Actual "Sneeze Effect" without backlighting

(Long/Full Version)

The potential spread of infection caused by aerosol 
contamination of surfaces after flushing a domestic toilet 

J. Barker1 and M.V. Jones2 
Department of Pharmaceutical and Biological Sciences, School of Life and Health Sciences, Aston University, Aston Triangle, 
Birmingham, UK, and 2Unilever Research Port Sunlight Laboratory, Bebington, Wirral, UK 

2004/0867: received 26 July 2004, revised and accepted 5 January 2005 

ABSTRACT 

J. BARKER AND M. V. JONES. 2005. 
Aims: To determine the level of aerosol formation and fallout within a toilet cubicle after flushing a toilet 
contaminated with indicator organisms at levels required to mimic pathogen shedding during infectious diarrhoea. 
Methods and Results: A semisolid agar carrier containing either Serratia marcesens or MS2 bacteriophage 
was used to contaminate the sidewalls and bowl water of a domestic toilet to mimic the effects of soiling after 
an episode of acute diarrhoea. Viable counts were used to compare the numbers of Serratia adhering to the porcelain 
surfaces and those present in the bowl water before and after flushing the toilet. Air sampling and settle plates 
were used to determine the presence of bacteria or virus-laden aerosols within the toilet cubicle. After seeding 
there was a high level of contamination on the porcelain surfaces both under the rim and on the sides of the 
bowl. After a single flush there was a reduction of 2Æ0–3Æ0 log cycles cm)2 for surface attached organisms. The 
number of micro-organisms in the bowl water was reduced by 2Æ0–3Æ0 log cycles ml)1 after the first flush and 
following a second flush, a further reduction of c. 2Æ0 log cycles ml)1 was achieved. Micro-organisms in the air 
were at the highest level immediately after the first flush (mean values, 1370 CFU m)3 for Serratia and 
2420 PFU m)3 for MS2 page). Sequential flushing resulted in further distribution of micro-organisms into the 
air although the numbers declined after each flush. Serratia adhering to the sidewalls, as well as free-floating 
organisms in the toilet water, were responsible for the formation of bacterial aerosols. 
Conclusions: Although a single flush reduced the level of micro-organisms in the toilet bowl water when 
contaminated at concentrations reflecting pathogen shedding, large numbers of micro-organisms persisted on the 
toilet bowl surface and in the bowl water which were disseminated into the air by further flushes. 
Significance and Impact of the Study: Many individuals may be unaware of the risk of air-borne dissemination 
of microbes when flushing the toilet and the consequent surface contamination that may spread infection within 
the household, via direct surface-to-hand-to mouth contact. Some enteric viruses could persist in the air after 
toilet flushing and infection may be acquired after inhalation and swallowing. 

Keywords: aerosols, environment, gastroenteritis, infection risk, MS2-bacteriophage, toilet. 

INTRODUCTION 

Infectious gastroenteritis is caused by a variety of microorganisms which have the potential to contaminate surfaces 

Correspondence to: J. Barker, Department of Pharmaceutical and Biological 
Sciences, School of Life and Health Sciences, Aston University, Aston Triangle, 
Birmingham B4 7ET, UK (e-mail: j.e.barker@aston.ac.uk). 

ª 2005 The Society for Applied Microbiology 

in toilets and bathrooms because they are excreted in large 
numbers during episodes of acute diarrhoea. Flushing the 
toilet is known to produce aerosols that are capable of 
causing surface contamination within the toilet and bathroom (Darlow and Bale 1959; Bound and Atkinson 1966; 
Newsom 1972; Gerba et al. 1975). Many enteric pathogens 
are spread by the faecal–oral route and it has been suggested 
that the fallout of droplets containing faecal material, after 


340 J. BARKER AND M.V. JONES 

flushing the toilet, is an important infection hazard within 
the bathroom (Hutchinson 1956; Darlow and Bale 1959; 
Gerba et al. 1975). 

Viruses are a significant cause of gastroenteritis worldwide 
and virtually all children aged 3–5 years acquire a rotavirus 
infection. Individuals with acute diarrhoea may shed >1010 
infectious rotavirus particles per ml of faeces (Hart and 
Cunliffe 1999) and toilet flushing could spread aerosols 
containing the virus onto surfaces in the bathroom. The 
virus spreads rapidly within families and adults also become 
infected, although they generally suffer from asymptomatic 
or mild illness. In the UK, over the last decade the reported 
incidence of norovirus has increased considerably and it is 
estimated that at least 3 million cases occur annually (Evans 
et al. 1998; Wheeler et al. 1999). The virus produces a rapid 
onset of diarrhoea and vomiting in both adults and children 
and large numbers of infectious virus particles are found in 
both vomit and faeces. The infective dose of both norovirus 
and rotavirus is presumed to be as low as 10–100 virus 
particles (LeBaron et al. 1990) which undoubtedly contributes to their high infectivity, spreading mainly through 
contact with infected individuals and virus-contaminated 
environmental fomites. Norovirus outbreaks can be difficult 
to control because the virus spreads rapidly in closed 
environments often resulting in secondary attack rates of 
>50% (Caul 1994). 

The risk of environmental contamination occurring in the 
bathroom is likely to be greatest during acute diarrhoeal 
illness when billions of micro-organisms are being flushed 
down the toilet. During such episodes faecal material is 
likely to contaminate not only the bowl water but also the 
porcelain surfaces within the toilet bowl. Flushing produces 
aerosols from the force of the water running down the 
surfaces of the bowl and from the turbulence caused by 
mixing with water contained in the bowl. Previous studies 
have shown that toilet design influences aerosol production. 
Bound and Atkinson (1966) found that a siphonic toilet 
produced much lower concentrations of contaminated 
particles than the older style
wash-down. pan by a ratio of 

1 : 14. Newsom (1972) reported that the splashing produced 
by flushing varied with cistern height and bowl design and 
noted that a double-trap siphonic toilet produced more 
splashes than a
wash-down. type. Obviously there is 
considerable variation in the design of modern flush toilets 
which is likely to affect the amount of turbulence, splashing, 
and aerosol production. 
This report considers the infection risk after flushing a 
toilet contaminated with indicator organisms at levels 
required to mimic pathogen shedding during infectious 
diarrhoea which could be >1010 particles per ml. A domestic 
close-coupled siphonic toilet, a type used widely in the UK, 
was used to examine the dynamics of aerosol formation and 
contamination of environmental surfaces after flushing. The 

separate effects of bacteria adhering to the porcelain 
sidewalls as opposed to bacteria present in the toilet bowl 
water on the formation of bacterial aerosols was determined. 
In addition, we investigated the effects of sequential flushing 
on environmental contamination. 

MATERIALS AND METHODS 
Toilet 

A domestic toilet, situated in a room of 2Æ6m)3, in the home 
of one of the authors (J.B.) was used throughout (see Fig. 1). 
The cistern had a reservoir containing 12 l of flush water 
and the toilet bowl contained 2 l of water. The surface area 
of the internal bowl sides above the water line was 1150 cm2. 
Before seeding with micro-organisms the toilet bowl water 
and porcelain surfaces were scrubbed with a chlorine-
containing disinfectant (50 000 ppm of free available chlorine) and flushed six times to eliminate traces of the cleaning 
compound. This procedure was also used to decontaminate 
the toilet after individual experiments. 

Organisms 

For bacterial contamination a pigment-producing strain of 
Serratia marcesens (NCTC 10211) was used throughout 

86 cm 

122 cm 

B 
C D 
EHeight of 
ceiling 248 cm 
Air sample 
at this 
position 
Fig. 1 The relative positions of settle plates which were exposed for 
30 min after flushing the toilet. A: a shelf behind the toilet, 83 cm 
above the seat; B: the cistern, 41 cm above the seat; C: toilet seat, left; 

D: toilet seat, right; E: 30 cm in front of the toilet, level with the toilet 
seat 
ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 339–347, doi:10.1111/j.1365-2672.2005.02610.x 


POTENTIAL SPREAD OF INFECTION FROM THE DOMESTIC TOILET 341 

because it can be easily identified on nonselective agar and it 
has a low decay constant when sprayed in aqueous 
suspension (Darlow and Bale 1959). The organism was 
grown to stationary phase in 100 ml buffered peptone water 
(Oxoid Ltd, Basingstoke, UK) on an orbital shaker at 37
C 
for 24 h to give c. 109 CFU ml)1. The suspension was 
centrifuged (2080 g for 30 min) before washing and suspending in 10 ml of 1/4 strength Ringer’s solution (RS; 
Oxoid Ltd). The washed suspension was added to 80 ml of 
semisolid agar (0Æ2% w/v, Technical Agar; Oxoid Ltd) and 
mixed thoroughly to produce a seed inoculum containing c. 
1010 bacteria. 

Virus contamination was achieved using MS2 bacteriophage (ATCC 15597-B1) which is a nonpathogenic virus 
that can be easily propagated in the laboratory. MS2 is a 
nonenveloped virus, known to be relatively stable in the 
environment, which has been used previously as an indicator 
for enteric viruses (Jones et al. 1991; Havelaar et al. 1993; 
Dore et al. 2000; Allwood et al. 2003). Bacteriophage 
propagation was performed using an agar-overlay technique 
using Escherichia coli (ATCC 15597) as the host (Adams 
1959). Briefly, a soft agar/host covering was prepared by 
overlaying agar plates (tryptone soya agar; Oxoid Ltd) with 
2Æ5 ml of melted 0Æ5% agar (same medium) which contained 
two drops of a 6-h culture of the host in tryptone soya broth 
(TSB; Oxoid Ltd). The soft agar was allowed to harden and 
the surface covered with c. 0Æ5 ml of the concentrated 
bacteriophage suspension. After 24 h incubation at 37
C, 
the soft agar was scraped off the surface of the plates and 
suspended in TSB. The extract was centrifuged at 3000 g 
for 20 min to sediment the cellular debris and agar. The 
supernatant containing the bacteriophage was passed 
through a 0Æ2-lm filter and the filtrate stored at 4
C. Prior 
to use the bacteriophage suspension was allowed to equilibrate to RT. To quantitate the virus, 10-fold dilutions of 
the stock suspension in TSB were assayed by the overlay 
method. Plaques were counted after 24 h incubation at 37
C 
and the results expressed as plaque-forming units 
(PFU ml)1). To seed the toilet with virus, 1Æ5 ml of stock 
bacteriophage suspension was added to 80 ml of semisolid 
agar (0Æ2% w/v) and mixed thoroughly to produce a seed 
inoculum containing c. 1010 PFU of virus. 

Toilet seeding 

Experiments were carried out to establish the dynamics of 
aerosol formation and surface contamination after seeding 
the toilet with S. marcesens. Key experiments were repeated 
using MS-2 bacteriophage to determine whether a similar 
pattern of contamination occurred when the toilet was 
seeded with a virus. Semisolid agar (0Æ2% w/v, 80 ml) was 
used as the carrier for the seed inoculum because it had the 
consistency of a loose stool. The inoculum was applied with 

a 50-ml syringe; either directly to sidewalls of the toilet bowl 
to give, as far as possible, an even coating on the porcelain 
surface above the water line, to simulate the splashing effects 
associated with acute diarrhoea or directly to the bowl water 
avoiding contamination of the sidewalls. The toilet was 
flushed 5 min after applying the inoculum. Preliminary tests 
showed that the toilet was not contaminated with pigment-
producing Serratia species or with MS2 bacteriophage prior 
to seeding. 

To study the aerosol formation produced by Serratia 
adhering to the sidewalls of the toilet, as opposed to the 
bacteria present in the bowl water, after applying the 
inoculum, the bowl water was disinfected with sodium 
hypochlorite at a final concentration of 5000 ppm of free 
available chlorine, before the toilet was flushed. After 
30 min disinfection the residual chlorine was neutralized 
for 15 min by adding 8 g of sodium thiosulfate to the bowl 
water (final concentration 0Æ4%). Preliminary experiments 
had shown that after this level of disinfection and neutralization Serratia was not detected in the bowl water nor did 
the water exhibit residual antibacterial activity. 

Microbiological sampling 

The contaminated toilet bowl surface was sampled using 
cotton swabs (25 cm2) moistened in RS which were rubbed 
over an area of 50 cm2. The swabs were placed in 6 ml of RS 
and homogenized for 30 s using a stomacher. To determine 
bacterial counts 10-fold dilutions were prepared in RS and 
0Æ1 ml aliquots spread onto nutrient agar plates (NA; Oxoid 
Ltd) which were incubated at 30
C for 18 h. Swabs for virus 
determination were also homogenized in RS and dilutions 
assayed by the agar overlay technique. The toilet bowl water 
was sampled by removing an aliquot with a disposable sterile 
plastic pipette into a 25-ml universal container. 

Bacterial air samples were collected onto NA immediately 
after flushing the toilet, using a portable, single-sieve, 
impacter MicroBio MB1 (FW Parrett Ltd, London, UK). 
This device meets the basic criteria for a suitable reference 
sampler although it does not differentiate particle sizes 
(Griffiths and Stewart 1999). The sampler was positioned 
30 cm in front of the toilet at a height of 20 cm above the 
toilet seat with the lid open. The door to the toilet cubicle 
was closed during sampling. Air sample volumes of between 
100 and 600 l were collected, depending on whether the 
samples were collected after the first, second or third flush 
after seeding. A control 500-l air sample was taken prior to 
flushing the toilet to establish that there were no Serratia 
species or MS2 bacteriophage particles present in the air. 
Virus-laden aerosols were detected using 0Æ2% semisolid 
agar for the entrapment medium. Bacteriophage was detected by thoroughly mixing the entrapment medium and 
assaying using the overlay technique as described above. 

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 339–347, doi:10.1111/j.1365-2672.2005.02610.x 


342 J. BARKER AND M.V. JONES 

MS2 bacteriophage PFU m)3 Serratia CFU m)3 
Bowl water 
Untreated disinfected 
Time Untreated bowl water bowl water and neutralized 
Before flush Not detected Not detected Not detected 
After flush 
1 min 2420 (691) 1370 (527) 351 (58) 
30 min 178 (91) 75 (25) 1 (0Æ25) 
60 min 27 (25) 13 (8Æ5) 2Æ6(0Æ5) 

Values given within parenthesis are standard error of the mean for three replicates. 

Table 1 MS2 bacteriophage and Serratia 
detected in air samples for up to 60 min after 
a single toilet flush. The effect of untreated 
bowl water and bowl water that was disinfected and neutralized prior to flushing on the 
dissemination of Serratia is shown 

Settle plates containing NA, exposed for 30 min after each 
flush, were used to determine the fallout of bacterial aerosols 
onto five surfaces surrounding the toilet (Fig. 1). Settle 
plates for virus capture contained 0Æ2% semisolid agar which 
was assayed as for the air samples. 

RESULTS 

The number of Serratia or MS2 bacteriophage disseminated 
into the air after a single flush of the toilet, 5 min after the 
inoculum had been applied to the sidewalls, for three 
replicate experiments is shown in Table 1. One minute after 
flushing, when the toilet bowl contained untreated water, the 
mean air count for Serratia was 1370 CFU m)3 which 
declined to 75 and 13 CFU m)3 after 30 and 60 min 
respectively. The toilet water contained c. 108 CFU ml)1 of 
Serratia prior to flushing and 60 min thereafter the numbers 
declined to c. 106 CFU ml)1 (data not shown). When the 
toilet bowl water was disinfected and neutralized prior to 
flushing the number of bacteria released into the air was 
greatly reduced. One minute after flushing the air count was 
351 CFU m)3 and this fell to <5 CFU m)3 after 30 and 
60 min. Compared with the number of bacteria released into 
the air, almost twice as many virus particles were detected. 
One minute after the first flush 2420 PFU m)3 of MS2 

bacteriophage were detected, declining to 178 and 
27 PFU m)3 after 30 and 60 min respectively. 

Table 2 reveals the level of contamination on surfaces 
surrounding the toilet 30 min after the toilet was flushed for 
three replicate experiments. The number of bacteria detected on the settle plates was greatest when the inoculum was 
applied to the sidewalls of the toilet. Counts were highest on 
the toilet seat (47 and 50 CFU per plate) which was more 
likely to have been contaminated by splashes but the shelf 
and the cistern which were 83 and 41 cm above the seat had 
mean counts of 38 and 45 CFU respectively. There was 
considerable variation in the counts obtained between three 
replicate experiments, presumably reflecting the variation in 
the distribution of the inoculum on the sidewalls and the 
flush hydrodynamics. The settle plate counts obtained after 
applying the inoculum directly to the water were less than 
half of those obtained when the inoculum was applied to the 
sidewalls. With one exception, the level of surface virus 
contamination was broadly similar to the bacterial contamination after the inoculum had been applied directly to the 
sidewalls. 

Figure 2 shows the number of MS2 bacteriophage 
particles or Serratia attached to the porcelain surfaces of 
the toilet bowl 5 min after applying the inoculum to the 
sidewalls and 60 min after flushing. Levels of contamination 

Settle plate counts 0–30 min after flush 
Table 2 MS2 bacteriophage and Serratia 
detected within 30 min of a single flush, on 
MS2 bacteriophage settle plates at various locations surrounding 
PFU per plate Serratia (CFU per plate) the toilet. For Serratia the inoculum was 
applied either to the sidewalls, or directly to 
Sample Inoculum applied Inoculum applied to Inoculum applied the bowl water but for MS2 bacteriophage the 
sites Location to the sidewalls the sidewalls to the bowl water inoculum as applied to the sidewalls only 
A Shelf 38 (20Æ5) 38 (21) 14 (8) 
B Cistern 45 (16Æ5) 45 (28) 11Æ5(4Æ5) 
C Seat (left) 171 (54) 47 (23Æ5) 20 (8) 
D Seat (right) 69 (31) 50 (18) 24Æ5 (4) 
E In front of toilet 42 (18Æ5) 34Æ5 (19) 11 (2Æ5) 

Values given within parenthesis are standard error of the mean for three replicate experiments. 

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 339–347, doi:10.1111/j.1365-2672.2005.02610.x 


POTENTIAL SPREAD OF INFECTION FROM THE DOMESTIC TOILET 343 

1·E + 09 
1·E + 08 
1·E + 07 
1·E + 06 
1·E + 05 
PFU CFU–1 m –2 
1·E + 04 
1·E + 03 
1·E + 02 
Fig. 2 Persistence of MS2 bacteriophage and 
Serratia on the porcelain surfaces of the toilet 
before and 60 min after flushing a seeded 
toilet. For Serratia, the effect of flushing the 1·E + 01 
toilet when the bowl water contained untreated bowl water is compared with disinfection and neutralization of the bowl water 
prior to flushing (bars represent the standard 
1·E + 00 
errors of the means for three replicate 
experiments). , Side (lt); side (rt); 
(, under rim (lt) and , under rim (rt) 

on the sidewalls and under the rim were broadly similar. 
Although the inoculum was not applied directly under the 
rim it was readily colonized with micro-organisms. This 
probably occurred from a
splash-back. effect as the force of 
the inoculum hitting the sides of the bowl bounced back 
under the rim, similar to
splash. effects that are likely to 
occur with acute episodes of diarrhoea. The initial level of 
contamination with MS2 bacteriophage on the bowl surface 
was c. 5 · 107 PFU cm)2 which was about 100-fold greater 
than for the initial bacterial contamination. Flushing the 
toilet reduced the level of surface attached bacteriophage by 

c. 3 log cycles cm)2. The initial surface counts of Serratia 
ranged from 4 · 104 to 5Æ8 · 105 CFU cm)2 and after 
MS2 phage untreated Serratia untreated Serratia bowl water 
bowl water bowl water disinfected and neutralised 

Before After Before After Before After 
Swab samples 
flushing, there was c. 100-fold cm)2 reduction. Following a 
second flush the number of surface attached Serratia 
declined by a further 10-fold cm)2 (data not shown). When 
the toilet was flushed after first disinfecting and neutralizing 
the bowl water prior to flushing the number of surface 
attached bacteria were broadly similar to the levels found 
when flushing in the presence of untreated bowl water. This 
indicates that the majority of surface attached bacteria are 
unlikely to have been derived from the bowl water splashing 
onto the bowl sides through turbulence. 

Figure 3a,b compares the reduction in the bacterial 
loading of the toilet water and the bacterial aerosol formation 
after three sequential flushes. Before flushing, the bowl 

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 339–347, doi:10.1111/j.1365-2672.2005.02610.x 


344 J. BARKER AND M.V. JONES 

(a) Serratia in toilet bowl water 
1·E + 09 
1·E + 08 
1·E + 07 
1·E + 06 
1·E + 05 
1·E + 04 
1·E + 03 
1·E + 02 
1·E + 01 
1·E + 00 

Before flush First flush Second flush Third flush 
Samples 
(b) Serratia in air samples 
CFU ml–3 CFU ml–1 

1·E + 04 

1·E + 03 

1·E + 02 

1·E + 01 

1·E + 00 

First flush Second flush Third flush 
Samples 

Fig. 3 Investigation of sequential flushes on bacteria persisting in the 
bowl water (a) and bacteria released into the air (b), comparing the 
effects of either; applying the inoculum to the sidewalls, or directly to 
the bowl water (bars represent the standard errors of the means for 
three replicate experiments). , Bowl water and (, side walls 

water contained c. 4 · 108 CFU ml)1, when the inoculum 
was applied directly to the water. The numbers in the bowl 
water were reduced to c. 2 · 108 CFU ml)1 after applying 
the inoculum to the sidewalls and clearly a considerable 
fraction of the semisolid agar inoculum ran down the walls 
contaminating the bowl water. A single flush reduced the 
number of bacteria in the bowl water by c. 2Æ0–3Æ0 log cycles 
and after the third flush the level had decreased to c. 
102 CFU m)1. Application of the inoculum to either the 
sides walls or the bowl water made little difference to the 
level of bacteria released into the air which was greatest after 
the first flush (@1300 CFU m)3). After the second flush the 
number of bacteria in the air declined to c. 500 CFU m)3.A 
third flush reduced the air count to 128 and 207 CFU m3, 
respectively, for the inoculum applied to either the bowl 
water or the sidewalls. Therefore, the reduction in air 
sample counts was not as great as in the bowl water, clearly, 
indicating the residual sidewall contamination as being a 
major contribution to the air loads. 

DISCUSSION 

This investigation simulated the effects of a person using a 
toilet during an attack of acute diarrhoea when there is likely 
to be substantial contamination of both the internal toilet 
bowl walls and the bowl water. We were able to show that 
considerable numbers of both bacteria and virus-laden 
particles were released into the air after flushing when 
seeded with c. 1010 micro-organisms, to mimic levels of 
bacterial/viral shedding that are known to occur during 
infectious diarrhoea (Thomson 1954; Hutchinson 1956; 
LeBaron et al. 1990; Caul 1994). One minute after the first 
flush c. 1370 CFU m)3 of Serratia were detected but 30 and 
60 min thereafter, the air count had declined by 20-and 
100-fold respectively. In contrast, when the toilet was 
flushed after first disinfecting and neutralizing the bowl 
water, the concentration in the air 1 min after flushing was c. 
350 CFU m)3. These data demonstrates that both the 
bacteria attached to the sidewalls and those present in the 
bowl water contribute to the aerosol formation. MS2 
bacteriophage was also released into air after toilet flushing 
with levels of contamination about twice that for bacteria, 
with 2240 PFU m)3 of virus particles detected in the air 
after the first flush. The air counts for both bacteria and 
viruses may have been considerably higher as a single-sieve 
impactor is known to be inefficient at capturing small 
particle sizes (Griffiths and Stewart 1999). Darlow and Bale 
(1959) estimated that c. 80% of air-borne particles released 
after flushing a toilet seeded with a liquid culture containing 
1011 Serratia were probably <4 lm. It is possible that our air 
sampling technique did not detect particles of <5 lm which 
are likely to remain suspended in the air for several hours 
but could, nevertheless, eventually settle onto surfaces. 

Closing the toilet lid had little effect in reducing the 
number of bacteria released into the air which was c. 
1000 CFU m)3 after the first flush (data not shown). 
Although splashes would probably have been contained by 
closing the lid, there was a gap of 15 mm between the top of 
the porcelain rim and the seat, and also a gap between the 
seat and the lid of 12 mm which would allow aerosols to 
escape into the room. Conversely, Darlow and Bale (1959) 
found that closing the lid reduced the aerosol concentration 
by a ratio of 1 : 2 but their measurements were performed 
using a
wash-down. toilet and an impinger air sampler. In 
contrast, Bound and Atkinson (1966) found that closing the 
lid did not significantly reduce the bacterial count in the air 
from a
wash-down. toilet seeded with E. coli using a slit 
sampler positioned at seat level. 

Sequential flushing of the seeded toilet resulted in 
prolonged air-borne transmission but with decreasing numbers of bacteria. Compared with the number of bacteria 
released into the air after the first flush, a second flush 
resulted in a threefold decrease and after the third flush the 

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 339–347, doi:10.1111/j.1365-2672.2005.02610.x 


POTENTIAL SPREAD OF INFECTION FROM THE DOMESTIC TOILET 345 

numbers had declined by almost 10-fold. The decline in airborne bacteria correlated with the decreasing numbers 
present in the bowl water. We found that reduction in 
numbers in the bowl water after flushing was similar to those 
reported by Newsom (1972). The first flush reduced the 
viable count in the water by 2–3 log cycles and by a similar 
amount after the second and third flushes. Even so, after the 
third flush up to 2 · 105 CFU were present in the 2 l 
volume of the bowl water. In contrast, the number of 
Serratia detected on the porcelain surfaces remained fairly 
constant after the initial flush showing that the organism had 
adhered to the surface. After applying the inoculum there 
was widespread contamination of the sidewalls; Serratia 
surface counts ranged from 104 to >105 CFU cm)2. 
Although flushing reduced the initial level of colonization 
by about 2 log cycles, c. 103 CFU cm)2 persisted on the 
surfaces despite repeated flushing (data not shown). When 
the bowl water was disinfected and neutralized prior to 
flushing it did not alter the level of bacteria attached to the 
sidewalls. Thus the bacteria surviving on the sidewalls are 
unlikely to have been derived from the bowl water splashing 
back onto the walls as the toilet was flushed. 

We also found that the recess under the rim of the toilet was 
heavily colonized with the test organisms. The recess under 
the rim of the toilet bowl has previously been found to be an 
area where Salmonella persisted in domestic homes where a 
family member had recently suffered an attack of salmonellosis with acute diarrhoea (Barker and Bloomfield 2000). The 
rim is an area of the toilet where limescale often accumulates, 
which aids bacterial retention and it can be difficult to clean 
effectively even with a toilet cleaner and scrubbing brush. 
Gerba et al. (1975) also found that a persistent fraction of 
seeded bacteria were absorbed onto the porcelain surface of 
the toilet and they concluded that subsequent elution of these 
organisms was responsible for continuing residual contamination in the toilet bowl water. In contrast, we found that 
after the initial seed inoculum was flushed from the sidewalls 
the numbers on the surface remained constant for several 
days of normal toilet use and thorough cleaning and 
disinfection using a toilet brush was required to remove the 
marker organisms to undetectable levels. 

Thirty minutes after flushing the toilet surface contamination was detected at various locations surrounding the 
toilet. The level detected was probably a minimum value 
because micro-organisms are subject to stress by aerosolization and can be further damaged by dehydration and 
impaction (Dark and Callows 1973; Griffiths and DeCosemo 1994; Griffiths 1998). The highest level of surface 
contamination was closet to the aerosol source, at the toilet 
seat level, however, the marker organisms were also found 
on the cistern and on a shelf, 41 and 83 cm above the toilet 
seat respectively. The particles captured by the settle plates 
were likely have been >20 lm because these are known to 

settle within a relatively short period compared with 
smaller-sized particles which can remain suspended for 
several hours (Chatigny et al. 1979). Our results support 
earlier studies (Darlow and Bale 1959; Gerba et al. 1975) 
that there is a risk that pathogens contaminating bathroom 
surfaces could spread to other family members. Organisms 
may be picked up by the clean hands of an uninfected 
person and cause infection, either by direct transfer from 
surface-to-hand-to-mouth, or transfer by handling ready-
to-eat foods (Barker et al. 2004). The number of bacteria/ 
viruses found in the toilet or on surrounding surfaces must 
be compared with the infectious dose. Although bacteria 
may multiply if they contaminate food and reach levels 
required for infection, clearly this does not happen with 
viruses. Nevertheless, many faecal–oral pathogens such as 
norovirus, rotavirus, Campylobacter and E. coli 0157 have 
infective doses as low as 10–100 micro-organisms (Dupont 
et al. 1972; LeBaron et al. 1990; Tauxe 1992; Caul 1994; 
Griffin et al. 1994; McDonnell et al. 1995) and we 
speculate that surface-to-hand-to-mouth transfer could 
occur with the levels of contamination that we found on 
the surfaces surrounding the toilet. 

The possibility that aerosols containing enteric pathogens 
could cause infection after being swallowed following 
deposition in the nose or pharynx was suggested by Darlow 
and Bale (1959) Recent epidemiological studies have provided convincing evidence to support this hypothesis. The 
likelihood of air-borne transmission of norovirus was 
demonstrated in an outbreak at a restaurant where no food 
source was implicated but analysis of the attack rate showed 
an inverse correlation with the distance from a person who 
had vomited (Marks et al. 2000). In infected persons up to 
1011 g)1 of virus particles have been detected in stools 
during viral gastroenteritis and with an average stool 
weighing 100 g the toilet bowl could contain 1013 virus 
particles. If there is a 2-log reduction in loading after an 
initial flush, the bowl water could still contain 1011 virus 
particles. Multiple trips to the toilet during diarrhoea are 
likely to result in large numbers of pathogens persisting in 
the toilet, both on the porcelain surfaces and in the bowl 
water. Our studies have shown that such contamination is 
likely to result in continuing air-borne spread on subsequent 
flushes. It would not be unreasonable to suggest that the 
persistence of enteric viruses within the air could be a 
potential infection risk via inhalation and swallowing. Airborne contamination could help to explain the high level of 
secondary spread of norovirus, within closed communities. 

In normal use the toilet is unlikely to present a great risk 
to health as formed stool is quickly washed away and does 
not create large numbers of bacterial aerosols (Newsom 
1972). In our opinion the health risk of using the toilet is 
likely to arise during acute episodes of gastroenteritis with 
the shedding of large numbers of pathogens. In this 

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 339–347, doi:10.1111/j.1365-2672.2005.02610.x 


346 J. BARKER AND M.V. JONES 

investigation, we were able to show when simulating loose 
stool that material deposited both on the sidewalls and in the 
bowl water were involved in the dissemination of microorganisms into the air and onto surrounding surfaces. 
Epidemiological studies from recurrent outbreaks of norovirus infection in successive cohorts of guests in hotels and 
on cruise ships (Ho et al. 1989; Gellert et al. 1994; 
Cheesbrough et al. 2000), suggests spread from infected 
persons after vomiting by settling of aerosol particles onto 
surfaces which are then touched by hands. In addition, these 
studies suggested that splashing or aerosol generation during 
toilet flushing may spread virus particles onto contact 
surfaces such as the toilet seat or flush handle. Combined 
with our experimental data we believe that the potential 
spread of enteric disease by contact with surfaces in 
bathrooms harbouring pathogens cannot be ignored and 
must be regarded as a serious infection risk.