Journal of Hospital Infection xxx (2014) 1e8 Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander E.L. Best a, P. Parnell a, M.H. Wilcox a, b, * a b
Microbiology Department, Old Medical School, Leeds General Infirmary, Leeds Teaching Hospitals NHS Trust, Leeds, UK University of Leeds, Leeds, UK
A R T I C L E
I N F O
Article history: Received 27 November 2013 Accepted 7 August 2014 Available online xxx Keywords: Air driers Bacteria Droplet dispersal Environmental contamination Hand drying Hand hygiene
S U M M A R Y
Background: The efficiency of hand drying is important in preventing pathogen spread, but knowledge surrounding which drying methods contribute least towards contamination of the environment and users is limited. Aim: To compare the propensity of three common hand-drying methods (jet air, warm air dryers, and paper towels) to contaminate the environment, users, and bystanders. Methods: Hands were coated in lactobacilli to simulate poorly washed, contaminated hands, and dried. The investigation comprised 120 air-sampling tests (60 tests and 60 controls), divided into close and 1 m proximity from the drying process. Separate tests used hands coated in paint to visualize droplet dispersal. Findings: Air bacterial counts in close proximity to hand drying were 4.5-fold higher for the jet air dryer (70.7 cfu) compared with the warm air dryer (15.7 cfu) (P ¼ 0.001), and 27-fold higher compared with use of paper towels (2.6 cfu) (P < 0.001). Airborne counts were also significantly different during use of towel drying versus warm air dryer (P ¼ 0.001). A similar pattern was seen for bacterial counts at 1 m away. Visualization experiments demonstrated that the jet air dryer caused the most droplet dispersal. Conclusion: Jet air and warm air dryers result in increased bacterial aerosolization when drying hands. These results suggest that air dryers may be unsuitable for use in healthcare settings, as they may facilitate microbial cross-contamination via airborne dissemination to the environment or bathroom visitors. ª 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction Hand hygiene is a fundamental component for controlling the spread of infection.1,2 Promotion of improved hand hygiene * Corresponding author. Address: Microbiology Department, Old Medical School, Leeds General Infirmary, Leeds Teaching Hospitals NHS Trust, Leeds LS1 3EX, UK. Tel.: þ44 113 392 6818; fax: þ44 113 392 2696. E-mail address: [email protected] (M.H. Wilcox).
is recognized as an important public health measure. There is much emphasis on the correct method for handwashing, but less so concerning the options for drying hands. Evidence suggests that efficiency of hand drying is important in the prevention of the transfer of micro-organisms from person to person or to the environment.3 However, the risk of aerosolizing micro-organisms during hand drying by various methods remains unclear. Methods for hand drying vary widely and include paper or cloth towels, warm air dryers or jet air dryers either singly or in
http://dx.doi.org/10.1016/j.jhin.2014.08.002 0195-6701/ª 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
2
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8 airborne bacteria. Experiments were carried out over a period of six weeks.
combination. Drying with towels may remove remaining microorganisms through friction, while moisture is wicked away into the absorbent material. Warm air dryers evaporate moisture and remove some micro-organisms during hand rubbing, although this process may take too long for efficient use, with hands consequently remaining damp. Newer jet air dryers rely on the passage of high speed, usually unheated, air to dry hands without rubbing, typically in 15 s.4 The selection of drying method may depend upon a number of factors including practicality, space availability, cost, or personal preference. Infection prevention considerations may influence the choice of hand-drying method, but the evidence base is weak to make informed decisions. Notably, the recent National Health Service (NHS) building guidance states that ‘Hot-air hand dryers reduce paper waste and may be considered for use in public areas of healthcare facilities, but should not be installed in clinical areas as they are noisy and could disturb patients’.5 It is clearly desirable to ensure that the process of hand drying does not increase the potential for micro-organism transmission, either directly to another person or indirectly by contamination of the bathroom environment. Evidence concerning whether hand-drying methods differ in their propensity to aerosolize, and so transmit microorganisms, is contradictory.6e11 Some studies suggest that drying hands with warmed air is associated with increased aerosolization of micro-organisms.11 However, others have suggested that there is little difference in aerosolization for the different drying methods.6 One of the reasons for discrepant results may be the use of relatively insensitive methods or experimental designs that fail to detect real differences. In this study, we aimed to compare the propensity of three widely used methods of hand drying (jet air dryer, warm air dryer, and paper towel drying) to contaminate the environment with bacteria via aerosolization during the drying process. Additionally, we investigated the extent of possible contamination of people drying their hands, as well as the possibility of contamination of a bystander.
Two electric hand dryers, a jet air dryer (Dyson Airblade) and warm air dryer (Pro-Elec GSQ250B), and a paper towel dispenser (Tork H3 classic dispenser containing Tork Advanced Towels, MRT213) were mounted on one wall at manufacturerrecommended heights for use. Air was collected using two Airtrace Environmental portable samplers (Biotrace, Microbial Contamination Control, Runcorn, UK) via a 1 m long Tygon tube placed at the left- and right-hand sides (LHS, RHS) of each drying unit, in close proximity and also at 1 m away. As air entered the sampler (28.3 L/min), it was forced through a fine slit (44 0.152 mm) at a velocity of 70 m/s thereby causing particulate matter (minimum size 0.4 mm) to impact upon a 140 mm diameter, lactobacillus-selective, agar plate (LAMVAB agar, Bioconnections, Wetherby, UK, supplemented with vancomycin). The plate rotated constantly from a known start point throughout the sampling period; thus, following culture, the location of the colonies represents the time of recovery from air. Following each test, the two plates (from the LHS and RHS) were immediately transported to the laboratory and incubated in an anaerobic cabinet at 37 C for 48e72 h, and colonies were then counted. Data from each drying method were compared using ManneWitney U-test (SPSS, IBM, Armonk, NY, USA) using a 5% confidence level to determine significance. In addition, four settle plates containing LAMVAB agar were sited at various locations around each dryer (Figure 1). At the end of each sampling session, the air-sampling machine was cleaned externally and internally using a disinfectant (Trigene, Medichem, Queenborough, UK) and run on a purge cycle to decontaminate the machine and tubing, as recommended by the manufacturer. Floor surfaces, the wall area around the dryers, and the dryer units were also thoroughly cleaned between tests using Trigene.
Methods
Testing procedure
Organization of air sampling
Air-sampling tubes were clamped at a height of 1.2 m (from floor level) on the LHS and RHS of the area where the drying of lactobacilli-contaminated gloved hands was taking place, in the outer edges of the airstream produced by the jet and warm air dryers. To ensure reproducibility, this was the same relative position for each of the hand-drying methods. For the tests 1 m away from the hand drying, the ends of the air-sampling tubes were clamped at the same height as before. The air samplers were switched on and hands were then immediately dried. For the warm air dryer, hands were rubbed together for 30e40 s until dry; for the jet air dryer, hands were placed into the unit and slowly drawn up until dry (15 s); for towel drying, four paper towels were taken from the dispenser and these were rubbed over hands until dry (15 s). The air samplers were left running for 15 min following each hand-drying process.
All tests were carried out in a room measuring 65 m3 with the door closed throughout experiments. Room air was maintained by standard ventilation without air-conditioning or negative or positive pressure ventilation. For each test, gloved hands were first coated by immersion in a suspension of lactobacilli (107 cfu/mL) that were cultured from a proprietary yoghurt (Actimel, Danone, Paris, France) and then dried in a standardized manner using one of each of the three drying methods: a warm air dryer, a jet air dryer, or paper towels. In total, this part of the study comprised 120 air-sampling tests (60 tests and 60 controls) with a total testing time of 15 min per test (comprising up to 40 s of drying time dependent on the method used and the remainder being the air-sampling time). The 60 air-sampling tests of contaminated hands comprised 20 collections for each drying method, of which 10 were in close proximity to the hands being dried and 10 were 1 m away. Control air sampling was carried out before every hand-drying test (with no hand drying taking place) both to provide baseline measurements and to minimize the risk of carryover of
Air-sampling process
Visual determination of the extent of contamination of the environment, user, and bystander To visualize the extent of potential contamination occurring during each drying process within the environment, and on to a
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
Left-hand side
Hand drier/paper towel dispenser mounted on wall
3
Right-hand side
Immediate vicinity (10 cm from each side of hand drier)
1 m away from hand-drying area
2 m away from hand-drying area
Figure 1. Locations of air-sampling tubes and settle plates. Diagram showing position of hand dryer and positions of the ends of the airsampling tubes for each of the three tests (black circles indicate the position of the end of the air sampler tubes) (mounted at a height of 1.2 m), white circles indicate position of settle plates on the floor.
user and a bystander, experiments were repeated as above. However, for each test, gloved hands were coated in a solution of black water-based paint (instead of lactobacilli), with the user wearing a white disposable Tyvek suit (DuPont, Stevenage, UK); suits were worn backwards so that the hood covered the face. Additionally, a bystander (wearing a suit in the same manner) stood diagonally adjacent to the dryer user (1 m distance) in order to replicate the scenario of someone waiting to dry his/her hands. In total, this user and bystander study comprised 30 drying tests (10 tests for each drying method). Following drying, the potential contamination within the environment was visualized and enumerated by the distribution of black paint splatters around each drying unit, and the distance travelled by the paint spots away from the drying unit/area was measured. The extent of potential contamination on the user and bystander was quantified by counting the numbers of paint spots on a number of predetermined areas marked out on the Tyvek suits (Figure 2).
Results Airborne lactobacilli counts during hand drying For each drying method, the airborne bacterial counts were very similar for samples collected on the LHS versus RHS of the subject/dryer. Table I shows the combined LHS and RHS mean counts (n ¼ 20) for each drying method. The air bacterial counts in tests performed in close proximity to hand drying
Figure 2. White disposable Tyvek suit showing predetermined areas used for counting paint spots.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
4
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
Table I Mean lactobacilli counts (cfu) in the air after 15 min sampling for each drying method (n ¼ 10) Distance
Close proximity 1 m away
Paper towels
Warm air dryer
Jet air dryer
LHS (n ¼ 10)
RHS (n ¼ 10)
Combined (n ¼ 20)
LHS (n ¼ 10)
RHS (n ¼ 10)
Combined (n ¼ 20)
LHS (n ¼ 10)
RHS (n ¼ 10)
Combined (n ¼ 20)
2.6 (2.2) 2.4 (2.7)
2.6 (1.7) 2 (2.1)
2.6 (1.9) 2.2 (2.4)
14.9 (14.5) 18.2 (4.7)
16.5 (16) 19.1 (7.9)
15.7 (14.9) 18.6 (6.4)
76.2 (40) 89.1 (34.2)
65.2 (37.5) 89.9 (37.6)
70.7 (38.2) 89.5 (34.9)
LHS, left-hand side; RHS, right-hand side. Values are mean (SD).
were 4.5-fold higher for the jet air dryer (70.7 cfu) compared with the warm air dryer (15.7 cfu) (P ¼ 0.001) and 27-fold higher compared with use of paper towels (2.6 cfu) (P < 0.001). Airborne counts were also significantly different during use of towel drying versus warm air dryer (P ¼ 0.001). A similar pattern was seen in mean bacterial counts recovered from air collected 1 m away from hand drying; bacterial air counts recovered following use of the jet air dryer, warm air dryer, and paper towels for hand drying were 89.5, 18.7, and 2.2 cfu, respectively. No lactobacilli were recovered in any control air-sampling experiments. The method enabled measurement of when bacteria were recovered from air during the 15 min sampling periods. Using the above test results, for the jet air dryer the mean proportions of bacteria recovered in the first, second, and third 5 min sectors (LHS and RHS, at all time-points, close to the dryer) were 52%, 28%, and 20%. The corresponding data for the warm air dryer were 50%, 24%, and 26%, and for the paper towels were 46%, 39% and 15%. Similar trends were seen for the bacteria recovered 1 m away from hand drying.
Settle plate lactobacilli counts Settle plates located under each drying unit yielded the greatest mean bacterial counts (Table II). Mean bacterial counts for the 1 m distant settle plates (LHS and RHS results) were combined, and results for the 10 tests at close proximity and 1 m away were combined (n ¼ 20) as the positions of settle plates remained the same. For the settle plates in close proximity to each drying method, counts under the dryer were the highest for the warm air dryer (190 cfu) compared with the jet air dryer (68.3 cfu) (P < 0.001) and the paper towel drying (11.9 cfu) (P < 0.001). Counts on settle plates 1 m away for the warm air dryer (7.8 cfu) were also higher than those for the jet air dryer (2 cfu) (P < 0.001) and the paper towels (0.7 cfu) (P < 0.001). There was no significant difference for the counts
for the one settle plate located 2 m away for the warm air dryer and jet air dryer (both 1.4 cfu) (P ¼ 0.940). There was a significant difference between the counts on settle plates for the paper towels (0.4 cfu) when compared to the warm air dryer (P ¼ 0.040) and jet air drier (P ¼ 0.008). All settle plates for the control experiments were negative.
Visualization tests to determine the extent of contamination of the environment Figures 3e5 show the distribution of paint following separate drying episodes with hands coated in paint. For the jet dryer, paint spots were found up to 120 cm away from the RHS of the unit, 140 cm away on the LHS, 30 cm above the unit, and 85 cm along the floor forwards towards the user (Figure 6). For the warm air dryer, the maximum distance travelled by paint spots was 30e40 cm on each side, 130 cm downwards along the wall and 40 cm forwards along the floor towards the user (Figure 7). For the hand towel drying process, markedly fewer paint spots were visualized; a few were seen on the towel dispenser and on the wall below, and some on the floor presumably due to dripping while rubbing hands (Figure 8). Table III shows mean numbers of paint spots detected during hand drying (n ¼ 10) on each of the demarcated areas of the body. No paint spots were seen on suits worn by the user or bystander during hand drying using towels. Numbers of spots on the user of the jet air dryer compared with the warm air dryer were overall higher for all body areas, except for both arm counts (mean counts: LHS arm 27.4 vs 14.8, P ¼ 0.061; RHS arm 8.7 vs 3.4, P ¼ 0.02). For both the jet air and warm air dryers, spots predominated in the upper body area, with the numbers
Table II Mean bacterial counts (mean cfu) on settle plates Drying process
Paper towels Warm air dryer Jet air dryer
Position of settle plates Under dryer (n ¼ 20)
1 m awaya (n ¼ 40)
2 m away (n ¼ 20)
11.9 190 68.3
0.7 7.8 2
0.4 1.4 1.4
Results for 10 tests at close proximity and 1 m away were combined as positions of settle plates remained the same. a Left- and right-hand-sided counts combined.
Figure 3. Visualization of paint spots following drying with the jet air dryer.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
70 cm
140 cm
5
30 cm 76 cm
Jet air drier
120 cm Wall Floor
100 cm
90 cm 85 cm
Figure 6. Furthest distance travelled by paint spots following one drying process with the jet air dryer. Black arrows indicate distance along the wall on which the unit was mounted; red arrows indicate distance forward along the floor.
Figure 4. Visualization of paint spots following drying with the warm air dryer.
of spots from the jet air dryer being significantly higher [mean counts of 144.1 and 65.8 (P ¼ 0.0005)]. Spot counts on upper legs were higher after use of the jet air compared with the warm air dryer [RHS mean counts 15 versus 1.6 (P ¼ 0.307) and LHS 15.8 versus 3.4 (P ¼ 0.006)]. Similar spot counts were seen on the lower legs (RHS mean counts 6.6 versus 4.3 (P ¼ 0.242) and LHS 6.3 versus 2.8 (P ¼ 0.794). Relatively few spots were detectable on the head area, with no significant difference between the mean counts for the jet air dryer and the warm air dryer [1.2 versus 0.5 (P ¼ 0.149)]. Overall, the mean number of
spots was significantly higher for the jet air dryer (mean counts 230.8) compared with the warm air dryer (135.1) (P ¼ 0.004). The number of paint spots detectable on the bystander was generally low for the air dryers (0.1e0.8). The body areas of the bystander where we recorded paint spots after using the jet air dryer were both arms (mean counts 0.1 for both), top of the right leg (mean count 0.8), and both lower legs (mean counts 0.3 for both). After using the warm air dryer, spots were most frequently seen on the head (mean counts 0.2), upper body (mean count 0.2), both upper legs (mean counts 0.2 and 0.4), and right lower leg (mean count 0.5). Overall, there was no significant difference when comparing the mean number of spots on the bystander with the jet air and warm air drying approach [mean counts 1.6 versus 1.5 (P ¼ 0.548)].
Discussion Hand drying is an integral part of the hand hygiene process, which aims to optimize the removal of potentially pathogenic
30 cm
Warm air drier
130 cm
40 cm
Wall Floor
40 cm
Figure 5. Visualization of paint spots following drying with the hand towels.
Figure 7. Furthest distance travelled by paint spots following one drying process with the warm air dryer. Black arrows indicate distance along the wall on which the unit was mounted; red arrow indicates distance forward along the floor.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
6
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
Paper towel dispenser 20 cm
Wall 20 cm
Floor
Figure 8. Furthest distance travelled by paint spots following one drying process with the paper towels. Black arrows indicate distance along the wall on which the unit was mounted; red arrow indicates distance forward along the floor.
micro-organisms that may be acquired during toileting and use of bathrooms. The published evidence concerning whether hand-drying methods may differ in their propensity to aerosolize and so transmit micro-organisms is contradictory.6e11 Some studies suggest that drying hands via warmed air is associated with increased aerosolization of micro-organisms, and others suggest there to be no difference.12 Methodological issues may explain these discrepancies. This study has demonstrated marked differences in the extent of aerosolization of bacteria during three different hand-drying methods, using a detection method that has been used in the clinical setting to examine the transmissibility of Clostridium difficile.13 In the present experiments, we used a presumed non-pathogenic bacterium that was isolated from a proprietary probiotic yoghurt. We have also shown clear differences between the methods in the degree of visible splashing during hand drying, using paint spots, resulting in contamination of both the user and bathroom environment. Taylor et al. found no difference in the numbers of microorganisms remaining on hands following washing and drying with paper towels or with a warm low-velocity air dryer.6 They also sampled the air in the immediate vicinity of hands being
dried in an enclosed cabinet, and reported that both methods contribute equally to aerosolization of bacteria. Matthews and Newsom compared dryers and towels using whole-hand impressions on agar plates and found no difference in the numbers of bacteria remaining on hands.8 No differences were observed in airborne bacterial counts sampled during drying with warm air dryers compared with hand towels. Meers and Leong collected air using a Casella slit air sampler before, during and after drying hands using a warm air hand dryer, and more bacteria were detected in the air during and after compared with before drying.11 Dispersal of marker bacteria during use of a warm air dryer was up to three feet away and included the investigator’s laboratory coat; no such spread was seen during paper towel drying.11 Gendron et al. reported no evidence of airborne transmission during paper towel dispensing.14 There are few studies comparing airborne contamination associated with jet air dryers versus other methods of drying. Snelling et al. compared warm air dryers to a jet air dryer and found that the latter was superior for reducing bacterial transfer from hands.15 Given the use of a standard method of contaminating hands prior to drying, these data can be interpreted to show that greater removal (aerosolization) of bacteria occurred during use of the jet air dryer. In a recent study of 100 volunteers using either paper towels or a jet air dryer, more droplets and greater dispersal was seen during jet air drying.16 Scoping experiments were carried out to determine the best way of reproducibly measuring the extent of airborne dispersal of micro-organisms during hand drying (data not shown). We found that airborne bacterial counts of skin-type micro-organisms during hand drying were difficult to distinguish from those measured when no hand drying occurs. This reflects the background counts of such bacteria (present on skin scales) that are liberated during normal human activity. We concluded that using an indicator bacterial strain (chosen as it can be identified amongst background contaminants that are found normally in air and because it is considered to be harmless) to seed gloved hands provided the most reproducible method to measure dispersed bacteria during hand drying. This approach allowed the same subject to carry out multiple repeats without the risk of skin damage (due to repeated washing/drying), which would risk confounding the results obtained, i.e. produce results that are erroneous and which potentially fail to
Table III Mean paint spot counts (n ¼ 10) on different body areas for each different drying method for the user of the dryer and bystandera Area of body (size of area used for counting spots) Head (14 9 cm) Arm (LHS) (24 13 cm) Arm (RHS) (24 13 cm) Upper body (36 26 cm) Lower body (24 13 cm) Right leg e upper (24 13 cm) Right leg e lower (24 13 cm) Left leg e upper (24 13 cm) Left leg e lower (24 13 cm)
Jet air dryer
Warm air dryer
Paper towels
User
Bystander
User
Bystander
User
Bystander
1.2 14.8 3.4 144.1 23.6 15 6.6 15.8 6.3
0 0.1 0.1 0 0 0.8 0.3 0 0.3
0.5 27.4 8.7 65.8 20.6 1.6 4.3 3.4 2.8
0.2 0 0 0.2 0 0.2 0.5 0.4 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
LHS, left-hand side; RHS, right-hand side. a See Figure 2 for body areas. Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8 detect real differences in microbe dispersal during hand drying. The capacity to carry out multiple repeats adds considerable rigour to the results. The use of an indicator bacterial strain, seeded in numbers that reflect microbe counts found on poorly washed hands (and also recommended for this type of evaluation in British Standard EN1500 test procedure), was used to measure the dispersal of microbes that are not normal skin inhabitants, but which may contaminate hands during toileting.3,17,18 In the present study, counts of airborne lactobacilli were four- and 26-fold higher in the immediate vicinity of the jet air dryer than those associated with the warm air dryer and paper towels, respectively. Bacteria persisted in the air well beyond the 15 s hand-drying time, with approximately half of the lactobacilli collected more than 5 min after drying ceased. Aerosolized bacteria were dispersed 1 m away from the jet air dryer within the first 5 min of air sampling. As air is emitted from the jet air dryer at speeds of up to 400 mph, it is not surprising that bacteria were recoverable in similar quantities in the immediate vicinity of, and 1 m away from, the machine. Bacterial counts on settle plates at 1 and 2 m away, as well as visual inspection of paint dispersed in the environment and on to the user, demonstrate widespread droplet release during drying with a jet air dryer. Droplets contaminated all areas of the body, including the head, albeit in smaller numbers. There is, therefore, the potential for splattering on to persons other than the person drying their hands, and possibly inhalation of microbes sheared off hands during drying. The extent of contamination during use of the warm air dryer was not as widespread as with the jet air dryer, likely reflecting the lower speed of the air being emitted and its downwards, as opposed to upwards, trajectory. However, the prolonged drying time required for drying hands with a warm air dryer may add to the dissemination of airborne bacteria. Bacteria recovered from the air were four-fold less than with the jet air dryer, but counts in air collected at close proximity and 1 m away were similar, suggesting that distribution of airborne bacteria can also be widespread with this drying method. The mean bacterial counts on the settle plates were highest after use of the warm air dryer, which probably reflects the downwards trajectory of air and droplets, as demonstrated by paint splatter patterns (Figure 4). We confirmed this finding by the distribution of paint spots on the suit of the user; whereas there were generally fewer spots than seen on the suits worn by the user of the jet air dryer, these were detectable all over the body. These findings were similar to those of Ngeow et al.19 Our data show that hand drying with paper towels contributed the least with respect to airborne contamination up to 1 m away, as reported elsewhere.14 Although a few bacteria were detectable in the air following paper towel drying, the low numbers were similar close and 1 m away from the point of hand drying. Likewise, very few bacteria were recovered on the settle plates. We detected negligible dispersal of paint droplets (up to 20 cm) during paper towel hand drying (Figures 6 and 8), and those detected mostly seem to relate to towel removal from the dispenser. We used four paper towels to dry hands in each test, and so the repetitive towel removal likely contributed towards the paint spots visualized on the dispenser and adjoining wall. However, paper towel use was not associated with paint dispersal to the body of the user or bystander. Our results suggest that paper towel use contributes less than either
7
of the air dryer machines to microbial contamination of the bathroom environment and individuals within. Some studies have reported an increase in the numbers of bacteria during drying with paper towels compared with drying with a warm air dryer.12,20,21 Soap can disrupt skin commensals, and vigorous hand rubbing while drying disrupts skin squames and brings bacteria from within pores to the surface, thereby increasing numbers of shed bacteria. Meers and Yeo confirmed this by air sampling, and reported that bacteria and skin squames were increased after washing and drying using paper towels.21 Yamamoto et al. evaluated the efficiency of warm air and paper towel drying for the removal of bacteria from washed hands.22 They concluded that holding hands stationary and not rubbing was desirable for removing bacteria and that paper towels were useful for removing bacteria from the fingertips but not from the hands and fingers. Snelling et al. also reported that rubbing hands during drying with a warm air dryer can increase the numbers of bacteria on the skin.15 There are some weaknesses and limitations to the present study. Experiments involved gloved hands, and therefore we potentially underestimated the effect of released skin squames and ensuing bacterial dissemination. We used relatively high inocula to simulate poorly washed hands; clearly lower levels of microbe aerosolization would be expected for only minimally contaminated hands. Unfortunately, however, there is a perception that inadequate handwashing and drying after toileting occurs rather frequently.23e26 Drying was carried out for a set length of time for each method, whereas in previous studies this has continued until hands felt dry (not practicable with gloved hands). Shorter ‘drying’ times could be expected to result in less microbe aerosolization, although at a possible cost of damp hands, itself a potential transmission risk. The subject carrying out the hand drying was of average height; results may have been altered for a smaller adult or children, as potential contamination could be greater on the upper body or head. We only examined two types of air dryers and many others are available; this may limit extrapolation of our results. We note that some dryers claim to filter air, but this applies only to incoming air and not to that which is ejected from the machine and can become contaminated with microbes from hands. Whereas this study has demonstrated that the use of paper towels may be superior in terms of microbe aerosolization risk, some have reported the possibility of transmission associated with paper towels or dispensers. Harrison et al. reported that transfer of bacteria between paper-towel dispensers and hands can take place if either one is contaminated, although this is dependent on the design, construction, and positioning of these devices.27,28 Gendron et al. detected a large community of culturable bacteria including toxin producers, on unused paper towels, which may be transferred to users.14 Others have demonstrated, via air sampling, that sacks of used paper towels were a possible source of contamination in operating theatres.29 Paper towels have drawbacks, which include the need to manage the accumulation of waste in bathrooms. Dispensers and air dryers need to be cleaned and/or designed in such a way that surface contamination and ensuing risk of transfer of microbes to hands is minimized. Short drying times associated with jet air dryers may encourage greater compliance with hand drying. However, our results demonstrate that jet air and warm air dryers increase the numbers of airborne bacteria
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
8
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
within the bathroom environment. Increased dispersal of microbes during hand drying could be an important consideration in the healthcare setting for bacteria, such as Staphylococcus aureus and Escherichia coli, and for viruses including norovirus and influenza viruses.30 There is a concern that droplets/particles released during hand drying could transmit respiratory viruses such as influenza from contaminated hands or air. Both possibilities would appear to be more likely with jet air dryers, particularly if these shower droplets towards the face of the user, although the magnitude of risk remains unclear. Jet air and warm air dryers result in increased bacterial aerosolization when drying hands. These results suggest that air dryers may be unsuitable for use in healthcare settings, as they may facilitate microbial cross-contamination via airborne dissemination to the environment or bathroom visitors. Conflict of interest statement M.H.W. has received honoraria from ETS for microbiological advice and travel expenses to attend an ETS meeting. Funding sources The project was funded by the European Tissue Symposium (ETS).
References 1. Larson EL. Persistent carriage of gram-negative bacteria on hands. Am J Infect Control 1981;9:112e119. 2. Lowbury EJ, Thom BT, Lilly HA, Babb JR, Whittall K. Sources of infection with Pseudomonas aeruginosa in patients with tracheostomy. J Med Microbiol 1970;3:39e56. 3. Patrick DR, Findon G, Miller TE. Residual moisture determines the level of touch-contact-associated bacterial transfer following hand washing. Epidemiol Infect 1997;119:319e325. 4. Todd EC, Michaels BS, Smith D, Greig JD, Bartleson CA. Outbreaks where food workers have been implicated in the spread of foodborne disease. Part 9. Washing and drying of hands to reduce microbial contamination. J Food Prot 2010;73:1937e1955. 5. Department of Health. HBN 00-09 e Infection control in the built environment. Gateway Reference 18521. 2013. 6. Taylor JH, Brown KL, Toivenen J, Holah JT. A microbiological evaluation of warm air hand driers with respect to hand hygiene and the washroom environment. J Appl Microbiol 2000;89:910e919. 7. Ansari SA, Springthorpe VS, Sattar SA, Tostowaryk W, Wells GA. Comparison of cloth, paper, and warm air drying in eliminating viruses and bacteria from washed hands. Am J Infect Control 1991;19:243e249. 8. Matthews JA, Newsom SW. Hot air electric hand driers compared with paper towels for potential spread of airborne bacteria. J Hosp Infect 1987;9:85e88. 9. Blackmore MA, Prisk EM. Is hot air hygienic? Home Econ 1984;4:14e15. 10. Blackmore MA. A comparison of hand drying methods. Cater Health 1989;1:189e198. 11. Meers PD, Leong KY. Hot-air hand driers. J Hosp Infect 1989;14:169e171.
12. Gustafson DR, Vetter EA, Larson DR, et al. Effects of 4 hand-drying methods for removing bacteria from washed hands: a randomized trial. Mayo Clin Proc 2000;75:705e708. 13. Best EL, Fawley WN, Parnell P, Wilcox MH. The potential for airborne dispersal of Clostridium difficile from symptomatic patients. Clin Infect Dis 2010;50:1450e1457. 14. Gendron LM, Trudel L, Moineau S, Duchaine C. Evaluation of bacterial contaminants found on unused paper towels and possible postcontamination after handwashing: a pilot study. Am J Infect Control 2012;40:e5ee9. 15. Snelling AM, Saville T, Stevens D, Beggs CB. Comparative evaluation of the hygienic efficacy of an ultra-rapid hand dryer vs conventional warm air hand dryers. J Appl Microbiol 2011;110:19e26. 16. Margas E, Maguire E, Berland CR, Welander F, Holah JT. Assessment of the environmental microbiological cross contamination following hand drying with paper hand towels or an air blade dryer. J Appl Microbiol 2013;115:572e582. 17. Fuls JL, Rodgers ND, Fischler GE, et al. Alternative hand contamination technique to compare the activities of antimicrobial and nonantimicrobial soaps under different test conditions. Appl Environ Microbiol 2008;74:3739e3744. 18. BS EN 1500:2013 Chemical disinfectants and antiseptics. Hygienic handrub. Test method and requirements (phase 2/step 2). 19. Ngeow YF, Ong HW, Tan P. Dispersal of bacteria by an electric air hand dryer. Malays J Pathol 1989;11:53e56. 20. Huang C, Ma W, Stack S. The hygienic efficacy of different handdrying methods: a review of the evidence. Mayo Clin Proc 2012;87:791e798. 21. Meers PD, Yeo GA. Shedding of bacteria and skin squames after handwashing. J Hyg (Lond) 1978;81:99e105. 22. Yamamoto Y, Ugai K, Takahashi Y. Efficiency of hand drying for removing bacteria from washed hands: comparison of paper towel drying with warm air drying. Infect Control Hosp Epidemiol 2005;26:316e320. 23. Hennessy P, Strong P, Morgan-Jones R. Environmental contamination following toilet use in the operating department. J Hosp Infect 2007;66:90e92. 24. Boyce JM, Pittet D. Guideline for hand hygiene in health-care settings. Recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. MMWR Recomm Rep 2002;51:1e45. 25. Anderson JL, Warren CA, Perez E, et al. Gender and ethnic differences in hand hygiene practices among college students. Am J Infect Control 2008;36:361e368. 26. Garbutt C, Simmons G, Patrick D, Miller T. The public hand hygiene practices of New Zealanders: a national survey. N Z Med J 2007;120:U2810. 27. Harrison WA, Griffith CJ, Michaels B, Ayers T. Technique to determine contamination exposure routes and the economic efficiency of folded paper-towel dispensing. Am J Infect Control 2003;31:104e108. 28. Harrison WA, Griffith CJ, Ayers T, Michaels B. Bacterial transfer and cross-contamination potential associated with paper-towel dispensing. Am J Infect Control 2003;31:387e391. 29. Speers Jr R, Shooter RA. Shedding of bacteria to the air from contaminated towels in paper sacks. Possible significance for operating-rooms. Lancet 1967;2:301e302. 30. Bean B, Moore BM, Sterner B, Peterson LR, Gerding DN, Balfour Jr HH. Survival of influenza viruses on environmental surfaces. J Infect Dis 1982;146:47e51.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander E.L. Best a, P. Parnell a, M.H. Wilcox a, b, * a b
Microbiology Department, Old Medical School, Leeds General Infirmary, Leeds Teaching Hospitals NHS Trust, Leeds, UK University of Leeds, Leeds, UK
A R T I C L E
I N F O
Article history: Received 27 November 2013 Accepted 7 August 2014 Available online xxx Keywords: Air driers Bacteria Droplet dispersal Environmental contamination Hand drying Hand hygiene
S U M M A R Y
Background: The efficiency of hand drying is important in preventing pathogen spread, but knowledge surrounding which drying methods contribute least towards contamination of the environment and users is limited. Aim: To compare the propensity of three common hand-drying methods (jet air, warm air dryers, and paper towels) to contaminate the environment, users, and bystanders. Methods: Hands were coated in lactobacilli to simulate poorly washed, contaminated hands, and dried. The investigation comprised 120 air-sampling tests (60 tests and 60 controls), divided into close and 1 m proximity from the drying process. Separate tests used hands coated in paint to visualize droplet dispersal. Findings: Air bacterial counts in close proximity to hand drying were 4.5-fold higher for the jet air dryer (70.7 cfu) compared with the warm air dryer (15.7 cfu) (P ¼ 0.001), and 27-fold higher compared with use of paper towels (2.6 cfu) (P < 0.001). Airborne counts were also significantly different during use of towel drying versus warm air dryer (P ¼ 0.001). A similar pattern was seen for bacterial counts at 1 m away. Visualization experiments demonstrated that the jet air dryer caused the most droplet dispersal. Conclusion: Jet air and warm air dryers result in increased bacterial aerosolization when drying hands. These results suggest that air dryers may be unsuitable for use in healthcare settings, as they may facilitate microbial cross-contamination via airborne dissemination to the environment or bathroom visitors. ª 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction Hand hygiene is a fundamental component for controlling the spread of infection.1,2 Promotion of improved hand hygiene * Corresponding author. Address: Microbiology Department, Old Medical School, Leeds General Infirmary, Leeds Teaching Hospitals NHS Trust, Leeds LS1 3EX, UK. Tel.: þ44 113 392 6818; fax: þ44 113 392 2696. E-mail address: [email protected] (M.H. Wilcox).
is recognized as an important public health measure. There is much emphasis on the correct method for handwashing, but less so concerning the options for drying hands. Evidence suggests that efficiency of hand drying is important in the prevention of the transfer of micro-organisms from person to person or to the environment.3 However, the risk of aerosolizing micro-organisms during hand drying by various methods remains unclear. Methods for hand drying vary widely and include paper or cloth towels, warm air dryers or jet air dryers either singly or in
http://dx.doi.org/10.1016/j.jhin.2014.08.002 0195-6701/ª 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
2
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8 airborne bacteria. Experiments were carried out over a period of six weeks.
combination. Drying with towels may remove remaining microorganisms through friction, while moisture is wicked away into the absorbent material. Warm air dryers evaporate moisture and remove some micro-organisms during hand rubbing, although this process may take too long for efficient use, with hands consequently remaining damp. Newer jet air dryers rely on the passage of high speed, usually unheated, air to dry hands without rubbing, typically in 15 s.4 The selection of drying method may depend upon a number of factors including practicality, space availability, cost, or personal preference. Infection prevention considerations may influence the choice of hand-drying method, but the evidence base is weak to make informed decisions. Notably, the recent National Health Service (NHS) building guidance states that ‘Hot-air hand dryers reduce paper waste and may be considered for use in public areas of healthcare facilities, but should not be installed in clinical areas as they are noisy and could disturb patients’.5 It is clearly desirable to ensure that the process of hand drying does not increase the potential for micro-organism transmission, either directly to another person or indirectly by contamination of the bathroom environment. Evidence concerning whether hand-drying methods differ in their propensity to aerosolize, and so transmit microorganisms, is contradictory.6e11 Some studies suggest that drying hands with warmed air is associated with increased aerosolization of micro-organisms.11 However, others have suggested that there is little difference in aerosolization for the different drying methods.6 One of the reasons for discrepant results may be the use of relatively insensitive methods or experimental designs that fail to detect real differences. In this study, we aimed to compare the propensity of three widely used methods of hand drying (jet air dryer, warm air dryer, and paper towel drying) to contaminate the environment with bacteria via aerosolization during the drying process. Additionally, we investigated the extent of possible contamination of people drying their hands, as well as the possibility of contamination of a bystander.
Two electric hand dryers, a jet air dryer (Dyson Airblade) and warm air dryer (Pro-Elec GSQ250B), and a paper towel dispenser (Tork H3 classic dispenser containing Tork Advanced Towels, MRT213) were mounted on one wall at manufacturerrecommended heights for use. Air was collected using two Airtrace Environmental portable samplers (Biotrace, Microbial Contamination Control, Runcorn, UK) via a 1 m long Tygon tube placed at the left- and right-hand sides (LHS, RHS) of each drying unit, in close proximity and also at 1 m away. As air entered the sampler (28.3 L/min), it was forced through a fine slit (44 0.152 mm) at a velocity of 70 m/s thereby causing particulate matter (minimum size 0.4 mm) to impact upon a 140 mm diameter, lactobacillus-selective, agar plate (LAMVAB agar, Bioconnections, Wetherby, UK, supplemented with vancomycin). The plate rotated constantly from a known start point throughout the sampling period; thus, following culture, the location of the colonies represents the time of recovery from air. Following each test, the two plates (from the LHS and RHS) were immediately transported to the laboratory and incubated in an anaerobic cabinet at 37 C for 48e72 h, and colonies were then counted. Data from each drying method were compared using ManneWitney U-test (SPSS, IBM, Armonk, NY, USA) using a 5% confidence level to determine significance. In addition, four settle plates containing LAMVAB agar were sited at various locations around each dryer (Figure 1). At the end of each sampling session, the air-sampling machine was cleaned externally and internally using a disinfectant (Trigene, Medichem, Queenborough, UK) and run on a purge cycle to decontaminate the machine and tubing, as recommended by the manufacturer. Floor surfaces, the wall area around the dryers, and the dryer units were also thoroughly cleaned between tests using Trigene.
Methods
Testing procedure
Organization of air sampling
Air-sampling tubes were clamped at a height of 1.2 m (from floor level) on the LHS and RHS of the area where the drying of lactobacilli-contaminated gloved hands was taking place, in the outer edges of the airstream produced by the jet and warm air dryers. To ensure reproducibility, this was the same relative position for each of the hand-drying methods. For the tests 1 m away from the hand drying, the ends of the air-sampling tubes were clamped at the same height as before. The air samplers were switched on and hands were then immediately dried. For the warm air dryer, hands were rubbed together for 30e40 s until dry; for the jet air dryer, hands were placed into the unit and slowly drawn up until dry (15 s); for towel drying, four paper towels were taken from the dispenser and these were rubbed over hands until dry (15 s). The air samplers were left running for 15 min following each hand-drying process.
All tests were carried out in a room measuring 65 m3 with the door closed throughout experiments. Room air was maintained by standard ventilation without air-conditioning or negative or positive pressure ventilation. For each test, gloved hands were first coated by immersion in a suspension of lactobacilli (107 cfu/mL) that were cultured from a proprietary yoghurt (Actimel, Danone, Paris, France) and then dried in a standardized manner using one of each of the three drying methods: a warm air dryer, a jet air dryer, or paper towels. In total, this part of the study comprised 120 air-sampling tests (60 tests and 60 controls) with a total testing time of 15 min per test (comprising up to 40 s of drying time dependent on the method used and the remainder being the air-sampling time). The 60 air-sampling tests of contaminated hands comprised 20 collections for each drying method, of which 10 were in close proximity to the hands being dried and 10 were 1 m away. Control air sampling was carried out before every hand-drying test (with no hand drying taking place) both to provide baseline measurements and to minimize the risk of carryover of
Air-sampling process
Visual determination of the extent of contamination of the environment, user, and bystander To visualize the extent of potential contamination occurring during each drying process within the environment, and on to a
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
Left-hand side
Hand drier/paper towel dispenser mounted on wall
3
Right-hand side
Immediate vicinity (10 cm from each side of hand drier)
1 m away from hand-drying area
2 m away from hand-drying area
Figure 1. Locations of air-sampling tubes and settle plates. Diagram showing position of hand dryer and positions of the ends of the airsampling tubes for each of the three tests (black circles indicate the position of the end of the air sampler tubes) (mounted at a height of 1.2 m), white circles indicate position of settle plates on the floor.
user and a bystander, experiments were repeated as above. However, for each test, gloved hands were coated in a solution of black water-based paint (instead of lactobacilli), with the user wearing a white disposable Tyvek suit (DuPont, Stevenage, UK); suits were worn backwards so that the hood covered the face. Additionally, a bystander (wearing a suit in the same manner) stood diagonally adjacent to the dryer user (1 m distance) in order to replicate the scenario of someone waiting to dry his/her hands. In total, this user and bystander study comprised 30 drying tests (10 tests for each drying method). Following drying, the potential contamination within the environment was visualized and enumerated by the distribution of black paint splatters around each drying unit, and the distance travelled by the paint spots away from the drying unit/area was measured. The extent of potential contamination on the user and bystander was quantified by counting the numbers of paint spots on a number of predetermined areas marked out on the Tyvek suits (Figure 2).
Results Airborne lactobacilli counts during hand drying For each drying method, the airborne bacterial counts were very similar for samples collected on the LHS versus RHS of the subject/dryer. Table I shows the combined LHS and RHS mean counts (n ¼ 20) for each drying method. The air bacterial counts in tests performed in close proximity to hand drying
Figure 2. White disposable Tyvek suit showing predetermined areas used for counting paint spots.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
4
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
Table I Mean lactobacilli counts (cfu) in the air after 15 min sampling for each drying method (n ¼ 10) Distance
Close proximity 1 m away
Paper towels
Warm air dryer
Jet air dryer
LHS (n ¼ 10)
RHS (n ¼ 10)
Combined (n ¼ 20)
LHS (n ¼ 10)
RHS (n ¼ 10)
Combined (n ¼ 20)
LHS (n ¼ 10)
RHS (n ¼ 10)
Combined (n ¼ 20)
2.6 (2.2) 2.4 (2.7)
2.6 (1.7) 2 (2.1)
2.6 (1.9) 2.2 (2.4)
14.9 (14.5) 18.2 (4.7)
16.5 (16) 19.1 (7.9)
15.7 (14.9) 18.6 (6.4)
76.2 (40) 89.1 (34.2)
65.2 (37.5) 89.9 (37.6)
70.7 (38.2) 89.5 (34.9)
LHS, left-hand side; RHS, right-hand side. Values are mean (SD).
were 4.5-fold higher for the jet air dryer (70.7 cfu) compared with the warm air dryer (15.7 cfu) (P ¼ 0.001) and 27-fold higher compared with use of paper towels (2.6 cfu) (P < 0.001). Airborne counts were also significantly different during use of towel drying versus warm air dryer (P ¼ 0.001). A similar pattern was seen in mean bacterial counts recovered from air collected 1 m away from hand drying; bacterial air counts recovered following use of the jet air dryer, warm air dryer, and paper towels for hand drying were 89.5, 18.7, and 2.2 cfu, respectively. No lactobacilli were recovered in any control air-sampling experiments. The method enabled measurement of when bacteria were recovered from air during the 15 min sampling periods. Using the above test results, for the jet air dryer the mean proportions of bacteria recovered in the first, second, and third 5 min sectors (LHS and RHS, at all time-points, close to the dryer) were 52%, 28%, and 20%. The corresponding data for the warm air dryer were 50%, 24%, and 26%, and for the paper towels were 46%, 39% and 15%. Similar trends were seen for the bacteria recovered 1 m away from hand drying.
Settle plate lactobacilli counts Settle plates located under each drying unit yielded the greatest mean bacterial counts (Table II). Mean bacterial counts for the 1 m distant settle plates (LHS and RHS results) were combined, and results for the 10 tests at close proximity and 1 m away were combined (n ¼ 20) as the positions of settle plates remained the same. For the settle plates in close proximity to each drying method, counts under the dryer were the highest for the warm air dryer (190 cfu) compared with the jet air dryer (68.3 cfu) (P < 0.001) and the paper towel drying (11.9 cfu) (P < 0.001). Counts on settle plates 1 m away for the warm air dryer (7.8 cfu) were also higher than those for the jet air dryer (2 cfu) (P < 0.001) and the paper towels (0.7 cfu) (P < 0.001). There was no significant difference for the counts
for the one settle plate located 2 m away for the warm air dryer and jet air dryer (both 1.4 cfu) (P ¼ 0.940). There was a significant difference between the counts on settle plates for the paper towels (0.4 cfu) when compared to the warm air dryer (P ¼ 0.040) and jet air drier (P ¼ 0.008). All settle plates for the control experiments were negative.
Visualization tests to determine the extent of contamination of the environment Figures 3e5 show the distribution of paint following separate drying episodes with hands coated in paint. For the jet dryer, paint spots were found up to 120 cm away from the RHS of the unit, 140 cm away on the LHS, 30 cm above the unit, and 85 cm along the floor forwards towards the user (Figure 6). For the warm air dryer, the maximum distance travelled by paint spots was 30e40 cm on each side, 130 cm downwards along the wall and 40 cm forwards along the floor towards the user (Figure 7). For the hand towel drying process, markedly fewer paint spots were visualized; a few were seen on the towel dispenser and on the wall below, and some on the floor presumably due to dripping while rubbing hands (Figure 8). Table III shows mean numbers of paint spots detected during hand drying (n ¼ 10) on each of the demarcated areas of the body. No paint spots were seen on suits worn by the user or bystander during hand drying using towels. Numbers of spots on the user of the jet air dryer compared with the warm air dryer were overall higher for all body areas, except for both arm counts (mean counts: LHS arm 27.4 vs 14.8, P ¼ 0.061; RHS arm 8.7 vs 3.4, P ¼ 0.02). For both the jet air and warm air dryers, spots predominated in the upper body area, with the numbers
Table II Mean bacterial counts (mean cfu) on settle plates Drying process
Paper towels Warm air dryer Jet air dryer
Position of settle plates Under dryer (n ¼ 20)
1 m awaya (n ¼ 40)
2 m away (n ¼ 20)
11.9 190 68.3
0.7 7.8 2
0.4 1.4 1.4
Results for 10 tests at close proximity and 1 m away were combined as positions of settle plates remained the same. a Left- and right-hand-sided counts combined.
Figure 3. Visualization of paint spots following drying with the jet air dryer.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
70 cm
140 cm
5
30 cm 76 cm
Jet air drier
120 cm Wall Floor
100 cm
90 cm 85 cm
Figure 6. Furthest distance travelled by paint spots following one drying process with the jet air dryer. Black arrows indicate distance along the wall on which the unit was mounted; red arrows indicate distance forward along the floor.
Figure 4. Visualization of paint spots following drying with the warm air dryer.
of spots from the jet air dryer being significantly higher [mean counts of 144.1 and 65.8 (P ¼ 0.0005)]. Spot counts on upper legs were higher after use of the jet air compared with the warm air dryer [RHS mean counts 15 versus 1.6 (P ¼ 0.307) and LHS 15.8 versus 3.4 (P ¼ 0.006)]. Similar spot counts were seen on the lower legs (RHS mean counts 6.6 versus 4.3 (P ¼ 0.242) and LHS 6.3 versus 2.8 (P ¼ 0.794). Relatively few spots were detectable on the head area, with no significant difference between the mean counts for the jet air dryer and the warm air dryer [1.2 versus 0.5 (P ¼ 0.149)]. Overall, the mean number of
spots was significantly higher for the jet air dryer (mean counts 230.8) compared with the warm air dryer (135.1) (P ¼ 0.004). The number of paint spots detectable on the bystander was generally low for the air dryers (0.1e0.8). The body areas of the bystander where we recorded paint spots after using the jet air dryer were both arms (mean counts 0.1 for both), top of the right leg (mean count 0.8), and both lower legs (mean counts 0.3 for both). After using the warm air dryer, spots were most frequently seen on the head (mean counts 0.2), upper body (mean count 0.2), both upper legs (mean counts 0.2 and 0.4), and right lower leg (mean count 0.5). Overall, there was no significant difference when comparing the mean number of spots on the bystander with the jet air and warm air drying approach [mean counts 1.6 versus 1.5 (P ¼ 0.548)].
Discussion Hand drying is an integral part of the hand hygiene process, which aims to optimize the removal of potentially pathogenic
30 cm
Warm air drier
130 cm
40 cm
Wall Floor
40 cm
Figure 5. Visualization of paint spots following drying with the hand towels.
Figure 7. Furthest distance travelled by paint spots following one drying process with the warm air dryer. Black arrows indicate distance along the wall on which the unit was mounted; red arrow indicates distance forward along the floor.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
6
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
Paper towel dispenser 20 cm
Wall 20 cm
Floor
Figure 8. Furthest distance travelled by paint spots following one drying process with the paper towels. Black arrows indicate distance along the wall on which the unit was mounted; red arrow indicates distance forward along the floor.
micro-organisms that may be acquired during toileting and use of bathrooms. The published evidence concerning whether hand-drying methods may differ in their propensity to aerosolize and so transmit micro-organisms is contradictory.6e11 Some studies suggest that drying hands via warmed air is associated with increased aerosolization of micro-organisms, and others suggest there to be no difference.12 Methodological issues may explain these discrepancies. This study has demonstrated marked differences in the extent of aerosolization of bacteria during three different hand-drying methods, using a detection method that has been used in the clinical setting to examine the transmissibility of Clostridium difficile.13 In the present experiments, we used a presumed non-pathogenic bacterium that was isolated from a proprietary probiotic yoghurt. We have also shown clear differences between the methods in the degree of visible splashing during hand drying, using paint spots, resulting in contamination of both the user and bathroom environment. Taylor et al. found no difference in the numbers of microorganisms remaining on hands following washing and drying with paper towels or with a warm low-velocity air dryer.6 They also sampled the air in the immediate vicinity of hands being
dried in an enclosed cabinet, and reported that both methods contribute equally to aerosolization of bacteria. Matthews and Newsom compared dryers and towels using whole-hand impressions on agar plates and found no difference in the numbers of bacteria remaining on hands.8 No differences were observed in airborne bacterial counts sampled during drying with warm air dryers compared with hand towels. Meers and Leong collected air using a Casella slit air sampler before, during and after drying hands using a warm air hand dryer, and more bacteria were detected in the air during and after compared with before drying.11 Dispersal of marker bacteria during use of a warm air dryer was up to three feet away and included the investigator’s laboratory coat; no such spread was seen during paper towel drying.11 Gendron et al. reported no evidence of airborne transmission during paper towel dispensing.14 There are few studies comparing airborne contamination associated with jet air dryers versus other methods of drying. Snelling et al. compared warm air dryers to a jet air dryer and found that the latter was superior for reducing bacterial transfer from hands.15 Given the use of a standard method of contaminating hands prior to drying, these data can be interpreted to show that greater removal (aerosolization) of bacteria occurred during use of the jet air dryer. In a recent study of 100 volunteers using either paper towels or a jet air dryer, more droplets and greater dispersal was seen during jet air drying.16 Scoping experiments were carried out to determine the best way of reproducibly measuring the extent of airborne dispersal of micro-organisms during hand drying (data not shown). We found that airborne bacterial counts of skin-type micro-organisms during hand drying were difficult to distinguish from those measured when no hand drying occurs. This reflects the background counts of such bacteria (present on skin scales) that are liberated during normal human activity. We concluded that using an indicator bacterial strain (chosen as it can be identified amongst background contaminants that are found normally in air and because it is considered to be harmless) to seed gloved hands provided the most reproducible method to measure dispersed bacteria during hand drying. This approach allowed the same subject to carry out multiple repeats without the risk of skin damage (due to repeated washing/drying), which would risk confounding the results obtained, i.e. produce results that are erroneous and which potentially fail to
Table III Mean paint spot counts (n ¼ 10) on different body areas for each different drying method for the user of the dryer and bystandera Area of body (size of area used for counting spots) Head (14 9 cm) Arm (LHS) (24 13 cm) Arm (RHS) (24 13 cm) Upper body (36 26 cm) Lower body (24 13 cm) Right leg e upper (24 13 cm) Right leg e lower (24 13 cm) Left leg e upper (24 13 cm) Left leg e lower (24 13 cm)
Jet air dryer
Warm air dryer
Paper towels
User
Bystander
User
Bystander
User
Bystander
1.2 14.8 3.4 144.1 23.6 15 6.6 15.8 6.3
0 0.1 0.1 0 0 0.8 0.3 0 0.3
0.5 27.4 8.7 65.8 20.6 1.6 4.3 3.4 2.8
0.2 0 0 0.2 0 0.2 0.5 0.4 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0
LHS, left-hand side; RHS, right-hand side. a See Figure 2 for body areas. Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8 detect real differences in microbe dispersal during hand drying. The capacity to carry out multiple repeats adds considerable rigour to the results. The use of an indicator bacterial strain, seeded in numbers that reflect microbe counts found on poorly washed hands (and also recommended for this type of evaluation in British Standard EN1500 test procedure), was used to measure the dispersal of microbes that are not normal skin inhabitants, but which may contaminate hands during toileting.3,17,18 In the present study, counts of airborne lactobacilli were four- and 26-fold higher in the immediate vicinity of the jet air dryer than those associated with the warm air dryer and paper towels, respectively. Bacteria persisted in the air well beyond the 15 s hand-drying time, with approximately half of the lactobacilli collected more than 5 min after drying ceased. Aerosolized bacteria were dispersed 1 m away from the jet air dryer within the first 5 min of air sampling. As air is emitted from the jet air dryer at speeds of up to 400 mph, it is not surprising that bacteria were recoverable in similar quantities in the immediate vicinity of, and 1 m away from, the machine. Bacterial counts on settle plates at 1 and 2 m away, as well as visual inspection of paint dispersed in the environment and on to the user, demonstrate widespread droplet release during drying with a jet air dryer. Droplets contaminated all areas of the body, including the head, albeit in smaller numbers. There is, therefore, the potential for splattering on to persons other than the person drying their hands, and possibly inhalation of microbes sheared off hands during drying. The extent of contamination during use of the warm air dryer was not as widespread as with the jet air dryer, likely reflecting the lower speed of the air being emitted and its downwards, as opposed to upwards, trajectory. However, the prolonged drying time required for drying hands with a warm air dryer may add to the dissemination of airborne bacteria. Bacteria recovered from the air were four-fold less than with the jet air dryer, but counts in air collected at close proximity and 1 m away were similar, suggesting that distribution of airborne bacteria can also be widespread with this drying method. The mean bacterial counts on the settle plates were highest after use of the warm air dryer, which probably reflects the downwards trajectory of air and droplets, as demonstrated by paint splatter patterns (Figure 4). We confirmed this finding by the distribution of paint spots on the suit of the user; whereas there were generally fewer spots than seen on the suits worn by the user of the jet air dryer, these were detectable all over the body. These findings were similar to those of Ngeow et al.19 Our data show that hand drying with paper towels contributed the least with respect to airborne contamination up to 1 m away, as reported elsewhere.14 Although a few bacteria were detectable in the air following paper towel drying, the low numbers were similar close and 1 m away from the point of hand drying. Likewise, very few bacteria were recovered on the settle plates. We detected negligible dispersal of paint droplets (up to 20 cm) during paper towel hand drying (Figures 6 and 8), and those detected mostly seem to relate to towel removal from the dispenser. We used four paper towels to dry hands in each test, and so the repetitive towel removal likely contributed towards the paint spots visualized on the dispenser and adjoining wall. However, paper towel use was not associated with paint dispersal to the body of the user or bystander. Our results suggest that paper towel use contributes less than either
7
of the air dryer machines to microbial contamination of the bathroom environment and individuals within. Some studies have reported an increase in the numbers of bacteria during drying with paper towels compared with drying with a warm air dryer.12,20,21 Soap can disrupt skin commensals, and vigorous hand rubbing while drying disrupts skin squames and brings bacteria from within pores to the surface, thereby increasing numbers of shed bacteria. Meers and Yeo confirmed this by air sampling, and reported that bacteria and skin squames were increased after washing and drying using paper towels.21 Yamamoto et al. evaluated the efficiency of warm air and paper towel drying for the removal of bacteria from washed hands.22 They concluded that holding hands stationary and not rubbing was desirable for removing bacteria and that paper towels were useful for removing bacteria from the fingertips but not from the hands and fingers. Snelling et al. also reported that rubbing hands during drying with a warm air dryer can increase the numbers of bacteria on the skin.15 There are some weaknesses and limitations to the present study. Experiments involved gloved hands, and therefore we potentially underestimated the effect of released skin squames and ensuing bacterial dissemination. We used relatively high inocula to simulate poorly washed hands; clearly lower levels of microbe aerosolization would be expected for only minimally contaminated hands. Unfortunately, however, there is a perception that inadequate handwashing and drying after toileting occurs rather frequently.23e26 Drying was carried out for a set length of time for each method, whereas in previous studies this has continued until hands felt dry (not practicable with gloved hands). Shorter ‘drying’ times could be expected to result in less microbe aerosolization, although at a possible cost of damp hands, itself a potential transmission risk. The subject carrying out the hand drying was of average height; results may have been altered for a smaller adult or children, as potential contamination could be greater on the upper body or head. We only examined two types of air dryers and many others are available; this may limit extrapolation of our results. We note that some dryers claim to filter air, but this applies only to incoming air and not to that which is ejected from the machine and can become contaminated with microbes from hands. Whereas this study has demonstrated that the use of paper towels may be superior in terms of microbe aerosolization risk, some have reported the possibility of transmission associated with paper towels or dispensers. Harrison et al. reported that transfer of bacteria between paper-towel dispensers and hands can take place if either one is contaminated, although this is dependent on the design, construction, and positioning of these devices.27,28 Gendron et al. detected a large community of culturable bacteria including toxin producers, on unused paper towels, which may be transferred to users.14 Others have demonstrated, via air sampling, that sacks of used paper towels were a possible source of contamination in operating theatres.29 Paper towels have drawbacks, which include the need to manage the accumulation of waste in bathrooms. Dispensers and air dryers need to be cleaned and/or designed in such a way that surface contamination and ensuing risk of transfer of microbes to hands is minimized. Short drying times associated with jet air dryers may encourage greater compliance with hand drying. However, our results demonstrate that jet air and warm air dryers increase the numbers of airborne bacteria
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
8
E.L. Best et al. / Journal of Hospital Infection xxx (2014) 1e8
within the bathroom environment. Increased dispersal of microbes during hand drying could be an important consideration in the healthcare setting for bacteria, such as Staphylococcus aureus and Escherichia coli, and for viruses including norovirus and influenza viruses.30 There is a concern that droplets/particles released during hand drying could transmit respiratory viruses such as influenza from contaminated hands or air. Both possibilities would appear to be more likely with jet air dryers, particularly if these shower droplets towards the face of the user, although the magnitude of risk remains unclear. Jet air and warm air dryers result in increased bacterial aerosolization when drying hands. These results suggest that air dryers may be unsuitable for use in healthcare settings, as they may facilitate microbial cross-contamination via airborne dissemination to the environment or bathroom visitors. Conflict of interest statement M.H.W. has received honoraria from ETS for microbiological advice and travel expenses to attend an ETS meeting. Funding sources The project was funded by the European Tissue Symposium (ETS).
References 1. Larson EL. Persistent carriage of gram-negative bacteria on hands. Am J Infect Control 1981;9:112e119. 2. Lowbury EJ, Thom BT, Lilly HA, Babb JR, Whittall K. Sources of infection with Pseudomonas aeruginosa in patients with tracheostomy. J Med Microbiol 1970;3:39e56. 3. Patrick DR, Findon G, Miller TE. Residual moisture determines the level of touch-contact-associated bacterial transfer following hand washing. Epidemiol Infect 1997;119:319e325. 4. Todd EC, Michaels BS, Smith D, Greig JD, Bartleson CA. Outbreaks where food workers have been implicated in the spread of foodborne disease. Part 9. Washing and drying of hands to reduce microbial contamination. J Food Prot 2010;73:1937e1955. 5. Department of Health. HBN 00-09 e Infection control in the built environment. Gateway Reference 18521. 2013. 6. Taylor JH, Brown KL, Toivenen J, Holah JT. A microbiological evaluation of warm air hand driers with respect to hand hygiene and the washroom environment. J Appl Microbiol 2000;89:910e919. 7. Ansari SA, Springthorpe VS, Sattar SA, Tostowaryk W, Wells GA. Comparison of cloth, paper, and warm air drying in eliminating viruses and bacteria from washed hands. Am J Infect Control 1991;19:243e249. 8. Matthews JA, Newsom SW. Hot air electric hand driers compared with paper towels for potential spread of airborne bacteria. J Hosp Infect 1987;9:85e88. 9. Blackmore MA, Prisk EM. Is hot air hygienic? Home Econ 1984;4:14e15. 10. Blackmore MA. A comparison of hand drying methods. Cater Health 1989;1:189e198. 11. Meers PD, Leong KY. Hot-air hand driers. J Hosp Infect 1989;14:169e171.
12. Gustafson DR, Vetter EA, Larson DR, et al. Effects of 4 hand-drying methods for removing bacteria from washed hands: a randomized trial. Mayo Clin Proc 2000;75:705e708. 13. Best EL, Fawley WN, Parnell P, Wilcox MH. The potential for airborne dispersal of Clostridium difficile from symptomatic patients. Clin Infect Dis 2010;50:1450e1457. 14. Gendron LM, Trudel L, Moineau S, Duchaine C. Evaluation of bacterial contaminants found on unused paper towels and possible postcontamination after handwashing: a pilot study. Am J Infect Control 2012;40:e5ee9. 15. Snelling AM, Saville T, Stevens D, Beggs CB. Comparative evaluation of the hygienic efficacy of an ultra-rapid hand dryer vs conventional warm air hand dryers. J Appl Microbiol 2011;110:19e26. 16. Margas E, Maguire E, Berland CR, Welander F, Holah JT. Assessment of the environmental microbiological cross contamination following hand drying with paper hand towels or an air blade dryer. J Appl Microbiol 2013;115:572e582. 17. Fuls JL, Rodgers ND, Fischler GE, et al. Alternative hand contamination technique to compare the activities of antimicrobial and nonantimicrobial soaps under different test conditions. Appl Environ Microbiol 2008;74:3739e3744. 18. BS EN 1500:2013 Chemical disinfectants and antiseptics. Hygienic handrub. Test method and requirements (phase 2/step 2). 19. Ngeow YF, Ong HW, Tan P. Dispersal of bacteria by an electric air hand dryer. Malays J Pathol 1989;11:53e56. 20. Huang C, Ma W, Stack S. The hygienic efficacy of different handdrying methods: a review of the evidence. Mayo Clin Proc 2012;87:791e798. 21. Meers PD, Yeo GA. Shedding of bacteria and skin squames after handwashing. J Hyg (Lond) 1978;81:99e105. 22. Yamamoto Y, Ugai K, Takahashi Y. Efficiency of hand drying for removing bacteria from washed hands: comparison of paper towel drying with warm air drying. Infect Control Hosp Epidemiol 2005;26:316e320. 23. Hennessy P, Strong P, Morgan-Jones R. Environmental contamination following toilet use in the operating department. J Hosp Infect 2007;66:90e92. 24. Boyce JM, Pittet D. Guideline for hand hygiene in health-care settings. Recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. MMWR Recomm Rep 2002;51:1e45. 25. Anderson JL, Warren CA, Perez E, et al. Gender and ethnic differences in hand hygiene practices among college students. Am J Infect Control 2008;36:361e368. 26. Garbutt C, Simmons G, Patrick D, Miller T. The public hand hygiene practices of New Zealanders: a national survey. N Z Med J 2007;120:U2810. 27. Harrison WA, Griffith CJ, Michaels B, Ayers T. Technique to determine contamination exposure routes and the economic efficiency of folded paper-towel dispensing. Am J Infect Control 2003;31:104e108. 28. Harrison WA, Griffith CJ, Ayers T, Michaels B. Bacterial transfer and cross-contamination potential associated with paper-towel dispensing. Am J Infect Control 2003;31:387e391. 29. Speers Jr R, Shooter RA. Shedding of bacteria to the air from contaminated towels in paper sacks. Possible significance for operating-rooms. Lancet 1967;2:301e302. 30. Bean B, Moore BM, Sterner B, Peterson LR, Gerding DN, Balfour Jr HH. Survival of influenza viruses on environmental surfaces. J Infect Dis 1982;146:47e51.
Please cite this article in press as: Best EL, et al., Microbiological comparison of hand-drying methods: the potential for contamination of the environment, user, and bystander, Journal of Hospital Infection (2014), http://dx.doi.org/10.1016/j.jhin.2014.08.002
