BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

Beskæftigelsesudvalget 2019-20
BEU Alm.del Bilag 39
Offentligt

Waste Management 101 (2020) 241–249

Contents lists available at

ScienceDirect

Waste Management

journal homepage: www.elsevier.com/locate/wasman

Measures to reduce the exposure of waste collection workers to

handborne and airborne microorganisms and inﬂammogenic dust

Anne Mette Madsen

⇑

, Margit W. Frederiksen

, Mette Bjerregaard

, Kira Tendal

The National Research Centre for the Working Environment, Lersø Parkallé 105, DK-2100 Copenhagen

Ø,

Denmark

The Danish Cancer Society, Strandboulevarden 49, DK-2100 Copenhagen

Ø,

Denmark

a r t i c l e

i n f o

a b s t r a c t

Waste collection is associated with various health symptoms. The aims of this study were to obtain

knowledge about exposure to bacteria, fungi, and endotoxin during waste collection, and to study

whether it is possible to reduce the exposures and the total inﬂammatory potential (TIP) of those expo-

sures through simple interventions. The study was performed with an initial baseline exposure assess-

ment, a second assessment with intervention workers only, and a third with intervention and

reference workers.

The waste collection workers were exposed to 7.8

cfu bacteria/m

, 1.4

cfu fungi/m

, and 92

endotoxin units/m

(geometric mean values). The potential exposures in the truck cabs were up to 23

times higher than outdoor reference concentrations. For the intervention trucks and workers, airborne

fungi in the truck cab were reduced; fungi, bacteria, and yeasts on the steering wheels were reduced;

and the concentration of fungi on the workers’ hands was reduced.

Exposures were typically highest during collection of mixed household waste, in the summer, and for

collection using trucks with low loading height. The TIP was highest for the reference group sampling

mixed household waste, using trucks with low loading height, in the summer. Endotoxin, bacteria, and

fungi contributed to the TIP of 42 personal exposure assessments.

Conclusion:

Motivating workers to reduce exposure through simple interventions improved hand and

truck cab hygiene, but only slightly reduced personal exposure to airborne bioaerosols. Exposure can

be reduced by only using trucks with high loading height.

2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Article history:

Received 2 July 2019

Revised 23 September 2019

Accepted 10 October 2019

Keywords:

Intervention

Bacteria

Domestic waste

Bioaerosol

Fungi

Waste collection workers

1. Introduction

Studies from the 1990s found that work with waste collection

was associated with various symptoms related to the airways, gas-

trointestinal complaints, and skin irritation (Allmers

et al., 2000;

Bünger et al., 2000; Poulsen et al., 1995; Yang et al., 2001).

Publi-

cations from 2010 and forward show that waste collection work

is still associated with respiratory symptoms (Athanasiou

et al.,

2010; Darboe et al., 2015; Kuijer et al., 2010; Poole and Wong,

2013; Schantora et al., 2014)

including e.g. reduced lung function

(Athanasiou

et al., 2010; Vimercati et al., 2016),

chronic bronchitis

(Schantora

et al., 2014),

symptoms of the eyes (Schantora

et al.,

2014),

nail infections, gastrointestinal complaints, and dermato-

logical problems (Kuijer

et al., 2010).

Only a few studies on expo-

sure to bioaerosols during waste collection have been published

since 2010 – even though the working environment in some coun-

⇑

Corresponding author.

E-mail address:

[email protected]

(A.M. Madsen).

tries has changed due to new and expanded waste sorting instruc-

tions in order to increase recycling, and even though this expanded

sorting may cause reduced waste collection frequencies for some

types of waste (Madsen

et al., 2019).

However, the few studies

show that collection of household waste is still associated with ele-

vated exposure to bioaerosols (Lavoie

et al., 2006; Madsen et al.,

2016; Ncube et al., 2017; Park et al., 2011).

For workers collecting household waste very different exposure

levels have been measured in different studies, thus e.g. exposures

to endotoxin ranging from below detection level (bd) to 53 endo-

toxin units (EU)/m

(medians between 3.6 and 25 EU/m

) have

been found for Danish waste collectors (Nielsen

et al., 1997),

between <4 and 7182 EU/m

(geometric mean values (GM) = 40

EU/m

) for waste collectors in the Netherlands, and an average of

1123 EU/m

for workers collecting and sorting waste in South

Korea (Park

et al., 2011).

It is not known whether these differences

are related to differences in e.g. waste collection frequencies, waste

types, collection systems, etc. between e.g. countries. However,

studies have shown that season and humidity (Park

et al., 2011),

https://doi.org/10.1016/j.wasman.2019.10.023

0956-053X/Ó 2019 The Authors. Published by Elsevier Ltd.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

242

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

temperature (Madsen

et al., 2019),

waste type (Nielsen

et al.,

1997),

and collection equipment (Breum

et al., 1996; Nielsen

et al., 1997)

have an effect on the exposure level to some of the

measured bioaerosol components. Furthermore, the concentra-

tions of volatile organic compounds released from urban waste

increase

with

decreased

waste

collection

frequency

(Statheropoulos

et al., 2005).

In developed countries, the work tasks of waste collection

workers typically consist of truck driving, collection and unloading

of waste containers or sacks, and unloading of the waste at the

waste receiving plants. The waste collection workers usually lack

access to hand washing during their workday. The importance of

hand hygiene is mentioned in some papers concerning waste

workers (El-Wahab

et al., 2014; Kiviranta et al., 1999),

and Ncube

et al. call for further studies on methods to provide opportunity for

efﬁcient hand washing for municipal solid waste workers (Ncube

et al., 2017).

For workers collecting dental solid waste, yeasts from

the waste seem to contaminate the workers’ hands (Vieira

et al.,

2018).

The aims of this study were to obtain knowledge on (a) waste

collection workers’ exposure levels to airborne bacteria, fungi,

and endotoxin, (b) hand hygiene, and c) whether intervention in

the form of a combination of attention to and knowledge of

hygiene, conveniently available hand sanitizer, and adherence to

a few basic, potentially exposure-reducing guidelines could reduce

the exposure to airborne microorganisms and improve hand and

truck cab hygiene. The study was performed with an initial base-

line exposure assessment before the implementation of any inter-

vention measures, followed by a second exposure assessment of

intervention workers only, and ﬁnally a third exposure assessment

of intervention and reference workers.

were taken: The intervention teams were allocated dedicated

trucks; from the interventions were implemented and until the

end of the study, the intervention group also had to attend a

biweekly seminar with the project group to improve their motiva-

tion to pay attention to their health in general; and ﬁnally, we

made sure they were all aware that hand sanitizer was freely avail-

able. All workers already routinely used gloves during the collec-

tion of the waste, and all workers worked as both loaders and

drivers during the same working day.

The municipal waste was primarily collected from residential

areas with some small businesses. Inside the homes, mixed house-

hold waste is usually placed in a plastic bag, which is then disposed

of in a lidded plastic container speciﬁcally for mixed household

waste outside each house or apartment complex. All the trucks

were compactor trucks and equipped with lifts which tipped the

contents of the container into the rear end of the truck, but some

trucks had high waste loading height and some low loading height.

In total, 37 measurements were done on workers collecting mixed

household waste (the residual household waste fraction and the

biological waste fraction; in the following called household waste),

3 on workers collecting bulky waste, 1 collecting cardboard, and 1

collecting paper waste (Table

1).

Bulky waste is typically collected

every second month and left at the curb the night before, while

cardboard and paper are each sorted into a plastic container and

collected every 30 days (Madsen

et al., 2019).

Data on these three

types are considered together as ‘other waste’.

2.2. Sampling

Airborne inhalable microorganisms were sampled using per-

sonal and stationary GSP samplers (Gesamtstaubprobenahme, CIS

by BGI, INC Waltham, MA, USA) at a ﬂow rate of 3.5 l/min. The

samplers were mounted with polycarbonate ﬁlters (pore size

mm,

SUEZ – Water Technologies & Solutions, Feasterville-

Trevose, PA, USA) for quantiﬁcations of colony forming units

(cfu) of bacteria, fungi, and endotoxin analysis. Air ﬂows of the

samplers were checked before and after air sampling. In addition,

one cassette with a ﬁlter was brought to the workplace each day.

It was not connected to a pump and was used as a blank ﬁlter. Each

waste worker carried a backpack with a pump inside, connected to

a sampler which was attached to the shoulder strap of the back-

packs, close to the breathing zone. In addition, samplers were

mounted inside a total of 15 truck cabs of the garbage trucks col-

lecting household waste and on the rear end of those trucks. How-

ever, one sampler on the back of a truck was not running at the end

of the shift, and the sample was not included in the analyses

(Table

1).

An outdoor reference sample was taken where the work-

ers picked up their trucks in the morning. When the waste collec-

tion shift was over, the pumps were switched off and the samplers

were removed and immediately transported to the laboratory.

2.3. Extraction of dust with bacteria, fungi, and endotoxin from ﬁlters

Within 2 h after sampling, the bacteria and fungi collected on

polycarbonate ﬁlters were extracted in 6.0 ml sterile solution

(0.05% Tween 80 and 0.85% NaCl) by orbital shaking (500 rpm)

for 15 min at room temperature. The personal samples for endo-

toxin analysis were centrifuged (1000g) for 15 min to remove

particles.

2.4. Sampling from hand and steering wheel

Sampling was done using the eSwab transport system (eSwab;

Copan, Brescia, Italy), consisting of a ﬂocked nylon swab in 1 ml of

modiﬁed Amies liquid transport medium. At the end of the three

working days, hand samples were taken from both palms (which

2. Material and methods

2.1. Study design and interventions

Two groups were formed from volunteers among waste collec-

tion workers from a waste collection company in Copenhagen,

Denmark. The ﬁrst group included workers who had volunteered

to work according to the suggested intervention procedures (inter-

vention workers). The other group, whose members were recruited

on an ad-hoc basis the day before each exposure assessment, con-

sisted of workers who worked according to their normal routine

(reference workers). The study was performed with an initial base-

line exposure assessment before any intervention measures were

implemented (Day A, 7 February 2018). The second assessment

of exposure included only the intervention workers (Day B, 1

March 2018), and the ﬁnal, third exposure assessment included

both the intervention and reference workers (Day C, 11 June

2018) (Table

1).

After the ﬁrst exposure assessment, the intervention group was

given a lecture (28 February 2018) on what previous studies have

found regarding the impact of exposure on the health of waste col-

lection workers and potential ways to reduce exposure. The lecture

concluded with a question and answer session and a discussion of

pertinent issues. The intervention measures were put forth as sug-

gestions: Try to keep as much distance as possible to the waste

when loading and unloading the truck; try to keep the inside of

the truck cab clean and tidy; try to use hand sanitizer multiple

times during the working day; try to use clean gloves every day,

and try not to dry wet gloves in the truck cab. New instructions

for the intervention workers were to collect only one container

at a time and to use the built-in cart lift, which was operated from

a panel on the side of the truck. To support the intervention group

in integrating these measures into their daily routine, several steps

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

Table 1

Temperatures, number of waste workers collecting different waste types, working hours, and types and numbers of samples taken.

Day A

Exposure measurement

Baseline

Avg. temperature during the day

Number of intervention workers

and waste type

Working hours

Number of reference workers and

waste type

Working hours

Numbers and types of intervention

truck samples

Sampling time

Numbers and types of reference

truck samples

Sampling time

Waste loading height of trucks

Numbers of outdoor reference samples

243

Day B

Exposure

measurement

–

Lecture on

exposure

À5°C

bulk,

household

4.1 h

–

Focus on

health

Day C

Exposure measurement

°C

bulk, 1

paper,

cardboard

household

6.2 h

household

5.4 h

3 w, 3 air, 3 bt

6.8 h

3 w, 3 air, 3 bt

5.1 h

high, 5

low

À3°C

bulk, 9

household

4.5 h

household

3.8 h

2 w, 3 air, 3 bt

4.8 h

3 w, 2 air, 2 bt

3.8 h

high, 9

low

–

3 w, 3 air, 3 bt

3.9 h

–

high, 10

low

–

The numbers of workers collecting each waste type; hand swabs and air samples were collected from each person.

Working hours during the day of sampling = sampling hours for personal air samples.

w = steering wheel sample, air = air sample inside truck cab, bt = air sample from the back of the truck.

For air samples.

counted as 1 sample), and in total 42 samples were taken. In addi-

tion, samples were taken from the surface of the steering wheels,

with the sampled area in the shape of a cylinder with a length of

5 cm; in total 14 steering wheel samples were taken.

2.5. Culturing and quantiﬁcation of bacteria and fungi

Two-fold and ten-fold dilution series of extracts from polycar-

bonate ﬁlters were prepared and these as well as the hand and

steering wheel samples were plated on agar plates for quantiﬁca-

tion of bacteria or fungi. The number of fungi culturable on Dichlo-

ran Glycerol agar (DG-18 agar; Thermo Fisher Scientiﬁc Oxoid,

Basingstoke, UK), and bacteria on Nutrient agar (NA; Thermo Fisher

Scientiﬁc Oxoid, Basingstoke, UK) with actidione (cycloheximide;

50 mg/l; Serva, Germany), both incubated at 25

°C,

were counted

after 3, 5, 7, and 14 days of incubation, while gram-negative bacte-

ria (only personal air samples) were plated on SSI Enteric medium

(SSI-agar, SSI Diagnostica, Denmark) at 37

°C

and counted after

24 h. The data from air samples are presented as time-weighted

average exposures (TWA) in cfu/m

air, hand samples as cfu/hand

sample, and surface samples as cfu/ml. The term ‘bacteria’ is used

for bacteria on NA agar and the term ‘fungi’ is used for fungi on DG-

18 agar and does not include yeasts. The term ‘yeasts’ is used for

yeast species on DG-18 agar. Yeast and SSI-agar bacteria data are

only presented sparsely as some samples had only one or no colo-

nies. The detections limits of the air samples depends on the sam-

pling time. The maximum detection limit was for fungi, bacteria,

and bacteria on SSI-agar all 9 cfu/m

. For hand and steering wheel

samples the dl was 5 cfu/ml.

2.6. Endotoxin

The dust suspensions from GSP samplers were centrifuged

(1000g) for 15 min, and the supernatants were analyzed (in dupli-

cate) for endotoxin from gram-negative bacteria using the kinetic

Limulus Amoebocyte Lysate test (Kinetic-QCL endotoxin kit, Lonza

Pharma & Biotech, Walkersville, Maryland, USA). A standard curve

obtained from an Escherichia coli O55:B5 reference endotoxin was

used to determine the concentrations in terms of EU (11.0 EU

1.0 ng). The data are presented as EU/m

air, and the limit of detec-

tion was 0.05 EU/ml corresponding to a maximum detection limit

of 0.46 EU/m

2.7. Measurement of the total inﬂammatory potential

Measurement of the total inﬂammatory potential (TIP) was con-

ducted using an assay based on granulocyte-like cells. The assay is

based on the differentiated HL-60 cell line (Human Promyelocytic

Leukaemia cell line) which, upon exposure to inﬂammogens, will

react by producing reactive oxygen species (ROS), quantiﬁable by

a luminol-dependent chemiluminometric assay. The cells were dif-

ferentiated for 6–7 days by adding Tretinoin (ATRA) without chang-

ing the growth medium (RPMI 1640, Biological Industries,

Cromwell, CT, USA). The cells were seeded at 3

cells/ml and

incubated at 5% CO

at 37

°C

as described previously (Frankel

et al., 2014).

Personal samples were diluted 20 times with sterile

water with 0.001% Tween 80 to reduce the amount of Tween in

the samples. A volume of 100

of the diluted sample suspensions

were in duplicate inoculated into 50

of the cells. The GM concen-

trations of bacteria, endotoxin, and fungi in the diluted samples

were 64 cfu bacteria/ml, 0.86 EU/ml, and 117 cfu fungi/ml, respec-

tively. The chemiluminesence reaction caused by sample activity

was measured by a thermostated (37

°C)

ORION II Microplate lumi-

nometer (Berthold Detection Systems, Germany), which measured

relative light units per second (RLU/s) for 1 s every 120 s for

180 min. For every sample, accumulated RLU/s were calculated by

summing the RLU/s measurements throughout the 180 min period.

As references in each run, two endotoxin concentrations (1 and 5

EU/ml), zymosan (5

mg/ml),

and the extraction solution diluted 20

times were used. These references were used to see whether the

cells consistently reacted to the same level each time and to test

for contamination. To account for variations in sensitivities of the

cells, all data were normalized to the reaction of the diluted extrac-

tion solution. The maximum detection limit was 2

AUC/m

2.8. Treatment of data

Concentrations of fungi, bacteria, and endotoxin were log trans-

formed to be normally distributed while the TIP normalized to the

reaction of Tween was already normally distributed. For measure-

ments below the detection limit 50% of the dl was used. SAS ver-

sion 9.4 (SAS Institute, Cary, NC, USA) was used for statistical

analysis. First, the exposures of what would become the reference

and the intervention groups (the baseline data) were compared in

GLM (generalised linear model). Then, the data for all three days

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

244

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

1.7

[1.2

–2.7

]

Reference

workers

3. Results

1.3

[9.1

–1.8

]

= 10

1.6

[8.8

–3.8

]

= 13

3.1. Personal exposure to bacteria, fungi, and endotoxin

TIP (AUC/m

)

3.2. Factors affecting personal exposure

At baseline exposure assessment (Day A), no signiﬁcant differ-

ence was found between the reference group and what would

become the intervention group for fungi (p = 0.14), bacteria

(p = 0.34), endotoxin (p = 0.99), SSI-agar bacteria (p = 0.95), and

for numbers of working hours (p = 0.074). In the following factor

analysis, the baseline exposure levels are all analyzed as reference

exposure level.

Statistical analysis with all factors in one model with backward

transformation showed that exposures to fungi were highest on

Day C (p = 0.0037) and for household waste (p = 0.0003;

Fig. 1a),

and tended to be highest for low loading height (p = 0.072). When

the factors were studied one by one only waste type had a signif-

icant effect on fungal exposure (p = 0.0083). As the outdoor refer-

ence measurement also showed high fungal concentrations on

Day C, we re-analyzed the personal exposure data for the workers

collecting household waste, but with the outdoor reference for

each day subtracted. The signiﬁcant effect of day on exposure

was conﬁrmed (p = 0.028).

The exposures to bacteria were highest for workers collecting

household waste (p = 0.0018) and for workers having a truck with

low loading height (p = 0.038;

Fig. 1b).

When the factors were stud-

ied one by one, only waste type had a signiﬁcant effect

(p = 0.0061). Exposure to the SSI-agar bacteria was highest for ref-

erence workers (p = 0.027) and on Day C (p = 0.022) (Fig.

1c).

When

the factors were studied one by one, no single factor had a signif-

icant effect. The household waste tended to cause a higher expo-

sure to SSI-agar bacteria than other waste types (p = 0.091). The

exposure to endotoxin was signiﬁcantly affected by waste type

(p = 0.042;

Fig. 1d).

On Day 3 the workers collecting ‘other waste’ had signiﬁcantly

longer working days than workers collecting household waste

(p = 0.025) (further data not shown). The whole working day expo-

sure (the potentially inhaled dose based on numbers of working

hours) has been calculated. However, this did not change the

results regarding the effects of the studied factors on exposure

level (further data not shown).

The waste collection workers worked in pairs, and for 15 truck

teams, both workers participated in the sampling, while for an

additional 12 truck teams, only one member of the team partici-

Table 2

Personal exposures

to fungi, bacteria, and endotoxin, and total inﬂammatory potential of the exposures.

Day

For endotoxin, fungi, and bacteria the geometric mean values are presented and for TIP the averages are presented in

bold

followed by [ranges].

Baseline, before introducing the interventions. cfu = colony forming units; EU = endotoxin units; TIP = total inﬂammatory potential; AUC = area under curve.

Endotoxin (EU/m

)

The personal exposure was between 620 and 4.7

cfu bac-

teria/m

(GM = 7.8

); 1.2

and 2.2

cfu fungi/m

(GM = 1.4

); and 9 and 3570 EU/m

(GM = 92 EU/m

;

Table 2).

Exposure to fungi correlated signiﬁcantly with exposure to bacteria

(r = 0.46,

= 0.0021), bacteria on SSI-agar (r = 0.42,

= 0.0052), and

endotoxin (r = 0.42,

= 0.005); and exposure to endotoxin corre-

lated signiﬁcantly with exposure to bacteria (r = 0.82,

< 0.0001)

and bacteria on SSI agar (r = 0.47,

= 0.0018). Bacteria did not cor-

relate signiﬁcantly with bacteria on SSI-agar (r = 0.25,

= 0.10).

Reference

workers

Intervention

workers

Intervention

workers

131

[49.3–506]

= 10

100

[11.1–798]

= 13

1.1

[1.5

–3.3

]

= 13

–

9.9

[1.2

–7.0

]

= 13

–

9.3

[5.8

–3.3

]

Reference workers

6.4

[1.6

–1.9

]

= 10

Bacteria (cfu/m

)

Intervention

workers

5.9

[1.3

–1.6

]

1.5

[3.9

–1.1

]

= 10

Intervention

workers

1.9

[1.2

–2.2

]

= 10

Fungi (cfu/m

)

6.7

[2.7

–1.8

]

Reference workers

5.6

[620–4.7

]

= 10

7.1

[1.7

–2.6

]

[8.9–3570]

= 10

[42–213]

130

[42–630]

–

1.6

[1.4

–2.5

]

= 10

2.1

[1.4

–2.5

]

–

were treated together and the effects of interventions, person, day

of measurement, waste loading height, and waste type on exposure

and TIP of exposure were analyzed in a stepwise regression with

backward regression. Concentrations on hands were analyzed

using the GLM method, and Pearson’s correlation coefﬁcients (r)

were calculated between exposures and TIP. The impacts of endo-

toxin, bacteria, and fungi on the TIP of the 42 personal exposure

samples were studied in GLM in one model with backward

transformation.

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

245

Fig. 1.

Personal exposures (GM with conﬁdence limits) to fungi (a), bacteria (b), bacteria on SSI-agar (c), and endotoxin (d) presented according to the factors affecting the

exposure signiﬁcantly. High = high waste loading height; low = low waste loading height; household = household waste; other (waste) = bulk, cardboard, and paper waste; for

number of samples see

Table 1.

At baseline, the interventions were not yet implemented.

3.3. Inﬂammatory potential

On Day A, the TIP values for what would become the reference

group versus the future intervention group were not signiﬁcantly

different (p = 0.12), and all measurements on Day A are in the factor

analysis considered as reference measurements. When the data

from the three days were studied together, signiﬁcant effects of

the intervention (p = 0.015), loading height (p = 0.0029), day

(p = 0.040), and type of waste handled (p = 0.046) were found. The

highest TIP was found for the reference group, low loading height,

Day C, and household waste (Fig.

2).

The TIP correlated signiﬁcantly

with exposure to endotoxin (r = 0.58,

< 0.0001), bacteria (r = 0.67,

< 0.0001), and fungi (r = 0.37,

= 0.016), but not with exposure to

bacteria on SSI-agar (r = 0.24,

= 0.12). When the impacts of endo-

toxin, bacteria, and fungi on the TIP were studied in one model with

backward transformation, bacteria (p = 0.0040) and endotoxin

(p < 0.0001) had a signiﬁcant effect on the TIP. When endotoxin

was excluded from the model, TIP was associated signiﬁcantly with

both fungi (p = 0.0034) and bacteria (p < 0.0001). When bacteria

were excluded from the analysis, TIP was associated signiﬁcantly

with both fungi (p = 0.0073) and endotoxin (p = 0.0016).

3.4. Concentrations of airborne bacteria, endotoxin, and fungi in

stationary samples

Concentrations from the back end of the truck and in outdoor

reference measurements are presented in

Fig. 3abc.

Bacterial

concentrations from the truck may be underestimated due to

Fig. 2.

The mean total inﬂammatory potential (TIP; expressed as the mean AUC/m

air, with conﬁdence limits) of the personal air samples shown for the four factors

which signiﬁcantly affected the TIP value: Intervention, day of measurement, waste

loading height, and type of waste. On Day A, interventions were not yet

implemented. High = high waste loading height; low = low waste loading height;

household = household waste; other waste = bulk, cardboard, and paper waste;

ref = reference group, AUC = area under the curve.

pated. There was no signiﬁcant correlation between the exposures

of two workers on the same truck team: Bacteria (r =

À0.10,

= 0.74), fungi (r = 0.44,

= 0.10), endotoxin (r =

À0.01,

= 0.87),

and TIP (r = 0.38,

= 0.17).

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

246

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

Fig. 3.

Concentrations (GM with conﬁdence limits) of airborne fungi (a), bacteria (b), and endotoxin (c) in the air measured from the back end of the trucks and in outdoor

reference samples, and for the parameters resulting in signiﬁcant differences inside the cabs (Intervention, Day, and Loading height; d, e, and f). At baseline, the interventions

were not yet implemented.

competition from fungi on some of the plates. The bacterial con-

centrations on SSI-agar corresponded to 57 to 5.3

cfu/m

(GM = 662 cfu/m

). None of the measured exposure parameters

differed signiﬁcantly between reference and intervention trucks

(ps > 0.05). The fungal concentration (p = 0.033) was higher on

Day C than on Day B (Fig.

3a).

The concentrations of bacteria and

endotoxin were not affected by the sampling day (ps > 0.05). The

three blank samples did neither show growth of microorganisms

nor contamination with endotoxin.

The bioaerosol concentrations inside the truck cabs were

between 236 and 1.9

cfu fungi/m

(GM = 1.2

cfu/m

between 87 and 5.1

cfu bacteria/m

(GM = 1.5

cfu/

), and for endotoxin between 0.6 and 52 EU/m

(GM = 10.0

EU/m

). At baseline, there were no signiﬁcant differences between

concentrations of any of the measured bioaerosols of what would

become the reference and intervention groups (ps > 0.05). The fun-

gal concentration was signiﬁcantly higher on Day C (p = 0.0066),

and it was higher for reference truck cabs (p = 0.019) than for inter-

vention truck cabs (Fig.

3d).

The concentration of airborne bacteria

inside the truck cab was lower on Day B than during Days A and C

(p = 0.047) (Fig.

3e).

The endotoxin concentration was signiﬁcantly

associated with day (p = 0.047) with the lowest concentration on

Day B, and with loading height (p = 0.0033) with the highest con-

centration for low loading height (Fig.

3f).

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

247

The GM ratios of the concentrations of bacteria, fungi, and endo-

toxin inside the truck cabins to outdoor reference concentrations

were 7.6, 2.3, and 23, respectively. The ratios for bacteria were asso-

ciated with loading height with the highest ratios for low loading

height (p = 0.013). The ratios for fungi were associated with inter-

vention with the highest ratios for reference workers (p = 0.046).

The ratios for endotoxin were or tended to be associated with load-

ing height (p = 0.017) and intervention (p = 0.063) with the highest

ratios for low loading height and reference workers.

3.5. Microorganisms on palms

The concentrations of fungi on the workers’ palms were

between bd and 270 cfu/sample (GM = 59 cfu/hand sample). The

concentrations of yeasts were between bd and 1800 cfu/sample

(GM = 23 cfu/hand sample), and the concentrations of bacteria

between 400 and 4.1

cfu (GM = 3507 cfu/hand sample). At

baseline, the concentrations of yeasts and fungi were not signiﬁ-

cantly different between the intervention and the reference group

(ps > 0.05), but for bacteria the concentration was highest for the

future intervention group (p = 0.028).

On Day C, 9 of the 10 intervention workers told us that they had

used hand sanitizers 1–7 times during the day of measurement.

None of the reference workers used hand sanitizer. The number

of fungi/hand sample was signiﬁcantly affected by the intervention

(p = 0.017), day of sampling (p = 0.0093), and type of waste han-

dled (p = 0.0044) (Fig.

4a).

The number of bacteria/sample tended

to be affected by the type of waste handled (p = 0.097) with highest

concentrations for household waste. The number of yeasts/hand

sample was not affected signiﬁcantly by the intervention

(p > 0.05). There was no signiﬁcant correlation between the num-

ber of bacteria/hand sample and the number of bacteria in the per-

sonal air samples (p > 0.05). However, the concentration of fungi

on hands and in the personal samples correlated signiﬁcantly

(r = 0.62,

< 0.0001). Concentrations of fungi and bacteria on the

workers’ hands did not correlate signiﬁcantly (r = 0.20,

= 0.20),

but concentrations of yeasts correlated with concentrations of

fungi (r = 0.39,

= 0.0098) and bacteria (r = 0.55,

= 0.0002) on

the workers’ palms.

3.6. Steering wheel surface

As with the hand samples, yeasts constituted a considerably

fraction of the fungi sampled from the steering wheel surface,

and yeast and fungus data are treated separately. The concentra-

tions of fungi and bacteria correlated signiﬁcantly (r = 0.70,

= 0.0057), but neither correlated with concentrations of yeasts

(ps > 0.05). At baseline, no difference was found between what

would become the reference and the intervention groups for the

steering wheel samples for fungi (p = 0.66), bacteria (p = 0.11),

and yeasts (p = 0.92). The interventions tended to be or were asso-

ciated with lower concentrations of fungi (p = 0.056) (Fig.

4b),

bac-

teria (p = 0.012) (Fig.

4c),

and yeasts (p = 0.0068) (Fig.

4d),

and in

addition the yeast concentrations tended to be associated with

the day (p = 0.066) while bacteria concentrations were associated

signiﬁcantly with day of measurement (p = 0.041).

Fig. 4.

Concentrations (GM with conﬁdence limits) of fungi on hands (a); on the steering wheels of intervention and reference trucks, fungi, bacteria, and yeasts (b, c, d). Other

waste = bulk, cardboard, and paper waste. For number of samples, see

Table 1.

At baseline, the interventions were not yet implemented.

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

248

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

4. Discussion

Waste workers are expected to be exposed to microorganisms

primarily via the air and hand-to-mouth. For exposure via air, this

study shows that the TIP of the personal exposure was reduced

post implementation of interventions. For the hand route, concen-

trations of fungi on the workers’ palms were reduced for the inter-

vention waste collection workers, and in addition, the truck cab

hygiene was improved.

The GM exposure to endotoxin during collection of ‘other waste’

was 22 EU/m

(n = 5) and the GM exposure for household waste

collection was 109 EU/m

(n = 37). In a previous study, workers

collecting different types of waste in the summer and fall were

exposed to 40 EU/m

(GM), and the group working as loaders col-

lecting residual waste were exposed to 49 EU/m

(GM) (Wouters

et al., 2006).

The workers in the present study worked as both dri-

vers and loaders throughout the day. The suggested occupational

exposure limits for endotoxin are 50 EU/m

(Douwes

and

Heederik, 1997)

and 150 EU/m

(Smid

et al., 1992),

which were

exceeded for respectively 35 and 11 out of 42 personal exposure

measurements.

In the winter, the waste collection workers were exposed to up

to 3.3

cfu bacteria/m

air (GM = 8.9

cfu/m

) while the

average outdoor reference was 124 cfu bacteria/m

. In a previous

study, we found that 13 waste workers were exposed to bacterial

concentrations between 112 and 4.8

(GM = 1.1

) cfu/

air in the winter (Madsen

et al., 2016).

In a study from Denmark

published in 1995, 20 waste workers collecting mixed household

waste were exposed to

cfu bacteria/m

air

(mean = 6.4

10 cfu/m air; the season was not mentioned

(Nielsen

et al., 1995).

In the winter, personal exposure to fungi

was 1.1

cfu/m

air (GM = 1.1

cfu/m

air) while the

average outdoor reference was 217 cfu/m

air. In the previously

mentioned study (Madsen

et al., 2016),

the 13 waste workers were

exposed to 4.6

cfu/m

air (GM = 5.7

cfu/m

air) in the

winter, while the 20 workers (Nielsen

and Breum, 1995)

were

exposed to 5

cfu fungi/m

air (mean = 7.7

cfu/m

air). Thus, the exposure levels seem not to have changed consider-

ably. In a study from South Africa, the exposure of 20 waste collec-

tion workers ranged from 5.8

to 1.36

cfu/m

fungi

(Ncube

et al., 2017).

There are no occupational exposure limits

for fungal and bacterial exposure, which is probably because the

health effects caused by the microorganisms differ greatly at spe-

cies level.

At baseline, the future intervention group’s exposure to air-

borne microorganisms was at the same level as that of the refer-

ence group, and therefore differences are expected to be related

to the interventions. The interventions had signiﬁcant effects on

the waste collection workers’ exposure to bacteria measured on

SSI-agar and to the TIP of the exposure, but not to fungi and bacte-

ria in general or to endotoxin. The workers themselves may also be

the sources of a, probably small, portion of their own bacterial

exposures and a fraction of the airborne bacteria in the truck cab,

but this alone cannot explain why the interventions did not affect

the personal exposure to bacteria. The TIP and exposures corre-

lated, and the TIP measurement combines the effects of all expo-

sures, and as the fungal exposure on Day C tended to be higher

for the reference group this may have contributed to the measured

effect of the intervention on TIP.

The fungal exposure inside the truck cabs was reduced for inter-

vention workers, and the bacterial and endotoxin concentrations

inside the truck cabs were lower on Day B than before the imple-

mentation of the interventions. A previous study shows that waste

collection workers spend 37% of their working day inside the truck

cab (Madsen

et al., 2016),

and therefore it is also relevant to reduce

exposure inside the truck cabs. On Day C, which was in the sum-

mer, windows were open in the truck cabs, and a high air change

rate is expected to cause a reduced bacterial concentration as seen

in indoor air (Madsen

et al., 2018).

This may have contributed to or

caused the lack of difference between intervention and reference

cabins for bacteria and endotoxin on Day C. The GM concentrations

of bacteria and fungi in the truck cabs were 1.5

cfu bacteria/

air and 1.2

cfu fungi/m

air, respectively, and thus these

exposures are at the same levels as previously found inside truck

cabs (Nielsen

et al., 1995).

In this study, samples from the workers’ hands had 270 cfu

fungi/hand sample,

1800 cfu yeasts/hand sample, and

3507 cfu bacteria/hand sample, which is lower than what was

found on the palms of three waste workers in South Korea

(means = 6.4

cfu fungi/cm

palm and 1.4

cfu bacteria/

palm) (Madsen

et al., 2016; Park et al., 2011).

The workers

with access to hand sanitizer had signiﬁcantly fewer fungi on their

hands, but there was no signiﬁcant effect on bacteria and yeasts.

Previous studies of health care workers have shown a signiﬁcant

effect of hand sanitizer on the total number of bacteria (Larocque

et al., 2016),

and

in vitro

studies have found that a 15 s exposure

of e.g.

Aspergillus ﬂavus

and

Penicillium citrinum

spores to hand san-

itizer kills the spores (Fendler

and Groziak, 2002).

The lack of effect

on bacteria and yeasts seen in this study could be because the bac-

teria and yeasts might be aggregated in clusters and thus are better

protected, as well as presence of yeasts and bacteria inside the

gloves, but it could also reﬂect the workers’ own skin bacteria

and yeasts. Some of the intervention workers told us that they

had introduced a routine where they cleaned the inside of the

truck cab while reﬁlling the gas tank. Accordingly, the concentra-

tions of fungi, yeasts, and bacteria on the steering wheels were also

lowest in the intervention truck cabs. In a study concerning bacte-

ria on smartphones, cleaning was also found to have a signiﬁcant

effect on the number of bacteria (Egert

et al., 2015).

In this study, using a truck with a high waste loading height for

collection of household waste in containers caused lower exposure

levels to bacteria and a lower TIP of the exposure compared to

using a truck with low waste loading height. The effect of loading

height was large enough that even inside the cabs of trucks with

high loading height, exposure to endotoxin was lower. Top loading

trucks have previously been observed to cause lower exposure to

fungi than trucks which are loaded at the same level with the

workers’ breathing zone (Nielsen

et al., 1997).

Waste type also

had an effect on the workers’ exposure to fungi, bacteria, endo-

toxin, and on the associated TIP with higher exposure levels during

collection of mixed household waste than during collection of

’other waste’ (cardboard, paper, and bulk categorized together).

Waste collection workers also had more fungi on their hands if

they collected household waste rather than ‘other waste’.

The exposures to fungi and bacteria on SSI-agar as well as the

TIP of the exposures were signiﬁcantly higher in the summer than

in the winter. This is expected to be related to a faster growth rate

of fungi during higher temperatures. Concentrations of bacteria,

fungi, and endotoxin were low inside the truck cabs on Day B when

the outdoor temperatures were low and the interventions had

been implemented. Seasonal or temperature variations in exposure

are in accordance with previous studies (Madsen

et al., 2019;

Nielsen et al., 1997)

and of relevance as temperatures are in gen-

eral rising in many regions of the world. In relation to this, it

should be noted that gastrointestinal symptoms among waste

workers collecting mixed household waste are most abundant in

the summer (Ivens

et al., 1999).

In conclusion, waste workers are exposed to high concentra-

tions of airborne fungi, bacteria, and endotoxin. By simple inter-

ventions, it was possible to reduce the TIP of the workers’

BEU, Alm.del - 2019-20 - Bilag 39: Orientering om NFA-artikel om tiltag for at reducere skraldemænds udsættelse for mikroorganismer, fra beskæftigelsesministeren

A.M. Madsen et al. / Waste Management 101 (2020) 241–249

249

exposures, the concentration of fungi on hands, concentrations of

airborne fungi inside the truck cabs, and concentrations of fungi,

bacteria, and yeasts on the steering wheels. However, the type of

waste handled, the type of truck used (high or low loading height),

and season also had signiﬁcant effects on more of the exposures. In

general, household waste, low waste loading height, and summer

were associated with the highest exposures.

Acknowledgments

The waste workers are highly acknowledged for their involve-

ment in the project. Vivian Bossen Mose is highly acknowledged

for her involvement.

Funding

This study has obtained economical support from the National

Research Centre of the Working Environment (NFA), the Munici-

pality of Copenhagen.

Declaration of Competing Interest

All authors declare no conﬂicts of interest in this paper.

References

Allmers, H., Huber, H., Baur, X., 2000. Two year follow-up of a garbage collector with

allergic bronchopulmonary aspergillosis (ABPA). Am. J. Ind. Med. 37, 438–442.

Athanasiou, M., Makrynos, G., Dounias, G., 2010. Respiratory health of municipal

solid waste workers. Occup. Med. 60, 618–623.

Breum, N.O., Nielsen, B.H., Nielsen, E.M., Poulsen, O.M., 1996. Bio-aerosol exposure

during collection of mixed domestic waste - an intervention study on

compactor truck design. Waste Manage. Res. 14, 527–536.

Bünger, J., Antlauf-Lammers, M., Schulz, T.G., Westphal, G.A., Müller, M.M., Ruhnau,

P., Hallier, E., 2000. Health complaints and immunological markers of exposure

to bioaerosols among biowaste collectors and compost workers. Occup.

Environ. Med. 58, 458–464.

Darboe, B., Kao, M.-Y., Tsai, D., 2015. Respiratory symptoms among municipal waste

workers in the Gambia: types of solid waste and working conditions. Int. J.

Health Promot. Educ. 53, 17–27.

Douwes, J., Heederik, D., 1997. Health-based recommended occupational exposure

limit for endotoxins (Draft document). Dutch expert committee

Gezondheidsraad, Rijswijk (the Netherlands).

Egert, M., Späth, K., Weik, K., Kunzelmann, H., Horn, C., Kohl, M., Blessing, F., 2015.

Bacteria on smartphone touchscreens in a German university setting and

evaluation of two popular cleaning methods using commercially available

cleaning products. Folia Microbiol. 60, 159–164.

El-Wahab, E.W.A., Eassa, S.M., Lotﬁ, S.E., El Masry, S.A., Shatat, H.Z., Kotkat, A.M.,

2014. Adverse health problems among municipality workers in Alexandria

(Egypt). Int. J. Prev. Med. 5, 545.

Fendler, E., Groziak, P., 2002. Efﬁcacy of alcohol-based hand sanitizers against fungi

and viruses. Infect. Control Hosp. Epidemiol. 23, 61–62.

Frankel, M., Hansen, E.W., Madsen, A.M., 2014. Effect of relative humidity on the

aerosolization and total inﬂammatory potential of fungal particles from dust-

inoculated gypsum boards. Indoor Air 24, 16–28.

Ivens, U.I., Breum, N.O., Ebbehoj, N., Nielsen, B.H., Poulsen, O.M., Wurtz, H., 1999.

Exposure-response relationship between gastrointestinal problems among

waste collectors and bioaerosol exposure. Scand. J. Work Environ. Health 25,

238–245.

Kiviranta, H., Tuomainen, A., Reiman, M., Laitinen, S., Nevalainen, A., Liesivuori, J.,

1999. Exposure to airborne microorganisms and volatile organic compounds in

different types of waste handling. Ann. Agric. Environ. Med. 6, 39–44.

Kuijer, P.P.F., Sluiter, J.K., Frings-Dresen, M.H., 2010. Health and safety in waste

collection: Towards evidence-based worker health surveillance. Am. J. Ind. Med.

53, 1040–1064.

Larocque, M., Carver, S., Bertrand, A., McGeer, A., McLeod, S., Borgundvaag, B., 2016.

Acquisition of bacteria on health care workers’ hands after contact with patient

privacy curtains. Am. J. Infect. Control 44, 1385–1386.

Lavoie, J., Dunkerley, C.J., Kosatsky, T., Dufresne, A., 2006. Exposure to aerosolized

bacteria and fungi among collectors of commercial, mixed residential,

recyclable and compostable waste. Sci. Total Environ. 370, 23–28.

Madsen, A.M., Alwan, T., Orberg, A., Uhrbrand, K., Jorgensen, M.B., 2016. Waste

workers’ exposure to airborne fungal and bacterial species in the truck cab and

during waste collection. Ann. Occup. Hyg. 60, 651–668.

Madsen, A.M., Frederiksen, M.W., Kurdi, I.M., Sommer, S., Flensmark, E., Tendal, K.,

2019. Expanded cardboard waste sorting and occupational exposure to

microbial species. Waste Manag. 87, 345–356.

Madsen, A.M., Moslehi-Jenabian, S., Islam, M.Z., Frankel, M., Spilak, M., Frederiksen,

M.W., 2018. Concentrations of

Staphylococcus

species in indoor air as associated

with other bacteria, season, relative humidity, air change rate, and

S. aureus-

positive occupants. Environ. Res. 160, 282–291.

Ncube, F., Ncube, E.J., Voyi, K., 2017. Bioaerosols, noise, and ultraviolet radiation

exposures for municipal solid waste handlers. J. Environ. Public Health 2017,

3081638.

Nielsen, B.H., Breum, N.O., 1995. Exposure to air contaminants in chicken catching.

Am. Ind. Hyg. Assoc. J. 56, 804–808.

Nielsen, E.M., Breum, N.O., Nielsen, B.H., Wurtz, H., Poulsen, O.M., Midtgaard, U.,

1997. Bioaerosol exposure in waste collection: A comparative study on the

signiﬁcance of collection equipment, type of waste and seasonal variation. Ann.

Occup. Hyg. 41, 325–344.

Nielsen, E.M., Nielsen, B.H., Breum, N.O., 1995. Occupational bioaerosol exposure

during collection of household waste. Ann. Agric. Environ. Med. 2, 53–59.

Park, D.U., Ryu, S.H., Kim, S.B., Yoon, C.S., 2011. An assessment of dust, endotoxin,

and microorganism exposure during waste collection and sorting. J. Air Waste

Manag. Assoc. 61, 461–468.

Poole, C., Wong, M., 2013. Allergic bronchopulmonary aspergillosis in garden waste

(compost) collectors—occupational implications. J. Occupat. Med. 63, 517–519.

Poulsen, O.M., Breum, N.O., Ebbehøj, N., Hansen, Å.M., Ivens, U.I., Van Lelieveld, D.,

Malmros, P., Mathiassen, L., Nielsen, B.H., Nielsen, E.M., Schibye, B., Skov, T.,

Stenb

k, E.I., Wilkins, C.K., 1995. Collection of domestic waste. Review of

occupational health problems and their possible causes. Sci. Total Environ. 170,

1–19.

Schantora, A.L., Casjens, S., Deckert, A., van Kampen, V., Neumann, H.D., Brüning, T.,

Raulf, M., Bünger, J., Hoffmeyer, F., 2014. Prevalence of work-related rhino-

conjunctivitis and respiratory symptoms among domestic waste collectors. In:

Pokorski, M. (Ed.), Environment Exposure to Pollutants. Springer International

Publishing, pp. 53–61.

Smid, T., Heederik, D., Houba, R., Quanjer, P.H., 1992. Dust- and endotoxin-related

respiratory effects in the animal feed industry. Am. Rev. Respir. Dis. 146, 1474–

1479.

Statheropoulos, M., Agapiou, A., Pallis, G.J.A.E., 2005. A study of volatile organic

compounds evolved in urban waste disposal bins. Atmospheric Environ. 39,

4639–4645.

Vieira, C.D., Tagliaferri, T.L., de Carvalho, M.A.R., de Resende-Stoianoff, M.A.,

Holanda, R.A., de Magalhães, T.F.F., Magalhães, P.P., dos Santos, S.G., de

Macêdo Farias, L., 2018. Investigating cross-contamination by yeast strains

from dental solid waste to waste-handling workers by DNA sequencing.

MicrobiologyOpen 7, e00554.

Vimercati, L., Baldassarre, A., Gatti, M.F., De Maria, L., Caputi, A., Dirodi, A.A.,

Cuccaro, F., Bellino, R.M., 2016. Respiratory health in waste collection and

disposal workers. Int. J. Environ. Res. Public Health 13, 631–638.

Wouters, I.M., Spaan, S., Douwes, J., Doekes, G., Heederik, D., 2006. Overview of

personal occupational exposure levels to inhalable dust, endotoxin,

b(1->3)-

glucan and fungal extracellular polysaccharides in the waste management

chain. Ann. Occup. Hyg. 50, 39–53.

Yang, C.Y., Chang, W.T., Chuang, H.Y., Tsai, S.S., Wu, T.N., Sung, F.C., 2001. Adverse

health effects among household waste collectors in Taiwan. Environ. Res. 85,

195–199.