62%
Ethyl Alcohol); the nonantimicrobial lotion soap used in
conjunction with the alcohol gel sanitizer; and the
antimicrobial lotion soap used in conjunction with the
alcohol gel sanitizer (Table 1). The immediate and per-
sistent antimicrobial efficacy of these 5 handwash prod-
uct configurations, as well as the degree of irritation to
the hands on repeated use were investigated. Both
PCMX-based antimicrobial soaps and waterless alcohol-
based hand sanitizers are in common use in the health
care industry, and we chose to evaluate these alone and
in combination. Future studies may evaluate triclosan or
chlorhexidine gluconate-based products in a similar way.
METHODS
Handwash configurations
The 5 product configurations used for this study are
described in Table 1.
Subjects
Twenty-five human subjects older than the age of 18
years but younger than the age of 70 years were recruit-
ed for this study. Subjects were of both sexes, and all
were free of clinically evident dermatosis or injuries to
the hands and forearms. No known immunocompro-
mised subjects were admitted into this study. Subjects
were randomly assigned 1 of the 5 product configura-
tions, 5 subjects per product configuration.
AJIC
August 1999
Marker microorganism
A 24-hour culture of S marcescens (American Type
Culture Collection #14756) with a population of
approximately 1.0 × 108 colony-forming units per mL
was used as the “marker” contaminant microorganism.
Because S marcescens colonies appear red on the tryp-
tic soy agar plates, they were clearly differentiated from
normal and transient microorganisms found on the
plate that, if counted, would have confounded the sam-
pling data. Only red colonies were tallied.
Neutralization
A neutralization assay was performed on each of the
test products to ensure that the neutralizers used in the
diluting and plating media inactivated the antimicro-
bial compounds.
Pretest period
The first 7-day period of this study was designated as
the “pretest period.” During this period, subjects avoid-
ed use of medicated soaps, lotions, detergents, acids,
and bases. Bathing in biocide-treated pools and hot
tubs, as well as use of ultraviolet light tanning beds, was
prohibited.
Experimental period
The second 7-day period (after the pretest period)
constituted the experimental period. Each subject was
used 1 day of that week for approximately 6 hours. On
entering testing, a 5 mL aliquot of the S marcescens cul-
ture was pipetted into each subject’s cupped hands. The
subject then distributed the inoculum evenly over both
hands by gentle massage. After a 1-minute air dry, the
Glove Juice Sampling Procedure was performed. This
first microbial inoculation/sampling procedure consti-
tuted the baseline sample and ensured that all of the
microorganisms inoculated onto the subjects’ hands
could be removed. It was followed by a handwash with
a nonantimicrobial soap. Subjects’ hands were reinocu-
lated with the microorganisms, and the hand-cleansing
product configuration, randomly assigned, was per-
formed. This microbial inoculation/assigned product-
use procedure was repeated 10 consecutive times by a
subject, with a minimum of 5 minutes between cycles.
Assays for transient “marker” microorganisms were
performed by using the Glove Juice Sampling
Procedure (see later) after washes 1, 3, 7, and 10.
Wash procedure
For the handwashing products, 3 mL aliquots were
dispensed into a subject’s hands. The subject then
washed the hands and one third of the forearms for 20
seconds, rinsed for 30 seconds with lukewarm water,
AJIC
Volume 27, Number 4
and allowed the hands and forearms to air dry thor-
oughly. The alcohol gel sanitizer (3 mL) was applied to
dry hands, rubbed to distribute evenly over the hands
and one third of the forearms, and allowed to air dry for
5 minutes before sampling. No water-wash was used in
conjunction with the test configuration specifying alco-
hol gel product alone.
Glove Juice Sampling Procedure
After the prescribed wash and rinse (or the drying of
the alcohol gel alone), nonpowdered, sterile surgical
gloves were donned. Seventy-five milliliters of a sterile
aqueous phosphate buffer (pH 7.8) solution containing
0.1% Triton X-100 (Union Carbide Corp, Danbury,
Conn) were instilled into the glove. The glove was
secured at the wrist, and the hand was massaged
through the glove for 60 seconds. Aliquots of the “glove
juice” were removed and serially diluted in Bacto
Tryptic Soy Broth (Soybean-Casein Digest Medium,
Difco Laboratories, Detroit, Mich) containing 1.0%
polysorbate 80 and 0.3% lecithin as neutralizing agents.
Duplicate Bacto Tryptic Soy Agar spread plates con-
taining 1.0% polysorbate 80 and 0.3% lecithin were pre-
pared for the 10-4, 10-5, and 10-6 dilutions (baseline sam-
ples) or the 10-0, 10-1, 10-2, and 10-3 dilutions (posttreat-
ment samples). The plates were incubated at 30 ± 2°C
until a red pigment was distinguishable in the colonies
on the surface of the agar (24 to 48 hours). Counts from
plates providing between 25 and 250 red-pigmented
colonies were used in this study. The number of viable
red-pigmented bacteria recovered was determined by
using the formula:
B = A [(ΣCi/n]10-D
Where: B = estimated number of microorganisms
A = glove juice volume = 75 mL
[ΣCi/n] = average colony-forming unit count
Ci = colony-forming unit count from a spread plate
n = number of replicate plates = 2
D = dilution level
Because data collected in this study were exponen-
tial, each B value was linearized by transforming it to
the log10 scale. This linearization process was a require-
ment of the statistical models employed.
Irritation study
After microbial inoculation/product-use cycle 10, the
subjects entered the irritation potential phase of the
study. The subjects continued to wash their hands with
their assigned product configuration for 15 consecutive
washes or product applications (a total of 25 for the
study). There was a minimum of 5 minutes and a max-
imum of 15 minutes between wash or product applica-
tion procedures. The subjects’ hands were examined for
irritation after washes 15, 20, and 25, or at the first sign
Paulson et al 335
Table 2. Hand irritation scoring system
Irritation Scoring
Erythema (redness) 0 = No reaction
1 = Mild or transient redness limited to
sensitive areas
2 = Moderate redness persisting over
much of the product-exposed area
3*= Severe redness extending over most
or all of the product-exposed area
Edema (swelling) 0 = No reaction
1 = Mild (just perceptible)/transient
2 = Moderate—definitely palpable
3*= Severe
Rash 0 = No reaction
1 = Mild—few, small eruptions
2 = Moderate—scattered eruptions >10
per hand
3 = Severe
Dryness (chafing) 0 = No reaction
1 = Mild—transient, generally limited to
cuticles, knuckles
2 = Moderate—persistent, extending
over much of the hand
3*= Severe—persistent, extending over
most of the hand, characterized by
cracking and roughness
*Represents significant irritation and requires subject’s removal from
study.
of irritation. Degree of irritation was scored according
to standardized criteria (Table 2) by a single, blinded
(unaware of subject’s product configuration) evaluator
trained in assessment of skin irritation.
Subject safety
On completion of the final product application (#5)
and the follow-up examination for skin irritation, all
subjects washed their hands and arms with 70% iso-
propyl alcohol for 2 minutes and rinsed under warm
tap water to eliminate any S marcescens that may have
remained on the hands.
RESULTS
The t test was used in this evaluation to determine the
level of significance between the baseline measurements
and configuration at a specific sample time. Because mul-
tiple 95% confidence intervals were used, the t test table
values were adjusted for multiple estimates by using a
correction factor described by Dixon and Massy.16 The α
value is modified to α* where α* = 1 – (1–α)k, and k is the
number of confidence intervals computed.
In addition, whereas the power of any statistic is less
than use of a greater sample size, if significant differ-
ence did occur, then it is even more probable that there
 AJIC
336 Paulson et al August 1999

Table 3. Log and percent reduction values from baseline

Mean log10 reduction from baseline (percent reduction)

Product Mean log10
configuration baseline Wash 1 Wash 3 Wash 7 Wash 10

1 7.98 ± 0.27 2.29 ± 0.52 (99.49) 2.09 ± 0.63 (99.19) 1.96 ± 0.72 (99.09) 2.09 ± 0.68 (99.19)
2 7.98 ± 0.27 2.50 ± 0.22 (99.68) 2.73 ± 0.41 (99.81) 2.91 ± 0.63 (99.88) 2.76 ± 0.27 (99.83)
3 7.98 ± 0.27 3.93 ± 0.04 (99.99) 3.79 ± 0.15 (99.98) 2.96 ± 0.17 (99.89) 2.15 ± 0.13 (99.29)
4 7.98 ± 0.27 3.27 ± 0.14 (99.95) 3.70 ± 0.18 (99.98) 3.64 ± 0.23 (99.98) 3.62 ± 0.29 (99.98)
5 7.98 ± 0.27 3.28 ± 0.13 (99.95) 4.13 ± 0.23 (99.99) 3.74 ± 0.27 (99.98) 3.65 ± 0.17 (99.98)
1, Nonantimicrobial lotion soap; 2, antimicrobial lotion soap; 3, alcohol gel sanitizer; 4, nonantimicrobial lotion soap with alcohol gel sanitizer; 5, antimicro-
bial lotion soap with alcohol gel sanitizer.

really was a significant difference. It should be remem-
bered that use of small sample sizes tends to promote
increased type II errors, not type I errors. Recall type II
error is stating that a difference does not exist when one
actually does. A type I error is the probability of stating
there was a difference when none existed. In certain
cases, a ranking and ordering statistical procedure was
used to determine whether nonsignificant differences
between product configurations calculated by using the
consecutive t test multiple adjustment factor actually
could be significant.17
Exploratory Data Analysis also was used to ensure
sample size normality (ie, no excessive skewing, no
bimodal samples, no outliers, etc) was attained. Stem-
Leaf Ordering, Letter Value displays, and Box-Plots
with notches were used.18
Table 3 presents the data in terms of both log10 reduc-
tion and percent reduction from baseline. As can be
seen, each of the 5 product configurations reduced the
baseline population by at least 2 logs, highly statistically
significant results (P < .05).
Nonantimicrobial lotion soap
The nonantimicrobial lotion soap reduced the base-
line microbial counts by about 2 logs (99.0%) after the
first wash (P < .05). Throughout the course of 10 micro-
bial inoculation/wash cycles, the contaminating
microorganism counts remained down from baseline
by nearly 2 logs (99%) (P < .05).
Antimicrobial lotion soap
The antimicrobial lotion soap reduced the initial
microbial counts by approximately 2.5 logs (99.68%)
after the first inoculation/wash cycle and demonstrated
antimicrobial consistency throughout the 10 consecu-
tive inoculation/wash cycles (P < .05).
Alcohol gel sanitizer
The alcohol gel sanitizer achieved an approximately
4-log (99.99%) reduction in microorganisms from base-
line levels after the first inoculation/product application
cycle and maintained a greater than 2 log reduction (P
< .05) after inoculation/application cycle 10. The slight
buildup in microorganism populations throughout the
course of 10 applications was an artifact of the test pro-
tocol, because the product is applied without a water
wash.
Nonantimicrobial lotion soap with alcohol gel
sanitizer
When used in combination, the nonantimicrobial
lotion soap and the alcohol gel sanitizer reduced counts
about 3.25 logs (99.94%) from baseline levels after the
first inoculation/product use cycle. After inoculation/
product-use cycles 3, 7, and 10, the microbial counts
were down from baseline by greater than 3.5 logs
(99.97%; P < .05).
Antimicrobial lotion soap with alcohol gel sanitizer
Used in conjunction with the alcohol gel sanitizer, the
antimicrobial lotion soap produced about a 3.25 log
(99.94%) reduction from baseline after the first inocu-
lation/product-use cycle, and counts remained at about
3.5 logs (99.97%) below baseline for the remainder of
the 10 inoculation/product-use cycles (P < .05).
Irritation
No significant skin irritation was observed through-
out the 25 consecutive wash/product application cycles
with the nonantimicrobial lotion soap, the antimicro-
bial lotion soap, or the alcohol gel sanitizer. Neither of
the soaps used in conjunction with the alcohol gel san-
itizer produced any significant skin irritation through-
out the course of the 25 handwashes.
DISCUSSION
Although each of the 5 product configurations was
effective in reducing transient microbial population lev-
els on the hands, the overall results illustrate the com-
plexity involved in setting an effective hand hygiene reg-
imen for the health care environment. Each configura-
tion has associated advantages and disadvantages.
AJIC
Volume 27, Number 4
Results from use of the nonantimicrobial soap illus-
trate the generally sound principle of standard hand-
washing for hygienic purposes. This test configuration
showed that mechanical removal of transient skin bac-
teria is relatively effective and consistent, but perhaps
not optimal. Importantly, no skin irritation was found
from use of this mild product intended for frequent
wash situations.
The antimicrobial lotion soap was not found statisti-
cally more effective than the plain soap in both imme-
diate and persistent antimicrobial properties by using a
t test (P < .05). However, by using a rank-ordering pro-
cedure found in Exploratory Data Analysis, it was con-
sistently different; that is, the antimicrobial soap was
consistently more effective than the plain soap and
water. A properly formulated, general purpose antimi-
crobial soap can exhibit both effective removal/kill of
transient bacteria and low skin irritation and thus can
be considered an attractive alternative to the nonan-
timicrobial soaps.
Again, relying on an Exploratory Data Analysis
selecting and ordering model the alcohol gel sanitizer
was demonstrated to be the most effective single prod-
uct for reducing transient microorganisms. And again,
this conclusion was based on consistently more effec-
tive results observed with the alcohol gel configuration.
Although the alcohol gel maintains significant log
reductions throughout a series of hand-cleansing pro-
cedures, the data’s upward trend (microbial rebound)
suggests that the product be supplemented by use of an
antimicrobial or nonantimicrobial handwash product
after 3 to 5 consecutive uses.
Visually, the alcohol gel sanitizer demonstrated low
skin irritation potential, comparable with that of the
plain soap. Other studies have shown that use of an
alcohol gel sanitizer helped maintain normal skin
hydration levels and reduced the irritation induced by
frequent handwashing.19,20
LIMITATIONS
This study provides health care professionals with
information and guidelines to help choose effective
hand-cleansing products and regimens that reduce the
risk of nosocomial infections. However, it is the respon-
sibility of the actual user to ensure that the products are
used appropriately and regimens are adhered to as
appropriate to each situation.
A potential limitation of this study is that it was con-
ducted in a laboratory setting, rather than in actual
field conditions. However, the correlation between the
glove juice test results and actual field results has been
established by studies in the health care industry. The
fact remains that glove juice testing is the sampling
methodology specified for testing products for use in
Paulson et al 337
the healthcare setting in the United States.21 Further,
any study, field or otherwise, must employ valid tech-
niques before accurate conclusions can be drawn. The
glove juice procedure has been shown to be more accu-
rate and precise than other sampling methods.
Finally, because the study results were based on
small sample sizes, precision was not as great as would
be seen for a study with a greater sample size.
CONCLUSION
The most effective product regimens, from both an
overall microorganism reduction profile and skin irrita-
tion potential, were the combinations of alcohol gel
with either an antimicrobial or a plain lotion soap. Both
of these configurations produced dramatic immediate
reductions in transient organisms as a result of
mechanical degerming by the soap handwash, coupled
with the alcohol’s tremendous microbicidal proper-
ties.19 Moreover, both combination regimens demon-
strated excellent persistent antimicrobial properties.
Effective handwashing is a cornerstone in preventing
nosocomial illness as a result of contamination by
health care personnel. Key issues are selection of effec-
tive hand-cleansing products, product-use regimens,
and well-managed handwashing/sanitizing training
programs, which include compliance measurement.
Ultimately, failure to address these key issues unques-
tionably risks the health of the patient.
References
1. Larson E. A causal link between handwashing and risk of infec-
tion? Infect Control Hosp Epidemiol 1988;9:28-36.
2. Bryan JL, Cohran J, Larson EL. Handwashing: a ritual revisited.
In: Rutala WA, editor. Chemical germicides in health care.
Washington (DC): Association for Professionals Infection Control
and Epidemiology; 1995. p. 163-78, 211-9.
3. Platt R, Goldmann DA, Hopkins CC. Epidemiology of nosocomial
infections. In: Gorbach SL, Bartlett JG, Blacklow NR. editors.
Infectious diseases. Philadelphia: WB Saunders Co.; 1992. p. 96-
106.
4. Beck-Sague C, Davis WR. The epidemiology and prevention of
nosocomial infections. In: Block SS, editor. Disinfection, steriliza-
tion, and preservation. Philadelphia: Lea & Fegier; 1991. p. 663-
75.
5. Wakefield DS. Understanding the cost of nosocomial infections.
In: Wenzel RP, editor. Prevention and control of nosocomial infec-
tions. Baltimore: Williams & Wilkins; 1993. p. 21-41.
6. Nettleman MD. The global impact of infection control. In: Wenzel
RP, editor. Prevention and control of nosocomial infections.
Baltimore: Williams & Wilkins; 1993. p. 13-20.
7. Doebbeling BN, Stanley GL, Sheetz CT, et al. Comparative effica-
cy of alternative hand-washing agents in reducing nosocomial
infections in intensive care units. N Engl J Med 1992;327:88-93.
8. Goldmann D, Larson E. Hand-washing and nosocomial infections.
N Engl J Med 1992;327:120-2.
9. Paulson DS. A broad-based approach to evaluating topical antimi-
crobial products. In: Asenzi JM, editor. Handbook of disinfectants
and antiseptics. New York: Marcel-Dekker; 1996. p. 17-33.
10. 21 CFR Sect 333.470 (1994).
 AJIC
338 Paulson et al August 1999

11. Paulson DS. Evaluation of three handwash modalities commonly
employed in the food industry. Dairy Food Environ Sanitation
1992;12:615-8.
12. Paulson DS. Evaluation of three microorganism recovery proce-
dures used to determine handwash efficacy. Dairy Food Environ
Sanitation 1993;13:520-3.
13. Maner F, Rotter M, Mittermayer H. Comparative investigation on
recovering bacteria from artificially contaminated hands.
Zentralbl Bakteriol [Orig B] 1975;160:412-31.
14. Rotter M, Wewalka G, Koller W. Influence of some variables on
the results of evaluations of procedures for hygienic hand disin-
fection. Hyg Med 1982;7:157-66.
15. Larson EL. APIC guideline for handwashing and hand antisepsis
in health care settings. AJIC Am J Infect Control 1995;23:251-69.
16. Dixon WJ, Massy FJ. Introduction to statistical analysis. 4th ed.
New York: McGraw-Hill; 1983. p. 131-3.
17. Gibbons JD, Olkin I, Sobel M. Selecting and ordering populations.
New York: John Wiley; 1977. p. 14-57.
18. Velleman PF, Hoaglin DC. Applications, basics and computing of
exploratory data analysis. Boston: Duxbury; 1981. p. 3-121.
19. Larson EL, Morton HE. Alcohols. In: Block SS, editor.
Disinfection, sterilization, and preservation. Philadelphia: Lea &
Febiger; 1991. p. 191-203.
20. Newman JL, Seitz JC. Intermittent use of an antimicrobial hand
gel for reducing soap irritation of health care personnel. AJIC Am
J Infect Control 1990;18:194-200.
21. Health Care Personnel Handwash Procedure, 21 CFR Part 333.470
(1994).

Availability of Journal back issues

As a service to our subscribers, copies of back issues of AJIC: American Journal of

Infection Control for the preceding 5 years are maintained and are available for purchase
from the publisher, Mosby, at a cost of $15.00 per issue until inventory is depleted. The fol-
lowing quantity discounts are available: 25% off on quantities of 12 to 23, and one third
off on quantities of 24 or more. Please write to Mosby, Inc, Subscription Services, 11830
Westline Industrial Dr, St Louis, MO 63146-3318, or call 800-453-4351 or 314-453-4351 for
information on availability of particular issues. If unavailable from the publisher, photo-
copies of complete issues may be purchased from Bell & Howell Information and
Learning, 300 N Zeeb Rd, Ann Arbor, MI 48106-1346, 734-761-4700.