- Email: [email protected]
Comparison of the antibacterial efficacy and acceptability of an alcohol-based hand rinse with two alcohol-based hand gels during routine patient care
Journal of Hospital Infection (2007) 66, 167e173
www.elsevierhealth.com/journals/jhin
Comparison of the antibacterial efficacy and acceptability of an alcohol-based hand rinse with two alcohol-based hand gels during routine patient care ´de ´ric Barbut a,*, Eric Maury b, Laurianne Goldwirt a, Fre ´rre-Yves Boe ¨lle c, Denis Neyme a, Rubina Aman b, Pie Beatrice Rossi b, Georges Offenstadt b a
Assistance Publique-Hoˆpitaux de Paris, Hoˆpital Saint-Antoine, Unite´ d’Hygie`ne et de Lutte contre les Infections Nosocomiales b Service de Re´animation me´dicale c Unite´ de Biostatistiques, INSERM U707 Received 6 March 2007; accepted 20 March 2007 Available online 21 May 2007
KEYWORDS Hand hygiene; Alcoholbased rinse; Alcoholbased gels; Compliance; Antibacterial efficacy
Summary The aims of this study were to compare the antibacterial efficacy of handrubbing with an alcoholic rinse (AHRR) and two different alcoholic gels (AHRG) in reducing hand contamination under practical use conditions. We wanted to assess the acceptability of the three products and to determine the effect of each product on overall hand hygiene compliance. A prospective alternating time-series clinical trial was performed in a medical intensive care unit. The study was divided into three six-week periods (P1, P2, P3). Handrubbing was achieved with Sterillium rinse (AHRR) during P1, Sterillium gel (AHRG-1) during P2 and Manugel Plus (AHRG-2) during P3. Pre- and post-rubbing hand contaminations were assessed immediately after a direct contact with a patient, using the glove juice technique. Health care workers (HCWs) evaluated the acceptability of the products through a self-administered anonymous questionnaire. Compliance of HCWs with hand hygiene was assessed during the three periods. We studied 242 handrubbing opportunities. The mean reduction factor (expressed as the Log10 CFU/mL) of the AHRR, AHRG-1 and AHRG-2 were 1.28 0.95, 1.29 0.84 and 0.51 0.73, respectively (p < 0.001). Assessment of the three products by HCWs indicated that
* Corresponding author. Address: UHLIN, Ho ˆpital Saint-Antoine, Assistance Publique-Ho ˆpitaux de Paris, 184 rue du faubourg SaintAntoine, 75012 Paris 12, France, Tel.: þ33 1 49 28 30 08; fax: þ33 1 49 28 30 09. E-mail address: [email protected] 0195-6701/$ - see front matter ª 2007 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2007.03.021
168
F. Barbut et al. AHRR and AHRG-1 were significantly better accepted than AHRG-2. The overall compliance of HCWs to hand hygiene was better when gel was available. Under practical use conditions, AHRG-1 and AHRR were more effective than AHRG-2, although all were claimed to pass the European standard EN1500. In vivo trials are essential to compare the antimicrobial efficacy of products for handrubbing. ª 2007 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction Hand hygiene is considered to be the most important measure to prevent cross transmission of microorganisms, and thus to prevent nosocomial infections.1,2 However, under routine hospital conditions, compliance with standard handwashing is unacceptably low, less than 50% in most studies published in the past 20 years.3e6 Guidelines now promote the use of alcoholic handrubbing in substitution of handwashing whatever the soap used, medicated or not.7 Alcoholic handrubbing is faster and easier to use due to ease of access at the bedside. It is more effective than handwashing with plain or antimicrobial soap, and is also better tolerated. Recent reports have clearly shown that the use of hand rinse is associated with a better compliance to hand hygiene.7e9 Pittet et al. have found an association between an increasing hand rub consumption following an intensive campaign for handrubbing and a decrease of both the incidence of nosocomial infections and the attack rate of meticillin-resistant Staphylococcus aureus (MRSA).2 Alcohol-based gels have been recently introduced in many hospitals worldwide for hand hygiene and seem to be a promising alternative to alcohol-based rinse in term of acceptability.10,11 Nevertheless, some reports using the European standard EN1500 suggested that gels might be less effective than rinses, especially because they usually contain less alcohol than rinses.12,13 This has led to controversy about the use of alcohol-based gels in hospitals and the significance of the European standard EN1500. Further studies are warranted to determine the relative efficacy of alcohol-based gels and rinses in daily clinical practice where duration and quality of application usually do not reach those required in the EN1500 standard.14 The objectives of the present study were: i) to compare the antibacterial efficacy of handrubbing with an alcoholic rinse versus handrubbing with two different alcoholic gels during ICU routine care
activities; ii) to assess the acceptability of the different products by health care workers (HCWs); and iii) to determine the effect of each product on the overall hand hygiene compliance.
Materials and methods Setting We conducted a prospective alternating timeseries trial from November 19th 2004 to March 11th 2005 in a 14-bed medical intensive care unit (MICU) of a tertiary care and university hospital (760 beds). In our hospital, alcohol-based rinses were introduced in daily clinical practice in 2000 with a hospital-wide campaign launched by the infection control committee. An additional mandatory teaching on hand hygiene was provided in September 2004 to all medical and non-medical HCWs.
Enrolment of HCWs for microbiological efficacy of hand hygiene Three consecutive six-week periods were defined (P1, P2, P3). During each period only one product was available for handrubbing. HCWs used handrubbing with Sterillium rinse (AHRR) (45% 2-propanol [isopropanol]; 30% 1-propanol; 0.2% mecetronium ethyl-sulphate, Bode Chemie, Hamburg, Germany distributed by Rivadis, Thouars, France) during P1, Sterillium gel (AHRG-1)(85% ethanol, Rivadis) during P2 and Manugel Plus (AHRG-2) (ethanol 53%, isopropanol 17%, Anios, Lille Hellemmes, France) during P3. Hand rinses and gels were available at the bed-side through wall-fixed dispensers in each room. HCWs could not differentiate the two gels because the same type of containers was used for delivering both gels. Microbiological studies started three days after the introduction of the new product. Conventional hand washing with a bland soap was available during the three periods of the study.
Antibacterial efficacy and acceptability of alcohol-based gels All nurses (n ¼ 28), nurses’ assistants (n ¼ 20) and medical doctors including senior (n ¼ 6), residents and students (n ¼ 14) agreed to participate. Each morning, 2 to 4 HCWs having had direct contacts with patients during a care were randomly chosen. A given HCW could not be included more than twice during the same period. For each HCW, hand contamination was assessed before and after alcoholic handrubbing. HCW group, type and duration of care, and duration of handrubbing were recorded on a standardized report form.
Microbiological techniques Immediately after a patient care, the left and right hands (in a randomised order) of HCWs were sampled before and after handrubbing using the glove-juice technique, as previously described.15 The HCW inserted the randomised pre-rub hand into a sterile glove containing 50 mL of a rinse with neutralizers (buffer rinse DNP [AES Laboraroires, Combourg, France] containing polysorbate 3%, cysteine 0.1%, histidine 0.1%, saponin 3%). The entire gloved hand was gently kneaded by the data collector (RA, BR) for 30 s. The hand was slowly removed from the glove and dried with a sterile paper hand-towel. The HCW rubbed hands with one or two doses of the product corresponding to the study period without any specific recommendations from the data collector. Thirty seconds after the end of handrubbing, the HCW inserted the unsampled post-rub hand into a sterile glove containing 50 mL of the rinse described as above. The entire hand was kneaded by the data collector for 30 s. Each glove juice was transferred into sterile containers and immediately sent to the technicians responsible for the bacterial count (DN, LG). Each glove juice was gently agitated and 10-fold dilutions were made in sterile saline. An aliquot of 0.1 mL from each dilution was plated on Trypticase soy agar (Oxoid, Wesel, Germany). The total bacterial count was performed after 48 h of incubation in aerobic atmosphere at 37 C. Results were expressed as the Log10 of total number of CFUs (colony forming units) per mL.
Compliance study During each of the three six-week periods of the study (P1, P2, P3), two members of the medical ICU staff (RA and BR) independently assessed the overall compliance of HCWs to hand hygiene. This was performed during 20 min observation sessions selected at random. Opportunities were broken
169
as follows: personal gesture (e.g. before and after duty hours), patient care without contact with body fluids and patient care with contact with body fluids.9 Three groups of HCWs were evaluated: paramedical staff (nurses and nurses’ assistants), physician (senior), and students (residents and medical students).
Acceptability study At the end of each period, all HCWs were asked to complete an anonymous self-administered questionnaire to assess their personal perception of the different products, using a visual analogical scale ranging from 0 to 10. The questionnaire included items about the skin feeling, pruritus, dryness, smell of each product and their general opinion about using the product on a regular basis.
Statistical analysis Considering the study hypothesis that rinses for handrubbing are more effective to reduce hand contamination than gels, the aim of the study was to obtain a difference in the reduction of bacterial contamination on HCWs hands of 20% between the two groups (rinse versus gel) with a 95% confidence interval and a ß-error of 20%. We calculated that 80 patient care activities in each period (P1, P2, P3) would be necessary. The main outcome variable was the reduction factor (RF) in hand contamination defined by the difference of bacterial counts before and immediately after applying handrubbing. The arithmetic means of all individual Log10 reduction factors were calculated. Analysis of variance (ANOVA) model and the Tukey tests were used to assess the difference between the RF of each technique (GraphPad Prism, San Diego, USA). For the acceptability study, non-parametric Kruskall Wallis and Dunn’s multiple comparison tests were used to compare the scores obtained from the analogical visual scale. Comparison of compliance rates was performed using contingency tables. A difference was considered significant when the p-value was <0.05.
Results Microbiological investigations A total of 242 handrubbing opportunities were studied: 81 for AHRR (P1), 81 for AHRG-1 (P2) and 80 for AHRG-2 (P3) (Table I). Duration of
170
F. Barbut et al.
Table I Characteristics of patient care and hand hygiene during P1, P2 and P3 and reduction factor (RF) of hand contamination Period 1 (P1) AHRR (n ¼ 81) HCWs group: - Doctors - Nurses - Nurses’ assistants - Others Duration of care (min) Mean SD (median) Duration of hand rubbing (s), Mean SD (median) Hand contamination before handrubbing (Log10 CFU/ml) Mean SD (median) Reduction factor (RF) Mean SD (median)
18 29 31 3
Period 2 (P2) AHRG-1 (n ¼ 81)
(22.2%) (35.8%) (38.3%) (3.7%)
28 23 25 5
(34.5%) (28.4%) (30.9%) (6.2%)
Period 3 (P3) AHRG-2 (n ¼ 80) 13 22 40 5
(16.2%) (27.5%) (50%) (6.3%)
18.6 21.7 (15)
15.5 14 (10)
15.5 16.5 (10)
30 11 (30)
31 14 (27)
34 15 (35)
3.11 0.94 (3.03)
3.27 0.92 (3.2)
3.22 0.87 (3.11)
1.28 0.95 (1.10)
1.29 0.84* (1.27)
0.51 þ 0.73** (0.5)
SD: standard deviation. *p > 0.05 compared to P1; **p < 0.001 compared to P1 and P2.
handrubbing was higher during P3 compared to P2 or P1 but the difference was not statistically significant. The initial levels of bacterial hand contamination before handrubbing were similar in the three groups (p > 0.2) (Table I). They ranged from 3.11 Log10 CFU/ml to 3.27 Log10 CFU/ml. The mean Log10 reduction factor with AHRR, AHRG-1 and AHRG-2 were 1.28 0.95, 1.29 0.84 and 0.51 0.73, respectively. AHRR and AHRG-1 were both more effective in reducing bacterial counts than AHRG-2 (p < 0.001).
Compliance Eight hundred and eighty two opportunities of hand hygiene were observed during P1, 971 during P2 and 826 during P3. More than 80% of the observation sessions were achieved during the day (between 08:30 am and 07:30 pm). Hand hygiene compliance by groups of HCWs and by types of contacts is described in Table II. Compliance to hand hygiene was higher in paramedical group than physicians group, whatever the period of observation. General compliance of HCWs to hand hygiene significantly increased from P1 to P3 reaching 68% when AHRG-2 was available. However, the proportion of alcoholic handrubbing decreased from 72% during P1 and 70 % during P2 to 64% during P3 (p < 0.05).
User acceptability and perception One hundred and eighty one questionnaires were filled out (62, 67, and 52 at the end of periods P1,
P2, and P3, respectively). They were completed by nurses (48.1%), nurses’ assistants (24.5%) or medical doctors (27.3%). The means and standard deviations of the scores obtained with the visual analogical scale for each parameter and for the three products are summarized in Figure 1. Dryness and pruritus were reported more frequently for AHRG-2 than for the 2 other products (p < 0.001). The smell was described more pleasant for the AHRG-1 compared to the two other products (p < 0.001). Skin feeling was considered softer with the AHRG-1 than with the AHRR (p < 0.001) and both were softer than AHRG-2 (p < 0.05). The overall assessment of the three products indicated that most HCWs would prefer to use on a regular basis AHRR or AHRG-1 rather than AHRG-2 (p < 0.001).
Discussion To the best of our knowledge, this is the first study addressing the antibacterial efficacy of alcoholbased gels compared to an alcohol-based rinse for hand hygiene under practical use conditions. Our results show that a gel containing 85% ethanol (AHRG-1) is equivalent to rinse containing 45% 2-propanol, 30% 1-propanol (AHRR) and that both are significantly more effective than a gel containing ethanol 53% and isopropanol 17% (AHRG-2) in reducing bacterial counts on hands. Hands were randomised for pre and post values in order to avoid the bias of a higher contamination associated to the dominant hand. The difference in
Antibacterial efficacy and acceptability of alcohol-based gels
171
Table II General and specific compliance during P1, P2 and P3 among all healthcare workers (in parentheses: number of opportunities observed) Period 1 (P1) AHRR % All health care workers General compliance Proportion of alcoholic handrubbing Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids Paramedical group General compliance Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids Physicians group General compliance Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids Students group General compliance Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids
53 72 43 47 71
(882)
Period 2 (P2) AHRG-1 %
(44) (455) (333)
63 70 30 66 61
67 64 46 82
(307) (14) (119) (174)
54 17 51 66 46 50 43 56
(971)
Period 3 (P3) AHRG-2 %
(27) (384) (560)
68 64 21 67 77
(826) (42) (419) (365)
63 40 68 62
(489) (5) (164) (320)
72 21 64 81
(363) (13) (167) (193)
(249) (14) (154) (81)
53 22 55 55
(158) (9) (75) (74)
47 7 57 52
(171) (28) (93) (50)
(276) (16) (182) (78)
65 31 69 63
(324) (13) (145) (166)
78 64 77 80
(292) (11) (159) (122)
p
a,b,c b,c b a,b a,c
c
NS a,b a,c
c
NS a,b a,c
a,b,c
NS a,b b,c
a
for p < 0.05 P1 vs P2. for p < 0.05 P1 vs P3. c for p < 0.05 P2 vs P3. NS: p > 0.05. b
antibacterial effect cannot be explained by the duration of handrubbing that was similar for both AHRR and AHRG-1 and even longer (but not significantly) for the less active product (AHRG-2). The longer time of handrubbing during P3 might be explained by a longer time for AHRG-2 to dry compared to the two other products. Gels have been previously tested in experimental conditions. Kramer et al. showed that none of 10 alcoholic gels tested with a content of alcohol ranging from 53% to 70% (v/v) met the EN 1500 requirements within 30 s (with a 60% 2-propanol as reference procedure).13 However, neither AHRG-1 nor AHRG-2 were assessed during this study. Conversely, Kampf et al. showed that AHRG-1 was as effective in 30 s as the reference method in 60 s and therefore was the first gel to pass the EN1500 standard.16 Our results confirmed the higher antimicrobial activity of a gel containing 85% ethanol compared to a gel with lower concentration of alcohol (70%). It has been hypothesized that differences in the antibacterial efficacy of alcohol-based gels might be explained both by the limited concentration (53%e70%) and the type of alcohol included in the formulations.17 The Log10 reduction factors (RF) ranged from 0.5 to 1.29. These results are in agreement with
those obtained during routine patient care. In a clinical trial performed in three intensive care units, the median percentage reduction in bacterial contamination after handrubbing with alcohol rinse (AHRR) was only 83%, corresponding to less than one Log10.18 In another study using the fingerprints method, the Log10 bacterial count reduction after handrubbing with AHRR was estimated to 1.40.19 Among a group of trained infection control practitioners and hospital epidemiologists, the mean reduction factor was 2 (range 0e3.85) after handrubbing with AHRR.14 The most powerful hand antiseptics may be useless if they are not accepted by HCWs whatever the reasons. Acceptability assessment showed wide differences between the three products. The overall assessment by HCWs of AHRG-1 was equivalent to that of ARHH and both were better accepted than AHRG-2 (p < 0.001). Skin dryness and smell were the major features that most users described unacceptable after repetitive use of AHRG-2. There are few data about the acceptability and perception of alcohol-based gel. Dermal tolerance of the AHRG-1 has been experimentally studied in 53 patients: results did not demonstrate a clinical relevant potential for dermal irritation or sensitization and significantly increased skin
172
F. Barbut et al.
10.00
Visual analogue scale
8.00
6.00
4.00
2.00
0.00
Soft skin feeling
Pleasant smell
Pruritus
Dryness
Figure 1 Acceptability and perception of the AHRG-1 (black bars), AHRR (white bars), and AHRG-2 (grey bars) assessed by a visual analogical scale. Difference in skin feeling, smell, pruritus and dryness were statistically significant (p < 0.001).
hydratation after repetitive use.11 In another multicentre prospective study, 65.6% of HCWs assessed AHRG-1 to be better than a comparator, irrespective of its type (gel or rinse).16 The acceptability and tolerance of gels are of primary importance because they can generate a high compliance to hand hygiene. General compliance of HCWs to hand hygiene was better during P3 (AHRG-2) than during the two other periods. However, the rate of hand hygiene achieved through conventional hand washing was significantly more important during P3 than during P2 and P1, suggesting that AHRG-2 was in fact less likely to be used than AHRR or AHRG-1. These data might be explained by the lower acceptability of AHRG-2 expressed by HCWs. In this study, several limitations must be underlined. Firstly, this study was conducted in only one unit and a few HCWs have been included two times within the same period. A multicentre study would have probably strengthened the results. Second, mean duration of handrubbing was around 30 s in the three groups. This duration is surprisingly high, especially in a ward where workloads are heavy. In the daily practice, the application time of hand hygiene is unlikely to exceed 15 s.20e22 This difference could be explained by the Hawthorn effect (external observer bias) but also by the glove juice technique, which is a time consuming method that forced HCWs to interrupt their activities for a while.
Third, the advantages and drawbacks of the glove juice technique need to be addressed. This tedious technique explores the entire hand surface contamination and not only the fingertips or thumbs and is therefore more suitable to show deficiencies in handrubbing technique.14 Nevertheless, the glove juice technique probably recovers the transient flora and part of the resident flora from the hands. The numeration of the resident flora might hamper the numeration of the transient flora and therefore mask the real activity of the antiseptics on the transient flora mostly responsible for cross contamination. Although considered as the reference method for hand numeration of bacteria, the glove juice technique might not be completely suitable for the evaluation of hand contamination during clinical practice. Fourth, the clinical relevance of the better efficacy of AHRR and AHRG-1 compared to AHRG2 needs to be addressed. Indeed, scientific studies have still not established the extent to which counts of bacteria need to be reduced in order to minimise transmission of pathogens in health care facilities and whether bacterial counts on the hands should be reduced by 1 Log10 (90% reduction), 2 Log10 (99% reduction) or more.23,24 In summary, in vivo trials are mandatory to compare the antimicrobial efficacy of alcoholbased products. Although all the products tested were claimed to pass the European standard
Antibacterial efficacy and acceptability of alcohol-based gels EN1500, our results showed significant differences in their antibacterial activity during routine care activities, AHRG-1 and AHRR being more effective than AHRG-2. Acceptability of AHRG-2 was also lower than AHRR and AHRG-1. This could be related to a less pleasant smell and a poorer dermal tolerance. Among products having a similar antimicrobial activity, the choice should take into account the HCWs’ acceptability. This point is absolutely essential to insure widespread appropriate use at the bedside.
9.
10.
11.
12.
13.
Acknowledgements 14.
This study was supported by a grant (9649341B) from the University Pierre et Marie Curie. 15.
References 1. Conly JM, Hill S, Ross J, Lertzman J, Louie TJ. Handwashing practices in an intensive care unit: the effects of an educational program and its relationship to infection rates. Am J Infect Control 1989;17:330e339. 2. Pittet D, Hugonnet S, Harbarth S, et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet 2000;356:1307e1312. 3. Albert RK, Condie F. Hand-washing patterns in medical intensive-care units. N Engl J Med 1981;304:1465e 1466. 4. Pittet D. Improving compliance with hand hygiene in hospitals. Infect Control Hosp Epidemiol 2000;21:381e386. 5. Pittet D. Compliance with hand disinfection and its impact on hospital-acquired infections. J Hosp Infect 2001;48 (Suppl. A):S40eS46. 6. Rotter ML. Arguments for alcoholic hand disinfection. J Hosp Infect 2001;48(Suppl. A):S4eS8. 7. Boyce JM, Pittet D. Guideline for Hand Hygiene in HealthCare Settings: recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Infect Control Hosp Epidemiol 2002;23(12 Suppl.):S3eS40. 8. Bischoff WE, Reynolds TM, Sessler CN, Edmond MB, Wenzel RP. Handwashing compliance by health care workers: the impact of introducing an accessible, alcohol-
16.
17.
18.
19.
20. 21. 22.
23.
24.
173
based hand antiseptic. Arch Intern Med 2000;160: 1017e1021. Maury E, Alzieu M, Baudel JL, et al. Availability of an alcohol solution can improve hand disinfection compliance in an intensive care unit. Am J Respir Crit Care Med 2000;162: 324e327. Girard R, Aho LS, Goetz ML, Labadie JC, Lejeune B. Alcoholbased hand gels and hand hygiene in hospitals. Lancet 2002; 360:1510e1511. Kampf G, Muscatiello M, Hantschel D, Rudolf M. Dermal tolerance and effect on skin hydration of a new ethanolbased hand gel. J Hosp Infect 2002;52:297e301. Pietsch H. Hand antiseptics: rubs versus scrubs, alcoholic solutions versus alcoholic gels. J Hosp Infect 2001; 48(Suppl. A):S33eS36. Kramer A, Rudolph P, Kampf G, Pittet D. Limited efficacy of alcohol-based hand gels. Lancet 2002;359:1489e1490. Widmer AE, Dangel M. Alcohol-based handrub: evaluation of technique and microbiological efficacy with international infection control professionals. Infect Control Hosp Epidemiol 2004;25:207e209. Larson EL, Aiello AE, Bastyr J, et al. Assessment of two hand hygiene regimens for intensive care unit personnel. Crit Care Med 2001;29:944e951. Kampf G, Rudolf M, Labadie JC, Barrett SP. Spectrum of antimicrobial activity and user acceptability of the hand disinfectant agent Sterillium Gel. J Hosp Infect 2002;52:141e147. Kampf G, Kramer A. Epidemiologic background of hand hygiene and evaluation of the most important agents for scrubs and rubs. Clin Microbiol Rev 2004;17:863e893. Girou E, Loyeau S, Legrand P, Oppein F, Brun-Buisson C. Efficacy of handrubbing with alcohol based solution versus standard handwashing with antiseptic soap: randomised clinical trial. BMJ 2002;325:362. Lucet JC, Rigaud MP, Mentre F, et al. Hand contamination before and after different hand hygiene techniques: a randomized clinical trial. J Hosp Infect 2002;50:276e280. Daschner FD. How cost-effective is the present use of antiseptics? J Hosp Infect 1988;11(Suppl. A):227e235. Fox MK, Langner SB, Wells RW. How good are hand washing practices? Am J Nurs 1974;74:1676e1678. Larson E, McGeer A, Quraishi ZA, et al. Effect of an automated sink on handwashing practices and attitudes in highrisk units. Infect Control Hosp Epidemiol 1991;12:422e428. Larson EL. APIC guideline for handwashing and hand antisepsis in health care settings. Am J Infect Control 1995; 23:251e269. Rotter M. Hand washing and hand disinfection. In: Mayhall CG, editor. Hospital epidemiology and infection control. 2nd ed. Philadelphia: Lippincott Williams and Wilkins; 1999. p. 1339e1355.
www.elsevierhealth.com/journals/jhin
Comparison of the antibacterial efficacy and acceptability of an alcohol-based hand rinse with two alcohol-based hand gels during routine patient care ´de ´ric Barbut a,*, Eric Maury b, Laurianne Goldwirt a, Fre ´rre-Yves Boe ¨lle c, Denis Neyme a, Rubina Aman b, Pie Beatrice Rossi b, Georges Offenstadt b a
Assistance Publique-Hoˆpitaux de Paris, Hoˆpital Saint-Antoine, Unite´ d’Hygie`ne et de Lutte contre les Infections Nosocomiales b Service de Re´animation me´dicale c Unite´ de Biostatistiques, INSERM U707 Received 6 March 2007; accepted 20 March 2007 Available online 21 May 2007
KEYWORDS Hand hygiene; Alcoholbased rinse; Alcoholbased gels; Compliance; Antibacterial efficacy
Summary The aims of this study were to compare the antibacterial efficacy of handrubbing with an alcoholic rinse (AHRR) and two different alcoholic gels (AHRG) in reducing hand contamination under practical use conditions. We wanted to assess the acceptability of the three products and to determine the effect of each product on overall hand hygiene compliance. A prospective alternating time-series clinical trial was performed in a medical intensive care unit. The study was divided into three six-week periods (P1, P2, P3). Handrubbing was achieved with Sterillium rinse (AHRR) during P1, Sterillium gel (AHRG-1) during P2 and Manugel Plus (AHRG-2) during P3. Pre- and post-rubbing hand contaminations were assessed immediately after a direct contact with a patient, using the glove juice technique. Health care workers (HCWs) evaluated the acceptability of the products through a self-administered anonymous questionnaire. Compliance of HCWs with hand hygiene was assessed during the three periods. We studied 242 handrubbing opportunities. The mean reduction factor (expressed as the Log10 CFU/mL) of the AHRR, AHRG-1 and AHRG-2 were 1.28 0.95, 1.29 0.84 and 0.51 0.73, respectively (p < 0.001). Assessment of the three products by HCWs indicated that
* Corresponding author. Address: UHLIN, Ho ˆpital Saint-Antoine, Assistance Publique-Ho ˆpitaux de Paris, 184 rue du faubourg SaintAntoine, 75012 Paris 12, France, Tel.: þ33 1 49 28 30 08; fax: þ33 1 49 28 30 09. E-mail address: [email protected] 0195-6701/$ - see front matter ª 2007 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2007.03.021
168
F. Barbut et al. AHRR and AHRG-1 were significantly better accepted than AHRG-2. The overall compliance of HCWs to hand hygiene was better when gel was available. Under practical use conditions, AHRG-1 and AHRR were more effective than AHRG-2, although all were claimed to pass the European standard EN1500. In vivo trials are essential to compare the antimicrobial efficacy of products for handrubbing. ª 2007 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction Hand hygiene is considered to be the most important measure to prevent cross transmission of microorganisms, and thus to prevent nosocomial infections.1,2 However, under routine hospital conditions, compliance with standard handwashing is unacceptably low, less than 50% in most studies published in the past 20 years.3e6 Guidelines now promote the use of alcoholic handrubbing in substitution of handwashing whatever the soap used, medicated or not.7 Alcoholic handrubbing is faster and easier to use due to ease of access at the bedside. It is more effective than handwashing with plain or antimicrobial soap, and is also better tolerated. Recent reports have clearly shown that the use of hand rinse is associated with a better compliance to hand hygiene.7e9 Pittet et al. have found an association between an increasing hand rub consumption following an intensive campaign for handrubbing and a decrease of both the incidence of nosocomial infections and the attack rate of meticillin-resistant Staphylococcus aureus (MRSA).2 Alcohol-based gels have been recently introduced in many hospitals worldwide for hand hygiene and seem to be a promising alternative to alcohol-based rinse in term of acceptability.10,11 Nevertheless, some reports using the European standard EN1500 suggested that gels might be less effective than rinses, especially because they usually contain less alcohol than rinses.12,13 This has led to controversy about the use of alcohol-based gels in hospitals and the significance of the European standard EN1500. Further studies are warranted to determine the relative efficacy of alcohol-based gels and rinses in daily clinical practice where duration and quality of application usually do not reach those required in the EN1500 standard.14 The objectives of the present study were: i) to compare the antibacterial efficacy of handrubbing with an alcoholic rinse versus handrubbing with two different alcoholic gels during ICU routine care
activities; ii) to assess the acceptability of the different products by health care workers (HCWs); and iii) to determine the effect of each product on the overall hand hygiene compliance.
Materials and methods Setting We conducted a prospective alternating timeseries trial from November 19th 2004 to March 11th 2005 in a 14-bed medical intensive care unit (MICU) of a tertiary care and university hospital (760 beds). In our hospital, alcohol-based rinses were introduced in daily clinical practice in 2000 with a hospital-wide campaign launched by the infection control committee. An additional mandatory teaching on hand hygiene was provided in September 2004 to all medical and non-medical HCWs.
Enrolment of HCWs for microbiological efficacy of hand hygiene Three consecutive six-week periods were defined (P1, P2, P3). During each period only one product was available for handrubbing. HCWs used handrubbing with Sterillium rinse (AHRR) (45% 2-propanol [isopropanol]; 30% 1-propanol; 0.2% mecetronium ethyl-sulphate, Bode Chemie, Hamburg, Germany distributed by Rivadis, Thouars, France) during P1, Sterillium gel (AHRG-1)(85% ethanol, Rivadis) during P2 and Manugel Plus (AHRG-2) (ethanol 53%, isopropanol 17%, Anios, Lille Hellemmes, France) during P3. Hand rinses and gels were available at the bed-side through wall-fixed dispensers in each room. HCWs could not differentiate the two gels because the same type of containers was used for delivering both gels. Microbiological studies started three days after the introduction of the new product. Conventional hand washing with a bland soap was available during the three periods of the study.
Antibacterial efficacy and acceptability of alcohol-based gels All nurses (n ¼ 28), nurses’ assistants (n ¼ 20) and medical doctors including senior (n ¼ 6), residents and students (n ¼ 14) agreed to participate. Each morning, 2 to 4 HCWs having had direct contacts with patients during a care were randomly chosen. A given HCW could not be included more than twice during the same period. For each HCW, hand contamination was assessed before and after alcoholic handrubbing. HCW group, type and duration of care, and duration of handrubbing were recorded on a standardized report form.
Microbiological techniques Immediately after a patient care, the left and right hands (in a randomised order) of HCWs were sampled before and after handrubbing using the glove-juice technique, as previously described.15 The HCW inserted the randomised pre-rub hand into a sterile glove containing 50 mL of a rinse with neutralizers (buffer rinse DNP [AES Laboraroires, Combourg, France] containing polysorbate 3%, cysteine 0.1%, histidine 0.1%, saponin 3%). The entire gloved hand was gently kneaded by the data collector (RA, BR) for 30 s. The hand was slowly removed from the glove and dried with a sterile paper hand-towel. The HCW rubbed hands with one or two doses of the product corresponding to the study period without any specific recommendations from the data collector. Thirty seconds after the end of handrubbing, the HCW inserted the unsampled post-rub hand into a sterile glove containing 50 mL of the rinse described as above. The entire hand was kneaded by the data collector for 30 s. Each glove juice was transferred into sterile containers and immediately sent to the technicians responsible for the bacterial count (DN, LG). Each glove juice was gently agitated and 10-fold dilutions were made in sterile saline. An aliquot of 0.1 mL from each dilution was plated on Trypticase soy agar (Oxoid, Wesel, Germany). The total bacterial count was performed after 48 h of incubation in aerobic atmosphere at 37 C. Results were expressed as the Log10 of total number of CFUs (colony forming units) per mL.
Compliance study During each of the three six-week periods of the study (P1, P2, P3), two members of the medical ICU staff (RA and BR) independently assessed the overall compliance of HCWs to hand hygiene. This was performed during 20 min observation sessions selected at random. Opportunities were broken
169
as follows: personal gesture (e.g. before and after duty hours), patient care without contact with body fluids and patient care with contact with body fluids.9 Three groups of HCWs were evaluated: paramedical staff (nurses and nurses’ assistants), physician (senior), and students (residents and medical students).
Acceptability study At the end of each period, all HCWs were asked to complete an anonymous self-administered questionnaire to assess their personal perception of the different products, using a visual analogical scale ranging from 0 to 10. The questionnaire included items about the skin feeling, pruritus, dryness, smell of each product and their general opinion about using the product on a regular basis.
Statistical analysis Considering the study hypothesis that rinses for handrubbing are more effective to reduce hand contamination than gels, the aim of the study was to obtain a difference in the reduction of bacterial contamination on HCWs hands of 20% between the two groups (rinse versus gel) with a 95% confidence interval and a ß-error of 20%. We calculated that 80 patient care activities in each period (P1, P2, P3) would be necessary. The main outcome variable was the reduction factor (RF) in hand contamination defined by the difference of bacterial counts before and immediately after applying handrubbing. The arithmetic means of all individual Log10 reduction factors were calculated. Analysis of variance (ANOVA) model and the Tukey tests were used to assess the difference between the RF of each technique (GraphPad Prism, San Diego, USA). For the acceptability study, non-parametric Kruskall Wallis and Dunn’s multiple comparison tests were used to compare the scores obtained from the analogical visual scale. Comparison of compliance rates was performed using contingency tables. A difference was considered significant when the p-value was <0.05.
Results Microbiological investigations A total of 242 handrubbing opportunities were studied: 81 for AHRR (P1), 81 for AHRG-1 (P2) and 80 for AHRG-2 (P3) (Table I). Duration of
170
F. Barbut et al.
Table I Characteristics of patient care and hand hygiene during P1, P2 and P3 and reduction factor (RF) of hand contamination Period 1 (P1) AHRR (n ¼ 81) HCWs group: - Doctors - Nurses - Nurses’ assistants - Others Duration of care (min) Mean SD (median) Duration of hand rubbing (s), Mean SD (median) Hand contamination before handrubbing (Log10 CFU/ml) Mean SD (median) Reduction factor (RF) Mean SD (median)
18 29 31 3
Period 2 (P2) AHRG-1 (n ¼ 81)
(22.2%) (35.8%) (38.3%) (3.7%)
28 23 25 5
(34.5%) (28.4%) (30.9%) (6.2%)
Period 3 (P3) AHRG-2 (n ¼ 80) 13 22 40 5
(16.2%) (27.5%) (50%) (6.3%)
18.6 21.7 (15)
15.5 14 (10)
15.5 16.5 (10)
30 11 (30)
31 14 (27)
34 15 (35)
3.11 0.94 (3.03)
3.27 0.92 (3.2)
3.22 0.87 (3.11)
1.28 0.95 (1.10)
1.29 0.84* (1.27)
0.51 þ 0.73** (0.5)
SD: standard deviation. *p > 0.05 compared to P1; **p < 0.001 compared to P1 and P2.
handrubbing was higher during P3 compared to P2 or P1 but the difference was not statistically significant. The initial levels of bacterial hand contamination before handrubbing were similar in the three groups (p > 0.2) (Table I). They ranged from 3.11 Log10 CFU/ml to 3.27 Log10 CFU/ml. The mean Log10 reduction factor with AHRR, AHRG-1 and AHRG-2 were 1.28 0.95, 1.29 0.84 and 0.51 0.73, respectively. AHRR and AHRG-1 were both more effective in reducing bacterial counts than AHRG-2 (p < 0.001).
Compliance Eight hundred and eighty two opportunities of hand hygiene were observed during P1, 971 during P2 and 826 during P3. More than 80% of the observation sessions were achieved during the day (between 08:30 am and 07:30 pm). Hand hygiene compliance by groups of HCWs and by types of contacts is described in Table II. Compliance to hand hygiene was higher in paramedical group than physicians group, whatever the period of observation. General compliance of HCWs to hand hygiene significantly increased from P1 to P3 reaching 68% when AHRG-2 was available. However, the proportion of alcoholic handrubbing decreased from 72% during P1 and 70 % during P2 to 64% during P3 (p < 0.05).
User acceptability and perception One hundred and eighty one questionnaires were filled out (62, 67, and 52 at the end of periods P1,
P2, and P3, respectively). They were completed by nurses (48.1%), nurses’ assistants (24.5%) or medical doctors (27.3%). The means and standard deviations of the scores obtained with the visual analogical scale for each parameter and for the three products are summarized in Figure 1. Dryness and pruritus were reported more frequently for AHRG-2 than for the 2 other products (p < 0.001). The smell was described more pleasant for the AHRG-1 compared to the two other products (p < 0.001). Skin feeling was considered softer with the AHRG-1 than with the AHRR (p < 0.001) and both were softer than AHRG-2 (p < 0.05). The overall assessment of the three products indicated that most HCWs would prefer to use on a regular basis AHRR or AHRG-1 rather than AHRG-2 (p < 0.001).
Discussion To the best of our knowledge, this is the first study addressing the antibacterial efficacy of alcoholbased gels compared to an alcohol-based rinse for hand hygiene under practical use conditions. Our results show that a gel containing 85% ethanol (AHRG-1) is equivalent to rinse containing 45% 2-propanol, 30% 1-propanol (AHRR) and that both are significantly more effective than a gel containing ethanol 53% and isopropanol 17% (AHRG-2) in reducing bacterial counts on hands. Hands were randomised for pre and post values in order to avoid the bias of a higher contamination associated to the dominant hand. The difference in
Antibacterial efficacy and acceptability of alcohol-based gels
171
Table II General and specific compliance during P1, P2 and P3 among all healthcare workers (in parentheses: number of opportunities observed) Period 1 (P1) AHRR % All health care workers General compliance Proportion of alcoholic handrubbing Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids Paramedical group General compliance Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids Physicians group General compliance Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids Students group General compliance Personal gestures Care without exposure to patient fluids Care with exposure to patient fluids
53 72 43 47 71
(882)
Period 2 (P2) AHRG-1 %
(44) (455) (333)
63 70 30 66 61
67 64 46 82
(307) (14) (119) (174)
54 17 51 66 46 50 43 56
(971)
Period 3 (P3) AHRG-2 %
(27) (384) (560)
68 64 21 67 77
(826) (42) (419) (365)
63 40 68 62
(489) (5) (164) (320)
72 21 64 81
(363) (13) (167) (193)
(249) (14) (154) (81)
53 22 55 55
(158) (9) (75) (74)
47 7 57 52
(171) (28) (93) (50)
(276) (16) (182) (78)
65 31 69 63
(324) (13) (145) (166)
78 64 77 80
(292) (11) (159) (122)
p
a,b,c b,c b a,b a,c
c
NS a,b a,c
c
NS a,b a,c
a,b,c
NS a,b b,c
a
for p < 0.05 P1 vs P2. for p < 0.05 P1 vs P3. c for p < 0.05 P2 vs P3. NS: p > 0.05. b
antibacterial effect cannot be explained by the duration of handrubbing that was similar for both AHRR and AHRG-1 and even longer (but not significantly) for the less active product (AHRG-2). The longer time of handrubbing during P3 might be explained by a longer time for AHRG-2 to dry compared to the two other products. Gels have been previously tested in experimental conditions. Kramer et al. showed that none of 10 alcoholic gels tested with a content of alcohol ranging from 53% to 70% (v/v) met the EN 1500 requirements within 30 s (with a 60% 2-propanol as reference procedure).13 However, neither AHRG-1 nor AHRG-2 were assessed during this study. Conversely, Kampf et al. showed that AHRG-1 was as effective in 30 s as the reference method in 60 s and therefore was the first gel to pass the EN1500 standard.16 Our results confirmed the higher antimicrobial activity of a gel containing 85% ethanol compared to a gel with lower concentration of alcohol (70%). It has been hypothesized that differences in the antibacterial efficacy of alcohol-based gels might be explained both by the limited concentration (53%e70%) and the type of alcohol included in the formulations.17 The Log10 reduction factors (RF) ranged from 0.5 to 1.29. These results are in agreement with
those obtained during routine patient care. In a clinical trial performed in three intensive care units, the median percentage reduction in bacterial contamination after handrubbing with alcohol rinse (AHRR) was only 83%, corresponding to less than one Log10.18 In another study using the fingerprints method, the Log10 bacterial count reduction after handrubbing with AHRR was estimated to 1.40.19 Among a group of trained infection control practitioners and hospital epidemiologists, the mean reduction factor was 2 (range 0e3.85) after handrubbing with AHRR.14 The most powerful hand antiseptics may be useless if they are not accepted by HCWs whatever the reasons. Acceptability assessment showed wide differences between the three products. The overall assessment by HCWs of AHRG-1 was equivalent to that of ARHH and both were better accepted than AHRG-2 (p < 0.001). Skin dryness and smell were the major features that most users described unacceptable after repetitive use of AHRG-2. There are few data about the acceptability and perception of alcohol-based gel. Dermal tolerance of the AHRG-1 has been experimentally studied in 53 patients: results did not demonstrate a clinical relevant potential for dermal irritation or sensitization and significantly increased skin
172
F. Barbut et al.
10.00
Visual analogue scale
8.00
6.00
4.00
2.00
0.00
Soft skin feeling
Pleasant smell
Pruritus
Dryness
Figure 1 Acceptability and perception of the AHRG-1 (black bars), AHRR (white bars), and AHRG-2 (grey bars) assessed by a visual analogical scale. Difference in skin feeling, smell, pruritus and dryness were statistically significant (p < 0.001).
hydratation after repetitive use.11 In another multicentre prospective study, 65.6% of HCWs assessed AHRG-1 to be better than a comparator, irrespective of its type (gel or rinse).16 The acceptability and tolerance of gels are of primary importance because they can generate a high compliance to hand hygiene. General compliance of HCWs to hand hygiene was better during P3 (AHRG-2) than during the two other periods. However, the rate of hand hygiene achieved through conventional hand washing was significantly more important during P3 than during P2 and P1, suggesting that AHRG-2 was in fact less likely to be used than AHRR or AHRG-1. These data might be explained by the lower acceptability of AHRG-2 expressed by HCWs. In this study, several limitations must be underlined. Firstly, this study was conducted in only one unit and a few HCWs have been included two times within the same period. A multicentre study would have probably strengthened the results. Second, mean duration of handrubbing was around 30 s in the three groups. This duration is surprisingly high, especially in a ward where workloads are heavy. In the daily practice, the application time of hand hygiene is unlikely to exceed 15 s.20e22 This difference could be explained by the Hawthorn effect (external observer bias) but also by the glove juice technique, which is a time consuming method that forced HCWs to interrupt their activities for a while.
Third, the advantages and drawbacks of the glove juice technique need to be addressed. This tedious technique explores the entire hand surface contamination and not only the fingertips or thumbs and is therefore more suitable to show deficiencies in handrubbing technique.14 Nevertheless, the glove juice technique probably recovers the transient flora and part of the resident flora from the hands. The numeration of the resident flora might hamper the numeration of the transient flora and therefore mask the real activity of the antiseptics on the transient flora mostly responsible for cross contamination. Although considered as the reference method for hand numeration of bacteria, the glove juice technique might not be completely suitable for the evaluation of hand contamination during clinical practice. Fourth, the clinical relevance of the better efficacy of AHRR and AHRG-1 compared to AHRG2 needs to be addressed. Indeed, scientific studies have still not established the extent to which counts of bacteria need to be reduced in order to minimise transmission of pathogens in health care facilities and whether bacterial counts on the hands should be reduced by 1 Log10 (90% reduction), 2 Log10 (99% reduction) or more.23,24 In summary, in vivo trials are mandatory to compare the antimicrobial efficacy of alcoholbased products. Although all the products tested were claimed to pass the European standard
Antibacterial efficacy and acceptability of alcohol-based gels EN1500, our results showed significant differences in their antibacterial activity during routine care activities, AHRG-1 and AHRR being more effective than AHRG-2. Acceptability of AHRG-2 was also lower than AHRR and AHRG-1. This could be related to a less pleasant smell and a poorer dermal tolerance. Among products having a similar antimicrobial activity, the choice should take into account the HCWs’ acceptability. This point is absolutely essential to insure widespread appropriate use at the bedside.
9.
10.
11.
12.
13.
Acknowledgements 14.
This study was supported by a grant (9649341B) from the University Pierre et Marie Curie. 15.
References 1. Conly JM, Hill S, Ross J, Lertzman J, Louie TJ. Handwashing practices in an intensive care unit: the effects of an educational program and its relationship to infection rates. Am J Infect Control 1989;17:330e339. 2. Pittet D, Hugonnet S, Harbarth S, et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet 2000;356:1307e1312. 3. Albert RK, Condie F. Hand-washing patterns in medical intensive-care units. N Engl J Med 1981;304:1465e 1466. 4. Pittet D. Improving compliance with hand hygiene in hospitals. Infect Control Hosp Epidemiol 2000;21:381e386. 5. Pittet D. Compliance with hand disinfection and its impact on hospital-acquired infections. J Hosp Infect 2001;48 (Suppl. A):S40eS46. 6. Rotter ML. Arguments for alcoholic hand disinfection. J Hosp Infect 2001;48(Suppl. A):S4eS8. 7. Boyce JM, Pittet D. Guideline for Hand Hygiene in HealthCare Settings: recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Infect Control Hosp Epidemiol 2002;23(12 Suppl.):S3eS40. 8. Bischoff WE, Reynolds TM, Sessler CN, Edmond MB, Wenzel RP. Handwashing compliance by health care workers: the impact of introducing an accessible, alcohol-
16.
17.
18.
19.
20. 21. 22.
23.
24.
173
based hand antiseptic. Arch Intern Med 2000;160: 1017e1021. Maury E, Alzieu M, Baudel JL, et al. Availability of an alcohol solution can improve hand disinfection compliance in an intensive care unit. Am J Respir Crit Care Med 2000;162: 324e327. Girard R, Aho LS, Goetz ML, Labadie JC, Lejeune B. Alcoholbased hand gels and hand hygiene in hospitals. Lancet 2002; 360:1510e1511. Kampf G, Muscatiello M, Hantschel D, Rudolf M. Dermal tolerance and effect on skin hydration of a new ethanolbased hand gel. J Hosp Infect 2002;52:297e301. Pietsch H. Hand antiseptics: rubs versus scrubs, alcoholic solutions versus alcoholic gels. J Hosp Infect 2001; 48(Suppl. A):S33eS36. Kramer A, Rudolph P, Kampf G, Pittet D. Limited efficacy of alcohol-based hand gels. Lancet 2002;359:1489e1490. Widmer AE, Dangel M. Alcohol-based handrub: evaluation of technique and microbiological efficacy with international infection control professionals. Infect Control Hosp Epidemiol 2004;25:207e209. Larson EL, Aiello AE, Bastyr J, et al. Assessment of two hand hygiene regimens for intensive care unit personnel. Crit Care Med 2001;29:944e951. Kampf G, Rudolf M, Labadie JC, Barrett SP. Spectrum of antimicrobial activity and user acceptability of the hand disinfectant agent Sterillium Gel. J Hosp Infect 2002;52:141e147. Kampf G, Kramer A. Epidemiologic background of hand hygiene and evaluation of the most important agents for scrubs and rubs. Clin Microbiol Rev 2004;17:863e893. Girou E, Loyeau S, Legrand P, Oppein F, Brun-Buisson C. Efficacy of handrubbing with alcohol based solution versus standard handwashing with antiseptic soap: randomised clinical trial. BMJ 2002;325:362. Lucet JC, Rigaud MP, Mentre F, et al. Hand contamination before and after different hand hygiene techniques: a randomized clinical trial. J Hosp Infect 2002;50:276e280. Daschner FD. How cost-effective is the present use of antiseptics? J Hosp Infect 1988;11(Suppl. A):227e235. Fox MK, Langner SB, Wells RW. How good are hand washing practices? Am J Nurs 1974;74:1676e1678. Larson E, McGeer A, Quraishi ZA, et al. Effect of an automated sink on handwashing practices and attitudes in highrisk units. Infect Control Hosp Epidemiol 1991;12:422e428. Larson EL. APIC guideline for handwashing and hand antisepsis in health care settings. Am J Infect Control 1995; 23:251e269. Rotter M. Hand washing and hand disinfection. In: Mayhall CG, editor. Hospital epidemiology and infection control. 2nd ed. Philadelphia: Lippincott Williams and Wilkins; 1999. p. 1339e1355.