Ignaz Semmelweis is widely regarded as the father of hand hygiene. He observed in 1846 that the mortality rate as a result of puerperal sepsis significantly reduced if physicians cleaned their hands with a chlorine-based solution between each patient (Noakes et al, 2008). Since this discovery, hand hygiene has become widely accepted as one of the most effective interventions to reduce infection transmission and, more recently, to decrease antibiotic-resistant infection rates (Mitsuboshi and Tsugita, 2019).
Hand hygiene in healthcare settings is important both to reduce nosocomial infection rates and, in the case of veterinary medicine, the risk of zoonotic disease transfer to staff. A study looking at the infection control practices and zoonotic disease risks among veterinarians in the USA concluded that most USA-based veterinarians do not engage in practices that may help reduce zoonotic disease transmission (Wright et al, 2008). This study by Wright et al, found that only 55.2% (590/1069) of participants reported always washing their hands before eating, drinking, or smoking at work, and only 48.4% (516/1066) reported always washing their hands between patient contacts (Wright et al, 2008). A questionnaire-based study of private veterinary clinics in the USA also concluded that the hand hygiene practices among veterinary professionals was inadequate, with less than half of the respondents (76/182) reporting washing their hands regularly between patients (Nakamura et al, 2012). These studies, however, did not investigate the efficacy of a hand hygiene event when it was performed.
The efficacy of alcohol-based hand sanitiser (ABHS) has been associated with both the alcohol concentration and the degree of skin coverage. A study into the volume of ABHS required to completely cover both hands found that volumes <2 ml yielded high rates of incomplete coverage (67–87%) whereas volumes ≥ 2 ml gave lower rates (13–53%) (Kampf et al, 2013). This study also found that application of 1.1 ml of 70% volume per volume (v/v) ethanol rubs reduced bacterial contamination by a maximum of 1.85 log10, which fails the US Food and Drug Administration (FDA) efficacy requirement of at least 2 log10. Application of 2 ml of the 85% weight per weight (w/w) ethanol rub reduced contamination by 2.06 log10, thus fulfilling the US FDA efficacy requirement (Kampf et al, 2013).
Faulty hand sanitisers are therefore, a major potential cause of ineffective hand sanitation. Previous studies into human healthcare establishments have found a large variation in the amount of gel dispensed by hand sanitisers — with a range of 0.5–6.0 ml in one study of acute hospital settings (Dalziel et al, 2018). This also raises a challenge for some monitoring practices, which use the volume of sanitiser used as a proxy for the number of usages of the sanitiser (Dalziel et al, 2018). A separate survey into the function of manual hand sanitiser dispensers in a human healthcare setting found that at a single time point only 77% were functional, and of those only 65% dispensed the sanitiser after a single stroke of the lever, with 13% requiring four or more strokes to before dispensing sanitiser (Kohan et al, 2002).
Effective hand sanitation, therefore, requires compliance by healthcare personnel, a correctly functioning dispenser and an appropriate volume of an efficacious alcohol gel or rub. If these criteria are not met, then the risk of ineffective sanitation presents a number of risks to both staff and patients. This includes a risk of nosocomial infections, zoonotic disease transfer, and an increase in antibiotic resistant strains of infection (Mitsuboshi and Tsugita, 2019). The aim of this study was, therefore, to investigate the amount of sanitiser dispensed, and to assess if this is fulfilling the requirement for effective sanitation. A second aim was to assess if the allocation of a specific role in maintaining the hand sanitisers improved their efficacy.
Materials and methods
The study was conducted between March 2018 and December 2018 in The Royal (Dick) School of Veterinary Studies Teaching Hospital, which is a large multidisciplinary referral and general practice hospital. The hospital is also a teaching facility with residents, interns and final year veterinary students. Therefore, with large numbers of staff and students, effective hand hygiene is important. Approximately 70 Purell White LTX-12 Touch Free Wall Mounted Pump Dispensers are distributed throughout the Small Animal Hospital. Purell Advanced Hygienic hand rub 70% ethyl alcohol sanitiser cartridges was used in all dispensers. A convenience sample of 15 sanitisers, based in separate rooms within the clinical floor, were used in the study. The Hospital for Small Animals uses automated motion activated sanitisers standardised to dispense a fixed volume of 1.2 ml, therefore a minimum of two aliquots would be required to fulfil the FDA requirements. Therefore, paired aliquots were collected onto a pre-weighed polyethylene grip seal bag (Figure 1). At the time of aliquot collection any perceived malfunctions were recorded. These were recorded as: no problems; empty; not empty but malfunctioning; or gel dispensed to side.
A fresh sample bag of the same weight was used for each sample and each sample was weighed immediately after obtaining it. A Myweigh i201 commercial scale was used to weigh the bag and sanitiser to the nearest 0.001 g, and the weight of each aliquot was calculated and recorded (Figure 2). The scale was in a central position to the sanitisers, which were carried to it for weighing, negating the need to recalibrate the scales for each aliquot. The mean volume/mass was calculated using seven paired aliquots and this value was then used to calculate the volume of paired aliquots in the data collection.
Aliquot collection was performed for 6 consecutive days at an initial time point (time point 1). Following this, a role was assigned to the animal care assistant team to routinely monitor and service the sanitisers, which included replacing the gel, cleaning the nozzle and removing any dried secretions. Sample collection was then repeated a second time immediately following the introduction of this role (time point 2) and again 8 months after the implementation of the role (time point 3).
Statistical analysis was performed using Minitab® Statistical Softwarea. Data were assessed for normality using an Anderson-Darling Analysis, and a p-value of <0.05 was considered significant. A one-way ANOVA was used to assess for a significant difference in the weight of sample at the three time points. A one-sample t-test was used to assess if a significant difference was present between the volume dispensed and the target of 1.2 ml. A Chi-squared table was used to assess if there was a significant change in the number of aliquots <1 ml at each time point, and to assess if a significant difference was present between aliquots of <1 ml and the number of perceived malfunctions when compared with aliquots ≥1 ml and the number of perceived malfunctions.
Results
Two hundred and seventy aliquots (135 pairs) were collected. The mean volume for all aliquots was 1.091 ml, with a standard deviation of 0.302 ml. The mean volume dispensed was 1.107 ml (0.9744 g) at time point 1, 1.085 ml (0.9553 g) at time point 2, and 1.082 ml (0.9518 g) at time point 3. No significant difference was found between the mean values at the three time points (p=0.829). A significant difference was found between the volume dispensed and the target of 1.2 ml (mean = 1.0912 ml, standard deviation (SD) = 0.3019, p<0.001). No significant difference was found in the number of aliquots <1 ml at the three time points (p=0.327). Of the 135 paired aliquots, there were 69 perceived malfunctions (Table 1 and 2). A significant difference was present between aliquots of <1 ml and the number of perceived malfunctions when compared with aliquots ≥1 ml and the number of perceived malfunctions (p<0.001).
Table 1. The number of perceived malfunctions for aliquots <1 ml and ≥1 ml
| <1 ml | ≥1 ml | |
|---|---|---|
| Gel dispensed to the side | 18 | 20 |
| Not empty but malfunctioning | 23 | 5 |
| Empty | 3 | 0 |
Table 2. The number of perceived malfunctions for aliquots <1 ml and ≥1 ml
| <1 ml | ≥1 ml | |
|---|---|---|
| Malfunction | 44 | 25 |
| No malfunction | 10 | 191 |
Discussion
This study addresses, for the first time, the efficacy of hand sanitiser dispensers in a veterinary healthcare setting. In this study, 20% of samples (paired-aliquots) fell below the volume previously shown to reduce the risk of incomplete hand coverage. This is consistent with previous studies assessing sanitiser dispenser function in human hospitals. One previous study found that 33% of manual hand sanitisers in a human healthcare setting were non-functional when surveyed, and of the functional dispensers 17% delivered an inadequate volume of sanitiser (estimated to be ≤1 ml) (Kohan et al, 2002). A separate study also found a significant variation from the assumed volume of 3 ml, with the median volume of ABHS per hand hygiene event of 1 ml (Dalziel et al, 2018). In the current study, the mean volume dispensed was significantly different to the target volume of 1.2 ml/aliquot (p=1.05x10-8), however, was above the 1 ml/aliquot value with a mean volume of 1.091 ml. This shows that although, on average, the volume of a paired aliquot will meet the requirement to reduce the risk of incomplete hand coverage, this may result in ineffective sanitation 20% of the time.
Perceived issues were documented in 69 paired-aliquot collections, and there were no perceived issues with 201 paired-aliquots. The most commonly identified malfunction differed between aliquots of ≥1 ml and <1 ml. In the group of aliquots ≥1 ml, the most commonly noted malfunction was a directional squirt, which suggests a partial obstruction of the nozzle that still allows the majority of the aliquot to be dispensed. Depending on the variation from the expected direction this poses the potential to miss a proportion of the aliquot. Once the bag was held under the sanitiser it was not adjusted to allow for movement of the jet, therefore, this method may be less likely to catch the jet than if someone was able to sense the aberrant jet and move their hand accordingly. In a previous study of manual hand sanitiser dispensers within a human hospital, 13% delivered product to the palm, 71% squirted the alcohol rinse between the fingers of the surveyors, and 16% squirted the product away from the hand onto the wall or floor (Kohan et al, 2002).
In the group of aliquots <1 ml, the most common malfunction was a perceived reduction in the volume of gel dispensed, often associated with increased dispenser noise, even though the dispenser was not empty. Three times an empty dispenser was noted, all occurred at time point 3, and involved two different dispensers, all correlating with a volume <1 ml. There was no statistical significance in the volume of ABHS dispensed at any of the three time points and the mean aliquot volume was significantly lower than the target volume (1.2 ml/aliquot). This suggests that the allocation of a specific role in servicing the sanitisers did not influence their function. The introduction of individualised feedback from emerging technologies may improve dispensed volumes of ABHS. A study by Scheithauer et al found that following the phased introduction of Wi-Fienabled hand sanitisers, the mean volume of ABHS increased from less than 50% of the 3 ml target to 2.66 ml (Scheithauer et al, 2018).
The number of perceived malfunctions was significantly higher when volumes fell below 1 ml/aliquot. This suggests that the user may be aware of the failure to dispense a sufficient volume. In multiple studies looking at behaviours of healthcare professionals and their use of hand sanitisers, convenience was a major factor in compliance (Kirk et al, 2016; Sadule-Rios and Aguilera, 2017). This suggests that the likelihood of someone noticing a perceived malfunction and sourcing alternative gel may be low, but has not been investigated in the current study. The association between inadequate volumes dispensed and a perceived malfunction at the time of dispensing suggests that if a simple user reporting system was developed, then this may reduce dispenser malfunctions.
Although hand hygiene has been shown to be a major factor in reducing hospital acquired infections and the spread of antibiotic resistant bacteria (Mitsuboshi and Tsugita, 2019), a multimodal approach is required to improve its efficacy. Compliance in the use of sanitisers remains a major factor and poor compliance rates have been identified in several studies (Scheithauer et al, 2009; FitzGerald et al, 2013; Pittet et al, 2000; Kirk et al, 2016). There are numerous factors associated with the compliance of sanitiser use, including but not limited to: knowledge of the transmission risk and impact of healthcare associated infections; accessibility of sanitisers; social pressures; perceived benefits; skin tolerance; and the effect of residue on the skin (Allegranzi and Pittet, 2009). The implementation of hand hygiene programmes in veterinary settings have achieved mixed results. A study into poster intervention on hand hygiene compliance found limited efficacy (Anderson et al, 2014), and a separate study into the effectiveness of a hand hygiene educational campaign at a small animal veterinary teaching hospital found no improvement in hand hygiene adherence or use of hand sanitisers (Smith et al, 2013). In contrast, a study with a lower baseline hand hygiene level found a significant increase in the proportion of interactions meeting the criteria for proper hand hygiene practices after an educational hand hygiene programme, with participants 4.1 times as likely to adhere to correct hand hygiene practices after the intervention than before (Shea and Shaw, 2012).
One of the major perceived issues in this hospital environment is education regarding volumes of sanitiser. With the current dispensers, a minimum of two aliquots are required to achieve an adequate volume. As user convenience is important (Kirk et al, 2016; Sadule-Rios and Aguilera, 2017), and increased time is needed to dispense a second aliquot, the propensity for users to obtain two may be reduced. A study by Dalziel et al looking at hand hygiene moments in Scottish hospitals found the mean number of dispenses per ABHS application were 1.37 (Dalziel et al, 2018), therefore this warrants further investigation as users are frequently observed obtaining single aliquots in the current study's hospital.
This study does not take into account the variance in individual hand sizes, the number of aliquots used by individuals, and if a perceived malfunction will prompt the user to source gel elsewhere. It also does not assess the correlation between volume dispensed and bacterial decontamination. Further studies looking into the number of aliquots used, the perception of percentage hand coverage and the actual percentage hand coverage, alongside correlating this with actual bacterial load would be useful. The limitations of this study lie in the lack of standardisation in ages of the sanitiser used, and the lack of subtyping of malfunctions (e.g. clogged nozzle, bag folded, viscosity of the gel). In addition, the hospital has around 70 Purell advance automated dispensers, and only a subset, which were all in clinical areas, were included in data collection. The sanitiser was carried to the scales prior to weighing, which may have induced some mild weight change with evaporation, however the sanitisers had limited variance in their distance to the scales, and there was no visible movement of the gel on the bag.
Conclusion
Of paired aliquots, 20% fell below the volumes previously shown to reduce the risk of incomplete hand coverage. The institution of a servicing role did not result in a significant improvement in the volume of gel dispensed. This study highlights that improving hand hygiene requires a holistic approach, including availability, compliance and correctly functioning dispensers. This study also suggests that if a protocol is put in place to ensure functionality of the dispensers, then its efficacy should be audited.
KEY POINTS
- Hand hygiene is critically important to reduce nosocomial disease transmission to patients, and zoonotic disease transmission to staff.
- The efficacy of alcohol-based hand sanitiser (ABHS) has been associated with both the alcohol concentration and the degree of skin coverage.
- 20% of paired hand sanitiser aliquots fell below the recommended volume for hand coverage.
- Instituting a role in cleaning and servicing the sanitisers did not result in a significant improvement in the volume of gel dispensed.
- Low volumes, <1ml, of dispensed hand sanitiser, are more likely to be associated with a perceived malfunction than those >1ml.