Clinical scenario report

1. INTRODUCTION.
About 200,000 healthcare-associated infections (HAIs) are acquired by patients whilst receiving some form of health care in Australia in a one year period (AIHW, 2011). National hand hygiene compliance rates are low, 68.3%, which is concerning given that hand hygiene is the single most important task health care workers can to do prevent HAIs (Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002, p.362). The purpose of this report is to investigate the best methods of hand hygiene to decrease the rate of cross-infection in a hospital setting. To focus the search, this report defines ‘best methods of hand hygiene’ as the best products to wash one’s hands with. Therefore, methods of hand-drying and hand washing protocols are excluded, as well as literature discussing compliance rates and educational interventions.
2. METHODOLOGY.
Practice Question: what are the best methods of hand hygiene to decrease the rate of cross-infection in a hospital setting?
PICOT:
Population: hospital-based personnel, i.e. health-care workers, students.
Intervention: products to wash your hands with, including liquid soap, 70% alcohol based hand rub, chlorhexidine gluconate 4%, and 10% povidone-iodine. (N.B. Interventions of compliance, education and hand-drying were NOT measured).
Control: other products to wash your hands with.
Outcome: infection rates and microbial counts on hands.
Type of study: randomised-controlled trial or experimental studies.
The Search Process: used keywords (or MESH/subject headings) combined with Boolean phrases “and”, then limited the final result to randomised-controlled trial.
1. Hand hygiene
2. Cross infection
3. Hospital units or hospitals
4. Hospital* or unit*
5. 3 or 4
6. 1 and 2 and 5
7. Limit 6 to randomised-controlled trials (RCTs), human subjects and English.
•Inclusion criteria: RCTs or experimental studies, English and human subjects.
•Exclusion criteria: non-English studies, non-human subjects, studies below level II (controlled trial without randomisation, case-control or cohort study, systematic review of qualitative or descriptive studies, qualitative or descriptive study, expert opinion or consensus).
Hand-drying, hand-washing protocols, compliance and educational interventions were not made exclusion criteria by applying limits to the database or negative keywords in the initial search. Instead the abstracts of each article from the final search result were read and assessed for appropriateness (screening), as the number generated was a manageable size to appraise (n=77, see flow diagram on next page). 71 of these 77 articles were excluded, essentially because they focused on compliance, educational interventions, hand drying methods, and hand-washing protocols. It was established at the beginning of the report that ‘best methods of hand hygiene’ was defined as the best products to wash one’s hands, which guided the inclusion and exclusion criteria. Six full-text articles were printed; with one of these articles being excluded as the setting was a university campus instead of a hospital setting (incorrect population). The five studies that were kept investigated the best methods of hand hygiene (i.e. different products to wash one’s hands) to decrease the rate of cross-infection in a hospital setting.
•Study types included: level II studies – RCTs or experimental studies.
•Places/databases searched: Pubmed, CINAHL, and Medline.
Critical Appraisal Process: Is the study design valid?
N.B. The criteria used to evaluate the remaining five study designs’ validity was that used by Fineout-Overholt, Melnyk, Stillwell & Williamson (2010), prescribed reading 2 of the week 7 reading list NURS1007.
Rapid Critical Appraisal Questions (Fineout-Overholt et al., 2010, pp. 44-45):
1. Are the results of the study valid?
A. Were the subjects randomly assigned to the intervention and control groups?
B. Was random assignment concealed from the individuals enrolling the subjects?
C. Were the subjects and providers blind to the study group?
D. Were reasons given to explain why subjects didn’t complete the study?
E. Were the follow-up assessments long enough to fully study the effects of the intervention?
F. Were the subjects analysed in the group to which they were randomly assigned?
This method of analysis is termed “intention-to-treat analysis”, whereby researchers keep participants in their assigned groups when conducting the analysis and intends to reduce possible bias (Fineout-Overholt et al., 2010, p. 46).
G. Was the control group appropriate?
H. Were the instruments used to measure the outcomes valid and reliable?
I. Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
2. What are the results?
A. How large is the intervention or treatment effect?
Indicated by the odds ratio (OR), which measures the association between an intervention and an outcome (Fineout-Overholt et al., 2010, p. 46).
B. How precise is the intervention or treatment?
The intervention or treatment is significant if the P-value is less than 0.05 (P < 0.05) and the confidence interval (CI) for OR does not include 1.0, meaning the findings are reliable. Findings that are not statistically significant could have been obtained by chance, so cannot be confidently put into practice (Fineout-Overholt et al., 2010, p. 48).
3. Will the results help me in caring for my patients?
A. Were all clinically important outcomes measured? E.g. cost etc.
B. What are the risks and benefits of the treatment?
C. Is the treatment feasible in my clinical setting? – a hospital.
D. What are my patients’ and their families’ values and expectations for the outcome and the treatment itself?
3. RESULTS
Flow diagram from: Moher, Liberati, Tetzlaff, Altman, & the PRISMA Group. (2009).
Records identified through database searching (n = 87): CINAHL, Medline, Pubmed.
Identification
Additional records identified through other sources (n = 0)
Records after duplicates removed (n = 77)
Records screened (n = 77)
Records excluded (n = 71) See exclusion criteria for reasons.
Full-text articles assessed for eligibility (n = 6)
Full-text articles excluded (n =1) Reason: the setting was a University campus & not a hospital
Studies included in qualitative synthesis (n = 0)
Studies included in quantitative synthesis (RCTs) (n = 5)
Screening
Eligibility
Included
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 1: Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
Yes
Was random assignment concealed from the individuals enrolling the subjects?
Unknown
Were the subjects and providers blind to the study group?
Yes
Were reasons given to explain why subjects didn’t complete the study?
N/A
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes – observed 114 patient care activities with HH.
Were the subjects analysed in the group to which they were randomly assigned?
Yes – intention-to-treat principle
Was the control group appropriate?
Yes
Were the instruments used to measure the outcomes valid and reliable?
Yes – bacterial counts (%) from agar plates before and after HH.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
Yes – tables provided.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 1 Continued: Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002.
What are the results?
Ask:
How large is the intervention or treatment effect?
Handrubbing 83% bacterial reduction vs handwashing with antiseptic soap (chlorhexidine gluconate 4%) 58% reduction. P value=0.012
How precise is the intervention or treatment?
95% CI (8% to 44%)
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes – mean and median bacterial counts before and after HH for handrubbing with alcohol based gel and handwashing with antiseptic soap. The median duration of HH for both hand products was also measured.
What are the risks and benefits of the treatment?
This RCT did not disrupt routine practice and can benefit patients by determining whether alcohol hand sanitiser or antiseptic soap is more effective in reducing hand contamination.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – HH is required before and after every patient encounter, so is integral to routine practice.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients and families expect quality care and safe care whilst in hospital, i.e. they expect health care workers to wash their hands before routine patient care.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 1 Continued: Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
No
Was random assignment concealed from the individuals enrolling the subjects?
No
Were the subjects and providers blind to the study group?
No
Were reasons given to explain why subjects didn’t complete the study?
n/a
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes
Were the subjects analysed in the group to which they were randomly assigned?
n/a: latin square statistical design, with two 5×4 randomised blocks. All volunteers used all hand products once.
Was the control group appropriate?
In each experiment one different volunteer acted as a control. Total participants: 2 males and 3 females.
Were the instruments used to measure the
Yes – obtained viable counts of
outcomes valid and reliable?
MRSA on mannitol salt agar with oxacillin from dilution by the modified drop-plate technique.
Calculated removal rate of MRSA using log10 reduction factors (RF).
Had a 7 day period inbetween experiments to allow normal skin flora to repair itself.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
In each experiment one different volunteer acted as a control, whose hands were contaminated but not treated with the hand cleansing product. Total participants: 2 males and 3 females.
Table above:
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2 Continued: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
What are the results?
Ask:
How large is the intervention or treatment effect?
RF=reduction factor.
FIRST BLOCK:
Significant difference between treatments as P<0.05. 10% PVP-I (RF 3.76) and 70% ethyl alcohol (RF 3.51) had significantly higher removal rates of MRSA than plain liquid soap (RF1.96) and 4% chlorhexidine (RF1.91). SECOND BLOCK: 10% PVP-I (RF4.39) and 70% ethyl alcohol (RF3.27) were significantly more effective in remove MRSA from hands than plain liquid soap (RF1.77) and 4% chlorhexidine (RF1.37; P<0.05). Plain liquid soap was significantly more effective than chlorhexidine 4% detergent in removing MRSA from hands.
How precise is the intervention or treatment?
Confidence intervals (CI) not given in this study.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
CI could have been measured – would have provided precision of intervention.
What are the risks and benefits of the treatment?
HH is the most important procedure in preventing HAIs. Health literature has proven that MRSA can be carried transiently on the hands of health care workers, so HH is important not only to prevent HAIs but antibiotic resistant HAIs.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – HH is an integral part of routine patient care.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients expect healthcare workers to wash their hands before and after delivery of patient care.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2 Continued: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 3: Hajipour, Longstaff, Cleeve, Brewster, Bint & Henman, 2006.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
Yes – using a randomisation table.
Was random assignment concealed from the individuals enrolling the subjects?
No. But microbiologist assessing bacterial colonies present on agar plate after 24h and 48h of incubation was blinded to the washing protocol used.
Were the subjects and providers blind to the study group?
No. But microbiologist assessing bacterial colonies present on agar plate after 24h and 48h of incubation was blinded to the washing protocol used.
Were reasons given to explain why subjects didn’t complete the study?
n/a
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes – 41 procedures and 82 episodes of handwashing included in the study.
Were the subjects analysed in the group to which
Yes
they were randomly assigned?
Was the control group appropriate?
Could have been more information, assumed orthopaedic surgeons are from the same hospital/surgery.
Were the instruments used to measure the outcomes valid and reliable?
Yes – Fisher’s exact test.
Surgeon scrubbed in appropriate product, gloved, performed procedure, then gloves were removed, and fingertips from each hand placed on agar plate.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
Unknown – orthopaedic surgeons – more information required, are they at the same hospital?
What are the results?
Ask:
How large is the intervention or treatment effect?
Four hands (8%) were contaminated in the chlorhexidine group compared to 19 (34%) in the alcohol group. Fisher’s exact test showed that there is a significantly higher risk of contamination using alcohol gel compared to chlorhexidine (P=0.002).
The average bacterial colony count was also much higher in the alcohol group (20) compared to the chlorhexidine group (5). There was no relationship between the length of surgery and the degree of contamination (P=1.12, not significant as P > 0.05).
How precise is the intervention or treatment?
Confidence intervals (CI) were not given in the results of this study.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes – bacterial contamination on surgeons’ hands when washed with alcohol compared to chlorhexidine.
What are the risks and benefits of the treatment?
One of the factors affecting patient’s infection risk during a surgical procedure is the surgeon’s method of hand washing prior to the surgery, as between 50-67% of gloves are perforated during joint replacement operations (Hajipour et al., 2006).
It is therefore of benefit to the patient to know the most effective handwashing product for reducing bacterial contamination of surgeon’s hands prior to surgery.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – RCT is the gold standard study. 82 episodes of handwashing were included in the study.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients’ expect a high standard of care, which includes an aseptic surgical scrub by the surgeon prior to surgery, to prevent them from acquiring a HAI from bacterial contamination of the surgeon’s hands in case the sterile glove perforates.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 3 Continued: Hajipour, Longstaff, Cleeve, Brewster, Bint & Henman, 2006.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 4: Larson, Cimiotti, Haas, Parides, Nesin, Della-Latta, & Saiman, 2005.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
No (Clinical trial using crossover design)
Was random assignment concealed from the individuals enrolling the subjects?
No – potential for bias.
Were the subjects and providers blind to the study group?
No – potential for researcher bias.
Were reasons given to explain why subjects didn’t complete the study?
Yes – 12 nurses withdrew: 9 left the neonatal ICU and 3 no longer wanted to participate.
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes – Data collection periods were March 1, 2001 to January 31, 2002, (year 1), and March 1, 2002, to January 31, 2003, (year 2), with a one month hiatus (February 2002) for product crossover.
Were the subjects analysed in the group to which they were randomly assigned?
N/A
Was the control group appropriate?
Crossover clinical trial – so same
group of nurses used both HH products.
Were the instruments used to measure the outcomes valid and reliable?
Yes – Interrater reliability first established with pilot work & infection data collected simultaneously and independently by a study nurse epidemiologist and each hospital’s nurse epidemiologist. Reliability measured throughout study during meetings between nurse epidemiologist and physician co-investigators (paediatric infectious disease specialist & neonatologist).
To measure transient flora rather than resident flora, nurses cleansed their hands using the assigned product, then a modified glove-juice technique was used for sampling. All microbiologic testing was performed
by the Clinical Microbiology Service of Columbia University Medical Centre, NY to determine microbial counts.
Nurses completed diary cards (1 shift per month) on the frequency of hand hygiene.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
n/a – crossover clinical trial
What are the results?
Ask:
How large is the intervention or treatment effect?
Odds ratio (OR) for alcohol compared with handwashing were 0.98 (95% CI, 0.77-1.25) for any neonatal infection. The mean microbial counts on nurses’ hands (3.21 and 3.11 colony- forming units for handwashing and alcohol respectively, P=0.38, therefore it is not significant as P > 0.05). But the condition of the nurses’ skin was improved using the
alcohol based solution (P=0.02 and P=0.049 for observer and self-assessments respectively).
How precise is the intervention or treatment?
The confidence interval (CI) for the OR includes 1.0, so the intervention is not significant.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes – Neonatal infection rates (bloodstream infections, pneumonia, conjunctivitis, skin and soft tissue infections, central nervous system infections), microbial counts on nurses’ hands and their skin condition.
What are the risks and benefits of the treatment?
There was no significant difference in neonatal infection risk during the use of chlorhexidine gluconate hand wash compared to alcohol based hand rub in this study.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure
Yes. Each phase of hand product was 11 consecutive months (i.e. 11
outcomes & the process of implementing the intervention).
months of using 61% alcohol based hand sanitiser followed by a cross over period, then 11 months of 2% chlorhexidine gluconate).
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
That the result will be significant, i.e. one product (alcohol hand sanitiser 61% or chlorhexidine gluconate 2%) will be more significant in reducing microbial counts on hands and reducing neonatal HAIs than the other.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 4 Continued: Larson, Cimiotti, Haas, Parides, Nesin, Della-Latta, & Saiman, 2005.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 5: Stahl, Morse, & Parks, 2007.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
No
Was random assignment concealed from the individuals enrolling the subjects?
No – so the personnel conducting microbiologic assessments of samples were blinded to the treatment assignments instead.
Were the subjects and providers blind to the study group?
No
Were reasons given to explain why subjects didn’t complete the study?
n/a
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes
Were the subjects analysed in the group to which they were randomly assigned?
n/a
Was the control group appropriate?
n/a each subject received both hand products.
Were the instruments used to measure the outcomes valid and reliable?
Yes – independent microbiologist.
Cup scrub methods were used to
obtain cultures; log reductions (reduction factors) of bacterial counts; chemical testing with bleach/colour change.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
n/a
What are the results?
Ask:
How large is the intervention or treatment effect?
Iodine log reduction saline soak 3.67 (mean), saline rinse 3.89 (mean).
Chlorhexidine log reduction saline soak 3.20 (mean), saline rinse 3.67 (mean).
35 out of 36 gauze samples had detectable chlorhexidine while no samples had detectable iodine (P < 0.0001). So chlorhexidine is removed by saline-soaked gauze while iodine povacrylex water-insoluble film remains intact under the same conditions.
How precise is the intervention or treatment?
CI not provided.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes except for costs.
What are the risks and benefits of the treatment?
Preventing HAIs with the most effective products to wash hands for hand hygiene.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – wound irrigation fluids are common in surgery, so whether saline exposure affects the antimicrobial activity of chlorhexidine and iodine surgical scrub is relevant research, the results of which can be easily implemented into practice.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients have an expectation that the surgeon will have done an aseptic surgical scrub prior to surgery (even if they don’t know the proper name, they will still have the expectation).
Summary of Findings from Articles.
First Author
Year
Setting
Participants
Interventions & Comparisons In Detail
Outcomes In Detail
Quality of Studies
Girou, E.
2002
3 intensive care units (ICUs) in a French university hospital.
23 healthcare workers.
Handrubbing with (75%) alcohol based solution (n=12) compared to handwashing with antiseptic soap (chlorhexidine gluconate 4%) (n=11) when hand hygiene (HH) was required before & after patient care. Imprints of nurses’ dominant hand were taken on an agar plate to quantify the bacterial count before &
Reduction in bacterial count from imprint before & after HH. Median % reduction in bacterial contamination of hands with handrubbing (alcohol) was 83% vs 58% for handwashing with antiseptic soap. P-value= 0.012. The median duration of HH was 30 seconds for both groups, but this only achieves bactericidal activity in the alcohol hand rub & is inadequate HH for a hand wash with antiseptic soap.
RCT (level II)
after HH.
Hand-rubbing with an alcohol based solution is more effective in reducing bacterial contamination than a hand wash with antiseptic soap, partly due to the inadequate time spent washing hands. 65% of HH lasted less than 30 seconds.
Guilhermetti, M.
2001
Hospital (unknown which one). Took place in same hospital where they isolated the strain of MRSA from
5 healthy adults, 2 males & 3 females, with no skin problems & who worked in a hospital.
The study aimed to compare the effectiveness of 4 hand products for removing hospital strain methicillin-resistant Staphylococcus aureus (MRSA) artificially applied to 5 participants’ hands. In
RF=reduction factor.
FIRST BLOCK:
Significant difference between treatments as P<0.05. 10% PVP-I (RF 3.76) and 70% ethyl alcohol (RF 3.51) had significantly higher removal rates of MRSA than plain liquid soap (RF1.96) and 4%
Experimental:
Latin square statistical design, with two 5×4 randomised blocks.
a hospitalised patient.
each experiment all hand-cleansing agents were tested, and volunteers used all products once. In each experiment one different volunteer acted as a control, whose hands were contaminated but not treated with the hand cleansing product. There was an interval of 7 days between experiments. The first block was light hand contamination and the second block was heavy hand contamination. The
chlorhexidine (RF1.91). SECOND BLOCK: 10% PVP-I (RF4.39) and 70% ethyl alcohol (RF3.27) were significantly more effective in remove MRSA from hands than plain liquid soap (RF1.77) and 4% chlorhexidine (RF1.37; P<0.05). Plain liquid soap was significantly more effective than chlorhexidine 4% detergent in removing MRSA from hands.
hand products tested were plain liquid soap, ethyl alcohol 70%, 10% povidone-iodine liquid soap (PVP-I), chlorhexidine gluconate 4% detergent.
Hajipour, L.
2006
Orthopaedic surgery, in England. No more information provided.
Orthopaedic surgeons.
Overall, 41 surgeries & 82 hand washing episodes were included in the study.
Orthopaedic surgeons were randomly allocated to two different handwashing protocols. All surgeons hand washed for 5 min with chlorhexidine for the first case, and depending on randomisation, washed for 3 min with either alcohol gel or chlorhexidine. The
Four hands (8%) were contaminated in the chlorhexidine group compared to 19 (34%) in the alcohol group. Fisher’s exact test showed that there is a significantly higher risk of contamination using alcohol gel compared to chlorhexidine (P=0.002). The average bacterial colony count was also much higher in the alcohol group
Prospective RCT.
surgeons gloved as per normal and performed surgery. Gloves were then carefully removed and the surgeon’s fingertips placed on an agar plate. An independent microbiologist assessed the bacterial colonies on the agar plates at 24h & 48h of incubation.
(20) compared to the chlorhexidine group (5). There was no relationship between the length of surgery and the degree of contamination (P=1.12, not significant as P > 0.05). Alcohol gel disinfectant is not a suitable alternative to chlorhexidine for handwashing prior to surgery due to the higher risk of bacterial contamination, which may lead to post-op infections in the event of glove perforation.
Larson, E. L.
2005
2 neonatal intensive care units
119 full-time nurses & neonates
Two HH products were tested: traditional antiseptic hand wash
Neonatal infection rates and microbial counts on nurses’ hands were equivalent
Clinical trial using a crossover
including 2932 neonatal hospital admissions. Located in Manhattan, NY.
hospitalised for more than 24 hours.
(chlorhexidine gluconate 2%) & alcohol hand sanitiser (61% ethanol & emollients). Each product was used for 11 consecutive months in each neonatal ICU in random order.
during handwashing with antiseptic soap and handrubbing with alcohol hand sanitiser. Odds ratio (OR) for alcohol compared with handwashing were 0.98 (95% CI, 0.77-1.25) for any neonatal infection. Bloodstream infections, pneumonia, conjunctivitis, skin and soft tissue infections & central nervous system infections were monitored because these represent more than 80% of all HAIs in neonates. The mean microbial counts on nurses’ hands (3.21 and 3.11 colony- forming units for handwashing and
design.
alcohol respectively, P=0.38, therefore it is not significant as P > 0.05). But the condition of the nurses’ skin was improved using the alcohol based solution (P=0.02 and P=0.049 for observer and self-assessments respectively).
Stahl, J. B.
2007
Hospital
36 participants, aged 18 to 65. Exclusion criteria: skin allergies, damaged skin, thyroid condition, antibiotic sensitivities,
Both iodine povacrylex in alcohol and chlorhexidine gluconate in alcohol were applied to the forearms of 36 participants and allowed to dry. The sites were exposed to either saline rinse or saline-saturated gauze. Saline rinse: An
Chlorhexidine is removed by saline-soaked gauze while the iodine povacrylex water-insoluble film remains intact under the same conditions. The implication is that similar results occur in surgery, e.g. wound irrigation.
Randomised blinded study
used tanning beds or chemically treated swimming pools, or are pregnant.
indicator organism was seeded onto the treated sites & after 30 minutes samples were collected. Log reductions calculated for surviving bacterial colonies.
Saline-saturated gauze: this gauze was chemically analysed for the presence of chlorhexidine of iodine.
4. DISCUSSION.
Hand hygiene (HH) is the single most important task health care workers can do to reduce bacterial hand contamination and decrease the rate of cross-infection in a hospital setting, therefore helping to prevent healthcare-associated infections (HAIs) (Girou et al., 2002; Guilhermetti, Hernandes, Fukushigue, Garcia, Cardoso, 2001, p. 107).
The research literature argues that when performing routine patient care in a hospital setting, handrubbing with an alcohol based solution, 75% ethyl alcohol, is significantly more efficient in reducing hand contamination than handwashing with antiseptic soap (chlorhexidine gluconate 4%), partly due to the inadequate time healthcare workers spend on hand hygiene (Girou et al., 2002). Girou et al. (2002) RCT’s intervention and control groups both washed their hands for 30 seconds, which was long enough to achieve bactericidal activity in the intervention group using the alcohol based hand-rub solution, but inadequate for the control group using the chlorhexidine gluconate antiseptic soap to hand wash with water. To achieve bactericidal activity with antiseptic soap and water hand hygiene is required to last 60 seconds.
It is interesting to compare Girou et al. (2002) RCT with Larson et al. (2005) clinical trial (crossover design) as both investigate the effect of antiseptic handwashing versus alcohol based handrubbing on microbial hand counts. As discussed above, while Girou et al. (2002) argue that handrubbing with alcohol based solution (75% ethanol and emollient) is significantly more efficient in reducing hand contamination compared to handwashing with chlorhexidine gluconate 4%, Larson et al. (2005) study concluded that there was no significant difference between the two hand products. However Larson et al. (2005) study used alcohol based solution that was only 61% ethanol and emollients compared to Girou et al. (2002) RCT’s 75%. This would have possibly significantly altered the results. Additionally, Larson et al. (2005) study had several potential biases, in that it was not possible to blind the nurses or the research team
to the study products (investigator bias). Hand hygiene practices were self-reported by the nurses, which is another potential bias. Larson et al. (2005) clinical trial has several limitations (biases), therefore Girou et al. (2002) RCT has more validity as it has been conducted with randomisation, the study design has been well conducted and therefore the results have worth and are significant and useful to practice (Fineout-Overholt, 2010, p. 47). Girou et al. (2002) results support existing literature, including Centre for Disease Control and Prevention Guideline for Hand Hygiene in Healthcare Settings, which recommends using alcohol based hand sanitisers rather than handwashing with antiseptic soap for patient encounters, unless hands are physically soiled (Boyce & Pittet, 2002).
While alcohol based hand sanitiser is the hand product of choice when performing routine patient care (Girou et al., 2002), it is not the product of choice for surgeons to scrub with prior to surgery (Hajipour et al., 2006). Hajipour et al. (2006) conclude that alcohol gel disinfectant is not a suitable alternative to chlorhexidine for handwashing prior to surgery due to the higher risk of bacterial contamination, which can lead to post-op infections in the event of glove perforation. This study is specific to the context of surgeons scrubbing prior to surgery and is not applicable to routine patient care in the general hospital setting. The results of this study are not surprising, due to the residual and cumulative effect of chlorhexidine (Guillhermetti et al., 2001), which is required for long surgeries in the event of glove perforation. Routine patient care encounters are generally short in nature, therefore alcohol based hand sanitiser is sufficient as bactericidal activity is achieved in 30 seconds, hand contamination is reduced, and a residual, cumulative effect is not required (Girou et al., 2002). Hajipour et al. (2006) RCT can only be applied to surgical settings of hospitals, not general wards where routine patient care is performed.
While Hajipour et al. (2006) compared alcohol gel disinfectant and chlorhexidine surgical scrubs, Stahl, Morse & Parks (2007) compared the
effectiveness of chlorhexidine and iodine povacrylex surgical scrubs when they are exposed to saline, which can occur in surgery. Iodine is the more effective hand product as the iodine povacrylex water-insoluble film remains intact when exposed to saline, whereas chlorhexidine is removed under these conditions (Stahl et al., 2007).
Guilhermetti et al. (2001) investigated different methods of hand hygiene and their effectiveness in removing a hospital strain of methicillin-resistant Staphylococcus aureus (MRSA). Not only is 75% ethyl alcohol hand-rub effective in reducing bacterial hand contamination in routine patient care where the infectious status of patients is unknown, but it is also effective in removing MRSA (hospital strain) from lightly or heavily contaminated hands (Guilhermetti et al., 2001). Guilhermetti et al. (2001) study found that ethyl alcohol 70% and PVP-I 10% were significantly more effective than 4% chlorhexidine and plain liquid soap in removing MRSA from contaminated hands. The study also found that plain liquid soap was significantly more effective than chlorhexidine 4% detergent in removing MRSA from contaminated hands. However the authors cited a bias in results due to significant differences in chlorhexidine preparations (since it is formula dependent) from different manufacturers, which could have explained the poor reduction factors of chlorhexidine gluconate compared to liquid soap (Guilhermetti et al., 2001). Guilhermetti et al. (2001) noted the conflicting evidence in the research literature regarding 4% chlorhexidine gluconate use with infectious patients. For example, while Guilhermetti et al. (2001) conclude chlorhexidine gluconate 4% to be not very effective in removing MRSA, Wade (as cited in Guilhermetti et al., 2001, p. 107) showed that it was a very reliable hand antiseptic and better than plain soap in removing HAI vancomycin-resistant Enterococcus faecium and gentamicin-resistant Enterobacter cloacae from contaminated fingertips.
5. RECOMMENDATIONS.
The research literature has shown the tendency of health care workers to spend an inadequate amount of time washing their hands with antiseptic soap when performing hand hygiene (Girou et al., 2002). Therefore in the hospital setting for routine patient care an alcohol based solution, importantly at least 75% ethanol and emollients, for handrubbing is recommended due to its rapid efficacy (30 seconds) and higher reduction in bacterial contamination compared to chlorhexidine gluconate 4% (Girou et al., 2002). An alcohol based hand-rub not only fits conveniently into the time pressures and heavy workload of the current hospital setting, but decreases the rate of cross-infection and HAIs (Girou et al., 2002).
Not only is 75% ethanol and emollient hand-rub recommended for routine patient care where the infectious status of patients is unknown, but it is also recommended for care of MRSA patients (Guilhermetti et al., 2001). PVP-I 10% is also effective in removing MRSA from hands, however 75% alcohol hand rub is recommended as it will not stain health care workers hands like PVP-I 10%, which is more ‘user-friendly’.
It should be noted that alcohol based solution is not appropriate for surgeons to scrub with prior to surgery. Chlorhexidine gluconate is better than alcohol based surgical scrub due to its residual effect, which reduces bacterial contamination, decreasing the risk of post-operative infections in the event of a glove perforation (Hajipour et al., 2006). However, if exposure to saline is likely during the surgery (e.g. wound irrigation); iodine povacrylex should be used due to its water insoluble nature, whereas the chlorhexidine will be removed by any saline exposure (Stahl et al., 2007).
To implement a hand hygiene protocol where 75% alcohol based solution for handrubbing is used, a multifactorial approach is required. The alcohol based solution could be mounted onto the wall at several patient-care points throughout the hospital unit. As well as supplying and providing easy access to the 75% alcohol based hand-rub
solution, wall posters should be mounted educating staff about the importance of hand hygiene, as well as running education seminars on the single importance of hand hygiene in preventing cross-infection and preventing HAIs (Girou et al., 2002). This implementation process needs the organisational support of the Nurse Unit Manager and the Hospital Administration to be successful.
6. REFERENCES.
•Boyce, J. M., & Pittet, D. (2002). Guideline for hand hygiene in health-care settings: recommendations of the Healthcare Infection Control Practices Advisory Committee and the HIPAC/SHEA/ APIC/ IDSA Hand Hygiene Task Force. American Journal of Infection Control, 30, S1-S6.
•Fineout-Overholt, E., Melnyk, B. M., Stillwell, S. B., & Williamson, K. M. (2010). Critical appraisal of the evidence: part I. American Journal of Nursing, 110(9), 47-52.
•Fineout-Overholt, E., Melnyk, B. M., Stillwell, S. B., & Williamson, K. M. (2010). Critical appraisal of the evidence: part II. American Journal of Nursing, 110(9), 41-48.
•Girou, E., Loyeau, S. Legrand, P., Oppein, F., & Brun-Buisson, C. (2002). Efficacy of handrubbing with alcohol based solution versus standard handwashing with antiseptic soap: randomised clinical trial. British Medical Journal, 325, 362-366.
•Guilhermetti, M., Hernandes, S. E. D., Fukushigue, Y., Garcia, L. B., & Cardoso, C. L. (2001). Effectiveness of hand-cleansing agents for removing methicillin-resistant staphylococcus aureus from contaminated hands. Infection Control and Hospital Epidemiology, 22(2), 105-108. doi: 10.1086/501872
•Hajipour, L., Longstaff, L., Cleeve, V., Brewster, N., Bint, D., & Henman, P. (2006). Hand washing rituals in trauma theatre: clean or dirty? The Royal College of Surgeons of England, 88(1), 13-15. doi: 10.1308/003588406X83032
•Larson, E. L., Cimiotti, J., Haas, J., Parides, M., Nesin, M., Della- Latta, P., & Saiman, L. (2005). Effect of antiseptic handwashing vs alcohol sanitizer on health care-associated infections in neonatal intensive care units. Archives of Pediatric and Adolescent Medicine, 159, 377-383
•Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G, & The PRISMA Group. (2009). Preferred reporting items for systematic reviews
and meta-analyses: the PRISMA statement. PLoS Med, 6(6): e1000097. doi: 10.1371/journal.pmed1000097
•Stahl, J. B., Morse, D., & Parks, P. J. (2007). Resistance of antimicrobial skin preparations to saline rinse using a seeded bacteria model. American Journal of Infection Control, 35(6), 367- 373. doi: 10.1016/j.ajic.2006.08.015

Clinical scenario report

1. INTRODUCTION.
About 200,000 healthcare-associated infections (HAIs) are acquired by patients whilst receiving some form of health care in Australia in a one year period (AIHW, 2011). National hand hygiene compliance rates are low, 68.3%, which is concerning given that hand hygiene is the single most important task health care workers can to do prevent HAIs (Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002, p.362). The purpose of this report is to investigate the best methods of hand hygiene to decrease the rate of cross-infection in a hospital setting. To focus the search, this report defines ‘best methods of hand hygiene’ as the best products to wash one’s hands with. Therefore, methods of hand-drying and hand washing protocols are excluded, as well as literature discussing compliance rates and educational interventions.
2. METHODOLOGY.
Practice Question: what are the best methods of hand hygiene to decrease the rate of cross-infection in a hospital setting?
PICOT:
Population: hospital-based personnel, i.e. health-care workers, students.
Intervention: products to wash your hands with, including liquid soap, 70% alcohol based hand rub, chlorhexidine gluconate 4%, and 10% povidone-iodine. (N.B. Interventions of compliance, education and hand-drying were NOT measured).
Control: other products to wash your hands with.
Outcome: infection rates and microbial counts on hands.
Type of study: randomised-controlled trial or experimental studies.
The Search Process: used keywords (or MESH/subject headings) combined with Boolean phrases “and”, then limited the final result to randomised-controlled trial.
1. Hand hygiene
2. Cross infection
3. Hospital units or hospitals
4. Hospital* or unit*
5. 3 or 4
6. 1 and 2 and 5
7. Limit 6 to randomised-controlled trials (RCTs), human subjects and English.
•Inclusion criteria: RCTs or experimental studies, English and human subjects.
•Exclusion criteria: non-English studies, non-human subjects, studies below level II (controlled trial without randomisation, case-control or cohort study, systematic review of qualitative or descriptive studies, qualitative or descriptive study, expert opinion or consensus).
Hand-drying, hand-washing protocols, compliance and educational interventions were not made exclusion criteria by applying limits to the database or negative keywords in the initial search. Instead the abstracts of each article from the final search result were read and assessed for appropriateness (screening), as the number generated was a manageable size to appraise (n=77, see flow diagram on next page). 71 of these 77 articles were excluded, essentially because they focused on compliance, educational interventions, hand drying methods, and hand-washing protocols. It was established at the beginning of the report that ‘best methods of hand hygiene’ was defined as the best products to wash one’s hands, which guided the inclusion and exclusion criteria. Six full-text articles were printed; with one of these articles being excluded as the setting was a university campus instead of a hospital setting (incorrect population). The five studies that were kept investigated the best methods of hand hygiene (i.e. different products to wash one’s hands) to decrease the rate of cross-infection in a hospital setting.
•Study types included: level II studies – RCTs or experimental studies.
•Places/databases searched: Pubmed, CINAHL, and Medline.
Critical Appraisal Process: Is the study design valid?
N.B. The criteria used to evaluate the remaining five study designs’ validity was that used by Fineout-Overholt, Melnyk, Stillwell & Williamson (2010), prescribed reading 2 of the week 7 reading list NURS1007.
Rapid Critical Appraisal Questions (Fineout-Overholt et al., 2010, pp. 44-45):
1. Are the results of the study valid?
A. Were the subjects randomly assigned to the intervention and control groups?
B. Was random assignment concealed from the individuals enrolling the subjects?
C. Were the subjects and providers blind to the study group?
D. Were reasons given to explain why subjects didn’t complete the study?
E. Were the follow-up assessments long enough to fully study the effects of the intervention?
F. Were the subjects analysed in the group to which they were randomly assigned?
This method of analysis is termed “intention-to-treat analysis”, whereby researchers keep participants in their assigned groups when conducting the analysis and intends to reduce possible bias (Fineout-Overholt et al., 2010, p. 46).
G. Was the control group appropriate?
H. Were the instruments used to measure the outcomes valid and reliable?
I. Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
2. What are the results?
A. How large is the intervention or treatment effect?
Indicated by the odds ratio (OR), which measures the association between an intervention and an outcome (Fineout-Overholt et al., 2010, p. 46).
B. How precise is the intervention or treatment?
The intervention or treatment is significant if the P-value is less than 0.05 (P < 0.05) and the confidence interval (CI) for OR does not include 1.0, meaning the findings are reliable. Findings that are not statistically significant could have been obtained by chance, so cannot be confidently put into practice (Fineout-Overholt et al., 2010, p. 48).
3. Will the results help me in caring for my patients?
A. Were all clinically important outcomes measured? E.g. cost etc.
B. What are the risks and benefits of the treatment?
C. Is the treatment feasible in my clinical setting? – a hospital.
D. What are my patients’ and their families’ values and expectations for the outcome and the treatment itself?
3. RESULTS
Flow diagram from: Moher, Liberati, Tetzlaff, Altman, & the PRISMA Group. (2009).
Records identified through database searching (n = 87): CINAHL, Medline, Pubmed.
Identification
Additional records identified through other sources (n = 0)
Records after duplicates removed (n = 77)
Records screened (n = 77)
Records excluded (n = 71) See exclusion criteria for reasons.
Full-text articles assessed for eligibility (n = 6)
Full-text articles excluded (n =1) Reason: the setting was a University campus & not a hospital
Studies included in qualitative synthesis (n = 0)
Studies included in quantitative synthesis (RCTs) (n = 5)
Screening
Eligibility
Included
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 1: Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
Yes
Was random assignment concealed from the individuals enrolling the subjects?
Unknown
Were the subjects and providers blind to the study group?
Yes
Were reasons given to explain why subjects didn’t complete the study?
N/A
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes – observed 114 patient care activities with HH.
Were the subjects analysed in the group to which they were randomly assigned?
Yes – intention-to-treat principle
Was the control group appropriate?
Yes
Were the instruments used to measure the outcomes valid and reliable?
Yes – bacterial counts (%) from agar plates before and after HH.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
Yes – tables provided.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 1 Continued: Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002.
What are the results?
Ask:
How large is the intervention or treatment effect?
Handrubbing 83% bacterial reduction vs handwashing with antiseptic soap (chlorhexidine gluconate 4%) 58% reduction. P value=0.012
How precise is the intervention or treatment?
95% CI (8% to 44%)
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes – mean and median bacterial counts before and after HH for handrubbing with alcohol based gel and handwashing with antiseptic soap. The median duration of HH for both hand products was also measured.
What are the risks and benefits of the treatment?
This RCT did not disrupt routine practice and can benefit patients by determining whether alcohol hand sanitiser or antiseptic soap is more effective in reducing hand contamination.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – HH is required before and after every patient encounter, so is integral to routine practice.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients and families expect quality care and safe care whilst in hospital, i.e. they expect health care workers to wash their hands before routine patient care.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 1 Continued: Girou, Loyeau, Legrand, Oppein, & Brun-Buisson, 2002.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
No
Was random assignment concealed from the individuals enrolling the subjects?
No
Were the subjects and providers blind to the study group?
No
Were reasons given to explain why subjects didn’t complete the study?
n/a
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes
Were the subjects analysed in the group to which they were randomly assigned?
n/a: latin square statistical design, with two 5×4 randomised blocks. All volunteers used all hand products once.
Was the control group appropriate?
In each experiment one different volunteer acted as a control. Total participants: 2 males and 3 females.
Were the instruments used to measure the
Yes – obtained viable counts of
outcomes valid and reliable?
MRSA on mannitol salt agar with oxacillin from dilution by the modified drop-plate technique.
Calculated removal rate of MRSA using log10 reduction factors (RF).
Had a 7 day period inbetween experiments to allow normal skin flora to repair itself.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
In each experiment one different volunteer acted as a control, whose hands were contaminated but not treated with the hand cleansing product. Total participants: 2 males and 3 females.
Table above:
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2 Continued: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
What are the results?
Ask:
How large is the intervention or treatment effect?
RF=reduction factor.
FIRST BLOCK:
Significant difference between treatments as P<0.05. 10% PVP-I (RF 3.76) and 70% ethyl alcohol (RF 3.51) had significantly higher removal rates of MRSA than plain liquid soap (RF1.96) and 4% chlorhexidine (RF1.91). SECOND BLOCK: 10% PVP-I (RF4.39) and 70% ethyl alcohol (RF3.27) were significantly more effective in remove MRSA from hands than plain liquid soap (RF1.77) and 4% chlorhexidine (RF1.37; P<0.05). Plain liquid soap was significantly more effective than chlorhexidine 4% detergent in removing MRSA from hands.
How precise is the intervention or treatment?
Confidence intervals (CI) not given in this study.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
CI could have been measured – would have provided precision of intervention.
What are the risks and benefits of the treatment?
HH is the most important procedure in preventing HAIs. Health literature has proven that MRSA can be carried transiently on the hands of health care workers, so HH is important not only to prevent HAIs but antibiotic resistant HAIs.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – HH is an integral part of routine patient care.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients expect healthcare workers to wash their hands before and after delivery of patient care.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 2 Continued: Guilhermetti, Hernandes, Fukushigue, Garcia, & Cardoso, 2001.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 3: Hajipour, Longstaff, Cleeve, Brewster, Bint & Henman, 2006.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
Yes – using a randomisation table.
Was random assignment concealed from the individuals enrolling the subjects?
No. But microbiologist assessing bacterial colonies present on agar plate after 24h and 48h of incubation was blinded to the washing protocol used.
Were the subjects and providers blind to the study group?
No. But microbiologist assessing bacterial colonies present on agar plate after 24h and 48h of incubation was blinded to the washing protocol used.
Were reasons given to explain why subjects didn’t complete the study?
n/a
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes – 41 procedures and 82 episodes of handwashing included in the study.
Were the subjects analysed in the group to which
Yes
they were randomly assigned?
Was the control group appropriate?
Could have been more information, assumed orthopaedic surgeons are from the same hospital/surgery.
Were the instruments used to measure the outcomes valid and reliable?
Yes – Fisher’s exact test.
Surgeon scrubbed in appropriate product, gloved, performed procedure, then gloves were removed, and fingertips from each hand placed on agar plate.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
Unknown – orthopaedic surgeons – more information required, are they at the same hospital?
What are the results?
Ask:
How large is the intervention or treatment effect?
Four hands (8%) were contaminated in the chlorhexidine group compared to 19 (34%) in the alcohol group. Fisher’s exact test showed that there is a significantly higher risk of contamination using alcohol gel compared to chlorhexidine (P=0.002).
The average bacterial colony count was also much higher in the alcohol group (20) compared to the chlorhexidine group (5). There was no relationship between the length of surgery and the degree of contamination (P=1.12, not significant as P > 0.05).
How precise is the intervention or treatment?
Confidence intervals (CI) were not given in the results of this study.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes – bacterial contamination on surgeons’ hands when washed with alcohol compared to chlorhexidine.
What are the risks and benefits of the treatment?
One of the factors affecting patient’s infection risk during a surgical procedure is the surgeon’s method of hand washing prior to the surgery, as between 50-67% of gloves are perforated during joint replacement operations (Hajipour et al., 2006).
It is therefore of benefit to the patient to know the most effective handwashing product for reducing bacterial contamination of surgeon’s hands prior to surgery.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – RCT is the gold standard study. 82 episodes of handwashing were included in the study.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients’ expect a high standard of care, which includes an aseptic surgical scrub by the surgeon prior to surgery, to prevent them from acquiring a HAI from bacterial contamination of the surgeon’s hands in case the sterile glove perforates.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 3 Continued: Hajipour, Longstaff, Cleeve, Brewster, Bint & Henman, 2006.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 4: Larson, Cimiotti, Haas, Parides, Nesin, Della-Latta, & Saiman, 2005.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
No (Clinical trial using crossover design)
Was random assignment concealed from the individuals enrolling the subjects?
No – potential for bias.
Were the subjects and providers blind to the study group?
No – potential for researcher bias.
Were reasons given to explain why subjects didn’t complete the study?
Yes – 12 nurses withdrew: 9 left the neonatal ICU and 3 no longer wanted to participate.
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes – Data collection periods were March 1, 2001 to January 31, 2002, (year 1), and March 1, 2002, to January 31, 2003, (year 2), with a one month hiatus (February 2002) for product crossover.
Were the subjects analysed in the group to which they were randomly assigned?
N/A
Was the control group appropriate?
Crossover clinical trial – so same
group of nurses used both HH products.
Were the instruments used to measure the outcomes valid and reliable?
Yes – Interrater reliability first established with pilot work & infection data collected simultaneously and independently by a study nurse epidemiologist and each hospital’s nurse epidemiologist. Reliability measured throughout study during meetings between nurse epidemiologist and physician co-investigators (paediatric infectious disease specialist & neonatologist).
To measure transient flora rather than resident flora, nurses cleansed their hands using the assigned product, then a modified glove-juice technique was used for sampling. All microbiologic testing was performed
by the Clinical Microbiology Service of Columbia University Medical Centre, NY to determine microbial counts.
Nurses completed diary cards (1 shift per month) on the frequency of hand hygiene.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
n/a – crossover clinical trial
What are the results?
Ask:
How large is the intervention or treatment effect?
Odds ratio (OR) for alcohol compared with handwashing were 0.98 (95% CI, 0.77-1.25) for any neonatal infection. The mean microbial counts on nurses’ hands (3.21 and 3.11 colony- forming units for handwashing and alcohol respectively, P=0.38, therefore it is not significant as P > 0.05). But the condition of the nurses’ skin was improved using the
alcohol based solution (P=0.02 and P=0.049 for observer and self-assessments respectively).
How precise is the intervention or treatment?
The confidence interval (CI) for the OR includes 1.0, so the intervention is not significant.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes – Neonatal infection rates (bloodstream infections, pneumonia, conjunctivitis, skin and soft tissue infections, central nervous system infections), microbial counts on nurses’ hands and their skin condition.
What are the risks and benefits of the treatment?
There was no significant difference in neonatal infection risk during the use of chlorhexidine gluconate hand wash compared to alcohol based hand rub in this study.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure
Yes. Each phase of hand product was 11 consecutive months (i.e. 11
outcomes & the process of implementing the intervention).
months of using 61% alcohol based hand sanitiser followed by a cross over period, then 11 months of 2% chlorhexidine gluconate).
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
That the result will be significant, i.e. one product (alcohol hand sanitiser 61% or chlorhexidine gluconate 2%) will be more significant in reducing microbial counts on hands and reducing neonatal HAIs than the other.
Table above: Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 4 Continued: Larson, Cimiotti, Haas, Parides, Nesin, Della-Latta, & Saiman, 2005.
Summary of Critical Appraisal Results (Fineout-Overholt et al., 2010 Criteria):
READING 5: Stahl, Morse, & Parks, 2007.
Are the results valid? Ask:
Were the subjects randomly assigned to the intervention and control groups?
No
Was random assignment concealed from the individuals enrolling the subjects?
No – so the personnel conducting microbiologic assessments of samples were blinded to the treatment assignments instead.
Were the subjects and providers blind to the study group?
No
Were reasons given to explain why subjects didn’t complete the study?
n/a
Were the follow-up assessments long enough to fully study the effects of the intervention?
Yes
Were the subjects analysed in the group to which they were randomly assigned?
n/a
Was the control group appropriate?
n/a each subject received both hand products.
Were the instruments used to measure the outcomes valid and reliable?
Yes – independent microbiologist.
Cup scrub methods were used to
obtain cultures; log reductions (reduction factors) of bacterial counts; chemical testing with bleach/colour change.
Were the demographics and baseline clinical variables of the subjects in each of the groups similar?
n/a
What are the results?
Ask:
How large is the intervention or treatment effect?
Iodine log reduction saline soak 3.67 (mean), saline rinse 3.89 (mean).
Chlorhexidine log reduction saline soak 3.20 (mean), saline rinse 3.67 (mean).
35 out of 36 gauze samples had detectable chlorhexidine while no samples had detectable iodine (P < 0.0001). So chlorhexidine is removed by saline-soaked gauze while iodine povacrylex water-insoluble film remains intact under the same conditions.
How precise is the intervention or treatment?
CI not provided.
Will the results help me in caring for my patients? Ask:
Were all clinically important outcomes measured?
Yes except for costs.
What are the risks and benefits of the treatment?
Preventing HAIs with the most effective products to wash hands for hand hygiene.
Is the treatment feasible in my clinical setting? (i.e. has enough time been allowed to measure outcomes & the process of implementing the intervention).
Yes – wound irrigation fluids are common in surgery, so whether saline exposure affects the antimicrobial activity of chlorhexidine and iodine surgical scrub is relevant research, the results of which can be easily implemented into practice.
What are my patients’ and their families’ values and expectations for the outcomes and the treatment itself?
Patients have an expectation that the surgeon will have done an aseptic surgical scrub prior to surgery (even if they don’t know the proper name, they will still have the expectation).
Summary of Findings from Articles.
First Author
Year
Setting
Participants
Interventions & Comparisons In Detail
Outcomes In Detail
Quality of Studies
Girou, E.
2002
3 intensive care units (ICUs) in a French university hospital.
23 healthcare workers.
Handrubbing with (75%) alcohol based solution (n=12) compared to handwashing with antiseptic soap (chlorhexidine gluconate 4%) (n=11) when hand hygiene (HH) was required before & after patient care. Imprints of nurses’ dominant hand were taken on an agar plate to quantify the bacterial count before &
Reduction in bacterial count from imprint before & after HH. Median % reduction in bacterial contamination of hands with handrubbing (alcohol) was 83% vs 58% for handwashing with antiseptic soap. P-value= 0.012. The median duration of HH was 30 seconds for both groups, but this only achieves bactericidal activity in the alcohol hand rub & is inadequate HH for a hand wash with antiseptic soap.
RCT (level II)
after HH.
Hand-rubbing with an alcohol based solution is more effective in reducing bacterial contamination than a hand wash with antiseptic soap, partly due to the inadequate time spent washing hands. 65% of HH lasted less than 30 seconds.
Guilhermetti, M.
2001
Hospital (unknown which one). Took place in same hospital where they isolated the strain of MRSA from
5 healthy adults, 2 males & 3 females, with no skin problems & who worked in a hospital.
The study aimed to compare the effectiveness of 4 hand products for removing hospital strain methicillin-resistant Staphylococcus aureus (MRSA) artificially applied to 5 participants’ hands. In
RF=reduction factor.
FIRST BLOCK:
Significant difference between treatments as P<0.05. 10% PVP-I (RF 3.76) and 70% ethyl alcohol (RF 3.51) had significantly higher removal rates of MRSA than plain liquid soap (RF1.96) and 4%
Experimental:
Latin square statistical design, with two 5×4 randomised blocks.
a hospitalised patient.
each experiment all hand-cleansing agents were tested, and volunteers used all products once. In each experiment one different volunteer acted as a control, whose hands were contaminated but not treated with the hand cleansing product. There was an interval of 7 days between experiments. The first block was light hand contamination and the second block was heavy hand contamination. The
chlorhexidine (RF1.91). SECOND BLOCK: 10% PVP-I (RF4.39) and 70% ethyl alcohol (RF3.27) were significantly more effective in remove MRSA from hands than plain liquid soap (RF1.77) and 4% chlorhexidine (RF1.37; P<0.05). Plain liquid soap was significantly more effective than chlorhexidine 4% detergent in removing MRSA from hands.
hand products tested were plain liquid soap, ethyl alcohol 70%, 10% povidone-iodine liquid soap (PVP-I), chlorhexidine gluconate 4% detergent.
Hajipour, L.
2006
Orthopaedic surgery, in England. No more information provided.
Orthopaedic surgeons.
Overall, 41 surgeries & 82 hand washing episodes were included in the study.
Orthopaedic surgeons were randomly allocated to two different handwashing protocols. All surgeons hand washed for 5 min with chlorhexidine for the first case, and depending on randomisation, washed for 3 min with either alcohol gel or chlorhexidine. The
Four hands (8%) were contaminated in the chlorhexidine group compared to 19 (34%) in the alcohol group. Fisher’s exact test showed that there is a significantly higher risk of contamination using alcohol gel compared to chlorhexidine (P=0.002). The average bacterial colony count was also much higher in the alcohol group
Prospective RCT.
surgeons gloved as per normal and performed surgery. Gloves were then carefully removed and the surgeon’s fingertips placed on an agar plate. An independent microbiologist assessed the bacterial colonies on the agar plates at 24h & 48h of incubation.
(20) compared to the chlorhexidine group (5). There was no relationship between the length of surgery and the degree of contamination (P=1.12, not significant as P > 0.05). Alcohol gel disinfectant is not a suitable alternative to chlorhexidine for handwashing prior to surgery due to the higher risk of bacterial contamination, which may lead to post-op infections in the event of glove perforation.
Larson, E. L.
2005
2 neonatal intensive care units
119 full-time nurses & neonates
Two HH products were tested: traditional antiseptic hand wash
Neonatal infection rates and microbial counts on nurses’ hands were equivalent
Clinical trial using a crossover
including 2932 neonatal hospital admissions. Located in Manhattan, NY.
hospitalised for more than 24 hours.
(chlorhexidine gluconate 2%) & alcohol hand sanitiser (61% ethanol & emollients). Each product was used for 11 consecutive months in each neonatal ICU in random order.
during handwashing with antiseptic soap and handrubbing with alcohol hand sanitiser. Odds ratio (OR) for alcohol compared with handwashing were 0.98 (95% CI, 0.77-1.25) for any neonatal infection. Bloodstream infections, pneumonia, conjunctivitis, skin and soft tissue infections & central nervous system infections were monitored because these represent more than 80% of all HAIs in neonates. The mean microbial counts on nurses’ hands (3.21 and 3.11 colony- forming units for handwashing and
design.
alcohol respectively, P=0.38, therefore it is not significant as P > 0.05). But the condition of the nurses’ skin was improved using the alcohol based solution (P=0.02 and P=0.049 for observer and self-assessments respectively).
Stahl, J. B.
2007
Hospital
36 participants, aged 18 to 65. Exclusion criteria: skin allergies, damaged skin, thyroid condition, antibiotic sensitivities,
Both iodine povacrylex in alcohol and chlorhexidine gluconate in alcohol were applied to the forearms of 36 participants and allowed to dry. The sites were exposed to either saline rinse or saline-saturated gauze. Saline rinse: An
Chlorhexidine is removed by saline-soaked gauze while the iodine povacrylex water-insoluble film remains intact under the same conditions. The implication is that similar results occur in surgery, e.g. wound irrigation.
Randomised blinded study
used tanning beds or chemically treated swimming pools, or are pregnant.
indicator organism was seeded onto the treated sites & after 30 minutes samples were collected. Log reductions calculated for surviving bacterial colonies.
Saline-saturated gauze: this gauze was chemically analysed for the presence of chlorhexidine of iodine.
4. DISCUSSION.
Hand hygiene (HH) is the single most important task health care workers can do to reduce bacterial hand contamination and decrease the rate of cross-infection in a hospital setting, therefore helping to prevent healthcare-associated infections (HAIs) (Girou et al., 2002; Guilhermetti, Hernandes, Fukushigue, Garcia, Cardoso, 2001, p. 107).
The research literature argues that when performing routine patient care in a hospital setting, handrubbing with an alcohol based solution, 75% ethyl alcohol, is significantly more efficient in reducing hand contamination than handwashing with antiseptic soap (chlorhexidine gluconate 4%), partly due to the inadequate time healthcare workers spend on hand hygiene (Girou et al., 2002). Girou et al. (2002) RCT’s intervention and control groups both washed their hands for 30 seconds, which was long enough to achieve bactericidal activity in the intervention group using the alcohol based hand-rub solution, but inadequate for the control group using the chlorhexidine gluconate antiseptic soap to hand wash with water. To achieve bactericidal activity with antiseptic soap and water hand hygiene is required to last 60 seconds.
It is interesting to compare Girou et al. (2002) RCT with Larson et al. (2005) clinical trial (crossover design) as both investigate the effect of antiseptic handwashing versus alcohol based handrubbing on microbial hand counts. As discussed above, while Girou et al. (2002) argue that handrubbing with alcohol based solution (75% ethanol and emollient) is significantly more efficient in reducing hand contamination compared to handwashing with chlorhexidine gluconate 4%, Larson et al. (2005) study concluded that there was no significant difference between the two hand products. However Larson et al. (2005) study used alcohol based solution that was only 61% ethanol and emollients compared to Girou et al. (2002) RCT’s 75%. This would have possibly significantly altered the results. Additionally, Larson et al. (2005) study had several potential biases, in that it was not possible to blind the nurses or the research team
to the study products (investigator bias). Hand hygiene practices were self-reported by the nurses, which is another potential bias. Larson et al. (2005) clinical trial has several limitations (biases), therefore Girou et al. (2002) RCT has more validity as it has been conducted with randomisation, the study design has been well conducted and therefore the results have worth and are significant and useful to practice (Fineout-Overholt, 2010, p. 47). Girou et al. (2002) results support existing literature, including Centre for Disease Control and Prevention Guideline for Hand Hygiene in Healthcare Settings, which recommends using alcohol based hand sanitisers rather than handwashing with antiseptic soap for patient encounters, unless hands are physically soiled (Boyce & Pittet, 2002).
While alcohol based hand sanitiser is the hand product of choice when performing routine patient care (Girou et al., 2002), it is not the product of choice for surgeons to scrub with prior to surgery (Hajipour et al., 2006). Hajipour et al. (2006) conclude that alcohol gel disinfectant is not a suitable alternative to chlorhexidine for handwashing prior to surgery due to the higher risk of bacterial contamination, which can lead to post-op infections in the event of glove perforation. This study is specific to the context of surgeons scrubbing prior to surgery and is not applicable to routine patient care in the general hospital setting. The results of this study are not surprising, due to the residual and cumulative effect of chlorhexidine (Guillhermetti et al., 2001), which is required for long surgeries in the event of glove perforation. Routine patient care encounters are generally short in nature, therefore alcohol based hand sanitiser is sufficient as bactericidal activity is achieved in 30 seconds, hand contamination is reduced, and a residual, cumulative effect is not required (Girou et al., 2002). Hajipour et al. (2006) RCT can only be applied to surgical settings of hospitals, not general wards where routine patient care is performed.
While Hajipour et al. (2006) compared alcohol gel disinfectant and chlorhexidine surgical scrubs, Stahl, Morse & Parks (2007) compared the
effectiveness of chlorhexidine and iodine povacrylex surgical scrubs when they are exposed to saline, which can occur in surgery. Iodine is the more effective hand product as the iodine povacrylex water-insoluble film remains intact when exposed to saline, whereas chlorhexidine is removed under these conditions (Stahl et al., 2007).
Guilhermetti et al. (2001) investigated different methods of hand hygiene and their effectiveness in removing a hospital strain of methicillin-resistant Staphylococcus aureus (MRSA). Not only is 75% ethyl alcohol hand-rub effective in reducing bacterial hand contamination in routine patient care where the infectious status of patients is unknown, but it is also effective in removing MRSA (hospital strain) from lightly or heavily contaminated hands (Guilhermetti et al., 2001). Guilhermetti et al. (2001) study found that ethyl alcohol 70% and PVP-I 10% were significantly more effective than 4% chlorhexidine and plain liquid soap in removing MRSA from contaminated hands. The study also found that plain liquid soap was significantly more effective than chlorhexidine 4% detergent in removing MRSA from contaminated hands. However the authors cited a bias in results due to significant differences in chlorhexidine preparations (since it is formula dependent) from different manufacturers, which could have explained the poor reduction factors of chlorhexidine gluconate compared to liquid soap (Guilhermetti et al., 2001). Guilhermetti et al. (2001) noted the conflicting evidence in the research literature regarding 4% chlorhexidine gluconate use with infectious patients. For example, while Guilhermetti et al. (2001) conclude chlorhexidine gluconate 4% to be not very effective in removing MRSA, Wade (as cited in Guilhermetti et al., 2001, p. 107) showed that it was a very reliable hand antiseptic and better than plain soap in removing HAI vancomycin-resistant Enterococcus faecium and gentamicin-resistant Enterobacter cloacae from contaminated fingertips.
5. RECOMMENDATIONS.
The research literature has shown the tendency of health care workers to spend an inadequate amount of time washing their hands with antiseptic soap when performing hand hygiene (Girou et al., 2002). Therefore in the hospital setting for routine patient care an alcohol based solution, importantly at least 75% ethanol and emollients, for handrubbing is recommended due to its rapid efficacy (30 seconds) and higher reduction in bacterial contamination compared to chlorhexidine gluconate 4% (Girou et al., 2002). An alcohol based hand-rub not only fits conveniently into the time pressures and heavy workload of the current hospital setting, but decreases the rate of cross-infection and HAIs (Girou et al., 2002).
Not only is 75% ethanol and emollient hand-rub recommended for routine patient care where the infectious status of patients is unknown, but it is also recommended for care of MRSA patients (Guilhermetti et al., 2001). PVP-I 10% is also effective in removing MRSA from hands, however 75% alcohol hand rub is recommended as it will not stain health care workers hands like PVP-I 10%, which is more ‘user-friendly’.
It should be noted that alcohol based solution is not appropriate for surgeons to scrub with prior to surgery. Chlorhexidine gluconate is better than alcohol based surgical scrub due to its residual effect, which reduces bacterial contamination, decreasing the risk of post-operative infections in the event of a glove perforation (Hajipour et al., 2006). However, if exposure to saline is likely during the surgery (e.g. wound irrigation); iodine povacrylex should be used due to its water insoluble nature, whereas the chlorhexidine will be removed by any saline exposure (Stahl et al., 2007).
To implement a hand hygiene protocol where 75% alcohol based solution for handrubbing is used, a multifactorial approach is required. The alcohol based solution could be mounted onto the wall at several patient-care points throughout the hospital unit. As well as supplying and providing easy access to the 75% alcohol based hand-rub
solution, wall posters should be mounted educating staff about the importance of hand hygiene, as well as running education seminars on the single importance of hand hygiene in preventing cross-infection and preventing HAIs (Girou et al., 2002). This implementation process needs the organisational support of the Nurse Unit Manager and the Hospital Administration to be successful.
6. REFERENCES.
•Boyce, J. M., & Pittet, D. (2002). Guideline for hand hygiene in health-care settings: recommendations of the Healthcare Infection Control Practices Advisory Committee and the HIPAC/SHEA/ APIC/ IDSA Hand Hygiene Task Force. American Journal of Infection Control, 30, S1-S6.
•Fineout-Overholt, E., Melnyk, B. M., Stillwell, S. B., & Williamson, K. M. (2010). Critical appraisal of the evidence: part I. American Journal of Nursing, 110(9), 47-52.
•Fineout-Overholt, E., Melnyk, B. M., Stillwell, S. B., & Williamson, K. M. (2010). Critical appraisal of the evidence: part II. American Journal of Nursing, 110(9), 41-48.
•Girou, E., Loyeau, S. Legrand, P., Oppein, F., & Brun-Buisson, C. (2002). Efficacy of handrubbing with alcohol based solution versus standard handwashing with antiseptic soap: randomised clinical trial. British Medical Journal, 325, 362-366.
•Guilhermetti, M., Hernandes, S. E. D., Fukushigue, Y., Garcia, L. B., & Cardoso, C. L. (2001). Effectiveness of hand-cleansing agents for removing methicillin-resistant staphylococcus aureus from contaminated hands. Infection Control and Hospital Epidemiology, 22(2), 105-108. doi: 10.1086/501872
•Hajipour, L., Longstaff, L., Cleeve, V., Brewster, N., Bint, D., & Henman, P. (2006). Hand washing rituals in trauma theatre: clean or dirty? The Royal College of Surgeons of England, 88(1), 13-15. doi: 10.1308/003588406X83032
•Larson, E. L., Cimiotti, J., Haas, J., Parides, M., Nesin, M., Della- Latta, P., & Saiman, L. (2005). Effect of antiseptic handwashing vs alcohol sanitizer on health care-associated infections in neonatal intensive care units. Archives of Pediatric and Adolescent Medicine, 159, 377-383
•Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G, & The PRISMA Group. (2009). Preferred reporting items for systematic reviews
and meta-analyses: the PRISMA statement. PLoS Med, 6(6): e1000097. doi: 10.1371/journal.pmed1000097
•Stahl, J. B., Morse, D., & Parks, P. J. (2007). Resistance of antimicrobial skin preparations to saline rinse using a seeded bacteria model. American Journal of Infection Control, 35(6), 367- 373. doi: 10.1016/j.ajic.2006.08.015