Bacterial Reduction of Hand Contamination: Hand Rubbing with Alcohol-Based Solution or Hand Washing with Soap and Water?

Health care-associated infections (HAIs) occur worldwide and affect both developed and developing countries. Hand hygiene, including hand washing (washing hands with plain soap), hygienic hand washing (washing hands with medicated soap) and hygienic hand rubbing (use of antiseptic rubs or solutions) had been recommended as effective in preventing and controlling HAIs. Thus a mini-review was conducted to answer the question is alcoholbased hand sanitizer more effective than standard hand washing with soap and water in bacterial reduction during patient care? Two studies were reviewed which showed statistically significant difference between before and after hand rubbing with alcohol-based and hand washing with soap and water. However bacterial reduction in alcoholbased solution was higher than that of soap and water (0.85 vs 0.58).


Inclusion and Exclusion Criteria
Primary studies comparing alcohol-based solution and hand wash with soap (either plain or antiseptic) and water and among HCWs were included. The study had to be randomised trial. In order for trials to be included, bacterial reduction had to be its primary outcome measured. The search resulted in 200 citations in which filters such as English Language and humans were applied bringing them down to 6 papers ( Figure 1) which 4 were excluded (Table 2).   Harbarth et al (2001) Measured compliance of alcohol-based hand gel in three ICUs Sandora et al (2005) Conducted a trial of a multifaceted intervention including alcohol-based hand sanitizer and hand-hygiene education to reduce illness transmission in the home Pittet et al (2007) Measured tolerability and acceptability among 3 alcohol-based hand rub formulations Weber et al (2009) Compared the effectiveness of alcohol hand rub after 1.5 and 3 minutes of application

Methodological Quality of Included Studies
The two studies reviewed included 73participants in total and both compared hand rubbing with alcohol-based solution and hand washing with soap and water. Table 3 gives the description of the main features of the two trials. An appraisal tool (CASP) adapted from Guyatt et al (1993Guyatt et al ( , 1994 was used to critically appraise the two trials. Bacterial reduction of hand contamination during routine patient care in wards and ICUs Zaragoza et al's (1999) study undertaken in Spain from February to March 1997, randomly recruited 50HCWs assigned to hand washing with liquid soap or hand rubbing with alcohol-based solution. Girou et al (2002) recruited 23HCWs, assigned hand washing with antiseptic soap and water and hand rubbing with alcohol-based solution to parallel groups in France from June to July 2000. Both studies took imprints from fingertips and palm of participants by researchers in the units without interfering with hand hygiene before and after patient activities. The microbiologists who examined the culture plates and did the reporting of the results were blinded in both trials. This made the results unbiased (Jadad 1998). Blinding of the researchers who collected the samples was not necessary as they did not examine the plates.
Both trials (table 4) were of reasonable methodological quality though not without flaws. The drop-out rate for Zaragoza et al (1999) was 14% which did not exceed the maximum rate considered to be tolerable by Sackett et al (2000) and so believed do not affect the validity of the results significantly (Gardener & Altman 1989). The two trials recruited similar participants, conducted in similar settings, received similar treatment at the start of the study as well as similar intervention been implemented. Clinical homogeneity exists between both studies.
Although both studies are clinical trials, Zaragoza et al (1999) used a crossover design while Girou et al (2002) used parallel group design. Girou et al (2002) analysis was based on intention-to-treat and Zaragoza et al (1999) adapted "as treated analysis". Methods of recruitment of HCWs were explicitly stated and random assignment of participants was clearly described but blinding was not clear in both studies. Methodologically some variations exist between both studies but the extent of heterogeneity was not determined and believed do not affect the outcome of both trials.

Findings
The two studies showed bacterial reduction after hand washing with both soap and water and hand rubbing with alcohol-based solution. Zaragoza et al (1999) reported mean colony forming unit (CFU) of bacterial before as 75 (SD 39) after as 9 (SD 11) for hand rubbing with alcohol-based solution compared to 82 (SD 75) before and 42 (SD 39) after hand washing. Mean CFU of bacterial in Girou et al (2002) before alcohol-based solution was 271 (SD 372) and 232 (SD 331) for soap and water. It reported 35 (SD 59) for alcohol-based solution and 69 (SD 106) for soap and water after their usage. The average reduction of bacterial was 49.6 % for soap and water and 88.2% for alcohol-based solution (Zaragoza et al 1999). Much difference was not seen in Girou et al (2002) trial, as 73% for soap and water and 86% for alcohol-based solution after usage. The difference in the effect size between after the intervention was statistically significant (Table 5). There was a high degree of statistical homogeneity between findings from both studies.

Discussion
Both trials showed statistical significant difference between before and after hand rubbing with alcohol-based and hand washing with soap and water. However bacterial reduction in alcohol-based solution was higher than that of soap and water. The effect size calculated shows that alcohol-based is more effective than soap and water bacterial reduction. A review undertaken by Picheansathian (2004) showed hand rubbing with alcohol-based solution significantly reduced bacterial more efficiently than did non-medicated soap. Girou et al (2002) used a medicated soap which resulted in 73% bacterial reduction, however Zaragoza et al (1999) using non-medicated liquid soap recorded 49.6% reduction but alcohol-based solution was still high in bacterial reduction in both trials (86% & 88.2 %).
Moreover compliance with hand hygiene was improved with the use of alcohol-based solution (Lona et al 2003, Picheansathian 2004, Yeng et al 2011. Alcohol amount in the agent to be tested would contribute to effectiveness of it. In Larson et al's (2001) study, hand rubbing with alcohol-based solution was equivalent to hand washing with antiseptic soap in bacterial reduction. The factor could be attributed to the amount of alcohol in the agent which was 61% to 75% in the alcohol-based solution used by Zaragoza et al (1999) and Girou et al (2002).

Conclusion
The objective of this mini review was to identify which is more effective; alcohol-based solution or soap and water. This review is in line with other studies that support the use of alcohol-based solution in routine patient care as it appears to be more effective than hand washing with soap and water. It is recommended to increase the usage of alcohol-based solution in clinical settings despite the cost as it has been proven to be more effective in reduction of hand contamination, causing less skin irritation and dryness and could be more accessible (Boyce & Pearson 2003, Picheansathian 2004. Further research and review is needed to assess the association between the use of alcohol-based solution and soap and water in HCAIs rates among patients. This review is limited in a number of ways. As the author is a novice in research and in doing a review, many flaws could be identified. Also the author assessed and appraised the trails alone which is in contrast to what is been suggested by Sackett et al (2000). Moreover the search of primary studies was conducted by the author alone, making it not without limitations. Time was a crucial factor in this review and so other databases and peer review journals could not be checked. However the findings of this review could not be affected much as it has been proven that alcohol-based solution is more effective than hand washing with soap and water.