Better Cleaning Technology
"Keeping school restrooms, cafeterias, and classrooms clean is an important step to protect the health of children. Having reviewed the data, I believe few technologies do this as effectively as a spray and vac system because - unlike old-fashioned mopping - the device removes soil rather than redistributing it, leaving surfaces clean and with dramatically lower microbial counts."
Michael Zuckerman, M.D.
By Dr. Charles P. Gerba, Ph.D.
Abstract
The presence of microorganisms on common classroom contact surfaces (fomites) was determined to identify the areas most likely to become contaminated. Six elementary classrooms were divided into control and intervention groups (cleaned daily with a quaternary ammonium wipe) and tested for heterotrophic bacteria. Three classrooms were also tested for norovirus and influenza A. Frequently used fomites were the most contaminated; water fountain toggles, pencil sharpeners, keyboards and faucet handles were the most bacterially contaminated; desktops, faucet handles and paper towel dispensers were the most contaminated with viruses. Influenza A virus was detected on up to 50% and norovirus on up to 22% of surfaces throughout the day. Children in the control classrooms were 2.32 times more likely to report absenteeism due to illness than children in the intervention classrooms and were absent longer (on average). Improved classroom hygiene may reduce the incidence of infection and thus student absenteeism.
Introduction
Bacteria and viruses may survive on environmental surfaces and may subsequently be transferred to a person's hands upon contact. Microbial survival on inanimate surfaces (fomites) depends upon a variety of factors including the species, the relative humidity or moisture content, the temperature, the surface materials and properties, and, in the case of viruses, whether they are enveloped or non-enveloped.
Non-enveloped viruses such as norovirus are fairly stable in the environment (Barker et al. 2001). For instance, the nonenveloped human astroviruses have been found to persist for up to 60 days on dry nonporous surfaces and for up to 90 days on dry porous surfaces (Abad et al. 2001).
Enveloped viruses are less stable; nevertheless, the enveloped influenza A virus may survive for up to 48 hours on dry surfaces (Bean et al. 1982) and the enveloped SARS coronavirus has been found to survive on fomites for up to 96 hours (Duan et al. 2003).
Bacteria tend to be found in higher numbers on porous surfaces and under moist conditions (Rusin et al. 2002). Transfer rates of microbes to hands are more efficient from hard, non-porous surfaces such as stainless steel (Rheinbaben et al. 2000; Rusin et al. 2002). A 40% transfer rate was observed for Escherichia coli from a non-porous laminate surface to fingers in one study (Scott and Bloomfield 1990). Rusin et al. (2002) found bacterial transfer rates of 38.5% to 41.8% from the telephone and rates of 27.6% to 40.0% from a sink faucet handle to a person's hand with minimal contact times. And in a study by Rheinbaben et al. (2000), the bacteriophage X174 was transferred from an inoculated doorknob to 14 successive people by contact and then to an additional six successive persons through the shaking of hands.
The transfer of bacteria and viruses from fomites to hands and from hands to fomites has been demonstrated in numerous studies (Ansari et al. 1988; Salvat et al. 1995; Boyce et al. 1997; Bures et al. 2000; Manning et al. 2001; Rusin et al. 2002).
Nevertheless, the contribution of such microbial transfer to the spread of diseases is unclear. In a study by Reynolds et al. (2005), 25 of 54 (46%) day care center surfaces tested positive for human biochemical markers (hemoglobin, urea, amylase) and 35 of 54 (65%) tested positive for protein levels greater than 200 g ml-1, suggesting contamination by blood, mucus, sweat, saliva or urine. Rotavirus has been found on 16% to 30% of surfaces in day care centers (Keswick et al. 1983; Wilde et al. 1992; Butz et al. 1993). In a study by Gwaltney and Hendley (1982), over 50% of individuals became ill after handling coffee cup handles and other objects that had been contaminated with rhinovirus, strongly suggesting transmission via fomites. In another study, the detection of influenza virus on greater than 50% of household and day care center fomites correlated with the seasonal incidence of disease in the community (Boone and Gerba 2005).
In the current study, disinfecting wipes (DW) containing quaternary ammonium were used daily to clean surfaces in a set of intervention classrooms. In addition, student illness and absenteeism records were kept to determine the impact of DW use. This study was undertaken to determine the levels of heterotrophic bacteria and the presence of norovirus and influenza A virus on common non-porous classroom surfaces and was not meant to be an exhaustive epidemiological study. Heterotrophic bacterial plate counts(HPC) were included as a general indicator of microbiological quality of an environmental surface to identify the areas in the classroom most likely to become contaminated by microorganisms. Norovirus and influenza A virus were included as indicators of the potential for spread of gastrointestinal and respiratory viruses via classroom fomites. Viruses such as norovirus that cause vomiting or diarrhea have the potential for contamination of the environment, especially as the result of poor personal hygiene. Evans et al. (1998) reported that 607 of 680 (89%) reported norovirus outbreaks were attributed to person-to-person transmission, including transmission due to poor hand hygiene and surface-to-surface transmission (Barker et al. 2001). Also, outbreaks of norovirus infections in passengers on cruise ships during successive trips have strongly implicated environmental contamination in the transmission of the virus (Barker et al. 2001). Although influenza A virus is not generally thought to be transmitted via fomites, other viruses such as the parainfluenza viruses, rhinoviruses and respiratory syncitial virus irritate the respiratory epithelium and induce coughing and sneezing, thereby facilitating their transmission by not only the airborne route, but also by environmental contamination (Hall et al. 1980; Barker et al. 2001). It is often difficult to test for the presence of specific respiratory viruses in schools unless there is an identified outbreak underway. Influenza A virus has a predictable seasonality (Monto 2004) during which it can be targeted for detection as an indicator of viral contamination. The presence of influenza A virus on fomites during the flu season highlights the potential for other respiratory viruses to contaminate classroom surfaces, possibly contributing to their transfer from person-to-person.
Materials and methods
Classroom intervention with disinfecting wipes
Between January 26 and March 15, 2004, six elementary school classrooms (with a total of 148 students) from one school in the Seattle, Washington (USA) area were included in an intervention study using quaternary ammonium disinfecting wipes (DW) to determine if the regular use of DW (Clorox Company, Oakland, CA, USA) reduced bacterial levels. Teachers in each classroom collected information regarding the incidence of gastrointestinal and respiratory illnesses (self-reported by the students) leading to student absenteeism during this period. Following the students' return to the classrooms, the teachers questioned them regarding their absence to determine if it was due to illness or for other reasons. Classroom A [5th grade (ages ten to eleven), 25 students], classroom B [1st grade (ages six to seven), 26 students], and classroom C (4th grade (ages nine to ten), 23 students] were included as control classrooms and did not receive any DW intervention. Classroom D [5th grade (ages ten to eleven), 26 students], classroom E [1st grade (ages six to seven), 24 students], and classroom F [3rd/4th grade (ages eight to ten), 24 students] were included as intervention classrooms in which adult volunteers wiped all test surfaces with DW in the morning each weekday prior to the arrival of the students. The use of such DW has been demonstrated to immediately reduce bacterial numbers on surfaces in previous experiments performed in our laboratory (unpublished data). The study was blind; students in the intervention classrooms were unaware of the DW being used in their classroom.
Surface sampling and enumeration of heterotrophic plate count bacteria
A total of 12 different types of surfaces were sampled in each classroom. Multiple desks, chair backs, computer keyboards, and computer mice were included. Samples were collected in the afternoons on January 26, February 2, February 23 and March 8 in each classroom by swabbing surfaces with BBLTM CultureSwabsTM with Amies medium without charcoal (Becton Dickinson, Franklin Lakes, NJ, USA). The samples were placed on ice for transport back to the laboratory. The samples were then serially diluted and the number of heterotrophic plate count (HPC) bacteria on each surface was determined by plating out appropriate dilutions from the swabs onto R2A medium (Difco, Sparks, MD, USA) in duplicate utilizing the spread plate technique. Agar plates were incubated at 30 degrees C for five days and then bacterial numbers were enumerated by counting colony-forming units (CFU). The number of HPC bacteria per square centimeter was then calculated for each surface.
Norovirus and influenza A virus detection on classroom surfaces collected on February 12, 2004 by swabbing surfaces with BBLTM CultureSwabsTM with Amies medium without charcoal (Becton Dickinson, Franklin Lakes, NJ, USA) in the morning, at midday recess, and in the afternoon in the three control classrooms A, B, and C. The surfaces sampled were a subset of those previously sampled for bacteria including student desktops, sink faucet handles, paper towel dispenser handles, soap dispenser levers, water fountain toggles, and entrance doorknobs. Fewer classroom surfaces were tested for viruses due to the additional costs of performing reverse transcriptase polymerase chain reaction (RT-PCR) on virus samples. The purpose of this sampling was to determine if norovirus and influenza A viral exposure was occurring in the classroom. Influenza strain A/Hong Kong/8/68 human Orthomyxovirus (ATCC VR-544) was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and used as an influenza A (H3N2) virus positive control throughout the procedure. A norovirus fecal isolate from a documented cruise ship outbreak was used as a positive control for norovirus. Viral RNA extraction and RT-PCR procedures were performed in separate rooms and in separate PCR hoods to prevent sample contamination. Also, samples and reagents were stored in separate rooms and separate freezers to further prevent contamination. QIAmp Viral RNA Mini kits (Qiagen Inc., Valencia, CA, USA) were used as recommended by the manufacturer to extract and concentrate viral RNA from fomite and control samples. All reagents used for the reverse transcriptase and the amplification steps were obtained from Applied Biosystems (Roche Molecular Systems Inc. Branchburg, NJ, USA).
Reverse transcriptase polymerase chain reaction (RT-PCR) for the detection of influenza A and norovirus
The reverse transcriptase reaction mixture was placed in an Applied Biosystems Gene Amp PCR System 9700 thermacycler (Roche Molecular Systems Inc. Branchburg, NJ, USA) and amplified using the following parameters: 10 minutes at 24ºC, 60 minutes at 44°C, 5 minutes at 99°C and 5 minutes at 5°C (Boone and Gerba 2005). This procedure was used for both viruses. The influenza A polymerase chain reaction (PCR) amplification step also employed the Applied Biosystems Gene Amp PCR System 9700 thermacycler. The primers used for this reaction were based on a nucleoprotein gene segment that is highly conserved among the various subtypes of type A influenza virus. The influenza A primers had the following nucleotide sequences: 5'-ATC-ACT-CAC-TGA-GTG-ACA-TC-3' (upstream), and 5'-CCT-CCA-GTT-TTC-TTA-GGA-TC-3' (downstream). These amplify a 306 base-pair segment of the virus genome (Wright et al. 1995; Vabret et al. 2000). This primer pair has a specificity of 100% and a sensitivity of 70% when compared with other methods of detection (Vabret et al. 2000). The following times and temperatures were utilized in the amplification step: 1) an initial period at 95°C for 10 minutes, 2) then 40 cycles with the following conditions: denaturation at 94°C for 30 seconds, annealing at 55°C for 30 seconds, then elongation at 72°C for 30 seconds, and 3) a final elongation step at 72°C for 10 minutes. The primers used for norovirus had the following sequences: JV12 5'-ATA-CCACTA-TGA-TGC-AGA-TTA-3' and JV13 5'-TCA-TCA-TCA-CCA-TAG-AAA-GAG-3' (Vinjé et al. 2003). These primers amplify a conserved region of open reading frame (ORF) 1 of the viral RNA polymerase including GLPSG and YGDD amino acid motifs and have an overall sensitivity of approximately 81% (Vinjé et al. 2003). The following times and temperatures were utilized in the norovirus amplification step: 1) an initial period at 94°C for 3 minutes, 2) then 40 cycles with the following conditions: denaturation at 94°C for 1 minute, annealing at 37°C for 1 minute and 30 seconds, then elongation at 74°C for 1 minute, and 3) a final elongation step at 74°C for 7 minutes. All PCR products were detected using 2% agarose gel eletrophoresis. An Alpha Imager 2000 (Alpha Innotech Company, San Leandro, CA, USA) was used to visualize the resulting product bands. Positive samples were purified using a QIAquick PCR Purification Kit (Qiagen Inc., Valencia, CA, USA) and sequenced at the University of Arizona's Genomic Analysis Technical Center using a 377 ABI sequencer from Applied Biosystems (Roche Molecular Systems Inc. Branchburg, NJ, USA). Sequencing is a commonly used method for confirmation of RT-PCR positive products to prevent falsepositive results (Vinjé et al. 2003).
Statistical analysis
One-way analysis of variance (ANOVA) tests were performed to determine if the bacterial counts from the control and intervention classrooms were statistically different (P 0.05). Information related to child absenteeism, in addition to concurrent gastrointestinal or respiratory illness (self-reported by students to teachers), was recorded during the entire 7-week duration of the study for students in both the control and intervention classrooms. The odds ratio of developing a respiratory or gastrointestinal illness leading to absenteeism was determined for the control classrooms relative to the intervention classrooms using Statcalc of EpiInfo version 6.0.
Results
HPC bacterial contamination of surfaces
The geometric means of the numbers of HPC bacteria found on each classroom surface were presented. The geometric mean was utilized due to the presence of outlying data values. Counts were grouped by the sample site (eg: desktop, keyboard) and by classroom type (control vs. intervention). The water fountain toggle and the manual pencil sharpener handle were determined to be the most contaminated surfaces (per square cm) and the classroom entrance and exit doorknobs were found to be the least contaminated surfaces. There were no statistically significant (P 0.05) differences between the bacterial numbers found on the surfaces of control and intervention classrooms, or between the three control classrooms or the three intervention classrooms (data not shown). There were also no significant differences between counts when classroom grade (e.g., 1st grade vs. 5th grade) was also considered (data not shown).
Viral occurrence
Influenza A virus was detected on surfaces in all three control classrooms. It was found on 24% (13 of 54) of all classroom surfaces tested, including 13.6% of surfaces in the mornings, 16.7% at midday recess, and 50% in the afternoon. The virus was most commonly detected on student desktops (5 of 27), sink faucet handles (4 of 7), paper towel dispensers (2 of 4), and entrance doorknobs (2 of 7), but was not detected on the water fountain toggle or the soap dispenser in any of the classrooms. Norovirus was detected on surfaces in only two of the three classrooms. The virus was found on 16.4% (9 of 55) of all classroom surfaces tested including 13.6% of surfaces in the morning, 22.2% at midday recess, and 13.3% of classroom surfaces in the afternoon. Norovirus was detected on student desktops (6 of 27), a paper towel dispenser (1 of 4), a sink faucet handle (1 of 7), and a water fountain toggle (1 of 7), but was not detected on the soap dispenser or the entrance doorknob in any of the classrooms. During the two weeks prior to the viral sampling date, from three to six children had been absent from each of these control classrooms due to illness. All three classrooms had at least one student that had been absent as a result of illness as recently as February 11th (the day before viral sampling).
Summary of illnesses
The numbers of students absent due to illness per week in the control and intervention classrooms were determined. During week three, a total of six students from the three control classrooms were absent due to illness; no students were absent from any of the intervention classrooms. This was the only week in which students were absent due to illness from only the control, but not the intervention classrooms. Otherwise, a total of at least two students were absent from the intervention classrooms and a total of at least 4 students were absent from the control classrooms in each week of the study, with the exception of week four in which no students were absent from any of the classrooms included in the study. Student illness summary data during the seven-week study period (from January 26 to March 15, 2004) were presented. The intervention classrooms were treated with DW during this entire period. In the control classrooms, 26 of 74 children (35.1%) became ill (with either gastrointestinal or respiratory symptoms) with a total of 57 days absent (mean = 2.23 days per ill student, median = 2.0 days per ill student); 14 of 74 children (18.9%) became ill in the intervention classrooms with a total of 22 days absent (mean = 1.57 days per ill student, median = 1.0 days per ill student). During this intervention study, children in the control classrooms were 2.32 times more likely to become ill than the children in the intervention classrooms (P = 0.026, 95% CI = 1.03 5.28, odds ratio determined using EpiInfo version 6 Statcalc).
Discussion
Based on existing literature, all of the classroom surfaces examined in this study are likely to have high transfer rates to the hands of children through normal contact and use during the day. All were made of non-porous materials such as plastics, veneers or laminates, and stainless steel. The water fountain toggle and the manual pencil sharpener handle are used by numerous students throughout the day and were found to be the two most bacterially contaminated classroom surfaces in this study. The sink faucet handle and the paper towel dispenser lever are likewise used by multiple students and were also among the most contaminated classroom surfaces. The computer keyboard was the third most contaminated fomite. Keyboards are frequently used by multiple students and experience an extended contact time with a person's hand during use. The student desktop is commonly used throughout the day and is touched quite frequently; however, most desks are utilized by only an individual student. The doorknob is not frequently touched during the average school day, as doors are often propped open. Frequently or heavily used fomites are most likely contaminated by such use and therefore carry higher heterotrophic bacterial loads than other lightly used fomites. Sink faucet handles (4 of 7 samples, 57%), paper towel dispenser levers (2 of 4 samples, 50%), and student desktops (5 of 27 samples, 18.5%) were the surfaces most often contaminated by influenza A virus. Norovirus was detected on these fomites as well. As mentioned previously, these are among the most frequently and heavily used fomites in the classroom. Interestingly, the water fountain toggle, the most heavily contaminated surface by bacteria, tested negative for influenza A virus in all samples (0 of 7) and only 1 of 7 (14.3%) tested positive for norovirus. Influenza A virus was detected on 2 of 7 (28.6%) of the entrance doorknobs tested though norovirus was not detected on any doorknobs (0 of 7). There is therefore no clear correlation between bacterial and viral contamination of classroom environmental surfaces. The manual pencil sharpener handle and the computer keyboard were not tested for the presence of viruses. Although the bacterial levels found on classroom surfaces in this study could not be used to model risk, there was an apparent effect from regular DW use. Classrooms undergoing daily intervention with DW were found to have a statistically significant reduction in student absenteeism due to illness. There are approximately 164 million lost school days each year in the USA among students in kindergarten (ages five to six years) through 12th grade (ages 17 to 18 years), with a yearly average of 4.5 absent days per student (Anonymous 1998). Numerous hand-washing and environmental hygiene studies have been conducted in daycare centers and hospital pediatric wards in which a decrease in the incidence of infection has been observed (Barker et al. 2001). In an elementary school, children involved in an experimental hand-washing program were absent 50.6% fewer times (277 versus 140) than children that did not participate (Guinan et al. 2002). Mandatory handwashing programs have also reduced the incidence of gastrointestinal illnesses and absenteeism in elementary school students (Kimel 1996; Master et al. 1997). Comparable results have been observed with similar hand-washing programs or the use of hand sanitizers in schools and day care centers (Butz et al. 1990; Hammond et al. 2000). Proper classroom hygiene such as the use of DW and frequent handwashing might therefore reduce the transfer of pathogenic bacteria and viruses between fomites and the hands of children and thereby reduce the spread of diseases. This preliminary work emphasizes the need for further epidemiological studies utilizing more classrooms and greater numbers of students to determine the potentially important role of microbial contamination of classroom fomites in the transfer and spread of infectious diseases and how this relates to student absenteeism rates.
References available upon request.
Popular Topics: MRSA | Staph | Norovirus | Flu | E. Coli | C. Difficile | Salmonella | Cleaning for Health | Nosocomial Infections | Disinfection | Bacteria | Viruses | Indoor Air Quality | Asthma | Allergies | Allergen | Mold
Only KaiShare partners are allowed to comment. Please sign-in to your KaiShare account to comment.
Sorry, there are currently no comments on this article...