Airborne molds and bacteria, microbial volatile organic compounds (MVOC), plasticizers and formaldehyde in dwellings in three North European cities in relation to sick building syndrome

Science of the Total Environment 444 (2013) 433–440

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

Airborne molds and bacteria, microbial volatile organic compounds (MVOC),plasticizers and formaldehyde in dwellings in three North European cities in relationto sick building syndrome (SBS)

Bo Sahlberg a,⁎, Maria Gunnbjörnsdottir b,f,g, Argo Soon c,d, Rain Jogi e, Thorarinn Gislason f,g,Gunilla Wieslander a, Christer Janson b, Dan Norback a

a Department of Medical Sciences, Occupational and Environmental Medicine, Uppsala University, Uppsala, Swedenb Department of Medical Science, Respiratory Medicine and Allergology, Uppsala University, Swedenc Department of Public Health, University of Tartu, Tartu, Estoniad Research and Cooperation Centre, Archimedes foundation, Tartu, Estoniae Foundation Tartu University Clinics, Lung Clinic, Tartu, Estoniaf Faculty of Medicine, University of Iceland, Icelandg Department of Respiratory Medicine and Sleep, Landspitali University Hospital, Iceland

H I G H L I G H T S

► We examine whether MVOCs, airborne bacteria, molds, formaldehyde and plasticizers in dwellings were associated with SBS.► We also study the associations between MVOCs and reports on dampness and mold.► The indoor levels of some MVOCs were positively associated with SBS.► Levels of airborne molds and bacteria and some MVOCs were higher in dwellings with a history of dampness and molds.► Information on levels of individual MVOCs is of more value than the total level of MVOCs.

⁎ Corresponding author at: Department of Medical SE-mail address: [email protected] (B. Sahlbe

0048-9697/$ – see front matter © 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.scitotenv.2012.10.114

a b s t r a c t

Article history:Received 11 June 2012Received in revised form 25 October 2012Accepted 25 October 2012Available online 29 December 2012

Keywords:Microbial volatile organic compounds(MVOCs)Indoor environmentSick building syndrome (SBS)DwellingDampnessMold

There are few studies on associations between airborne microbial exposure, formaldehyde, plasticizers indwellings and the symptoms compatible with the sick building syndrome (SBS). As a follow-up of the EuropeanCommunity Respiratory Health Survey (ECRHS II), indoor measurements were performed in homes in threeNorth European cities. The aim was to examine whether volatile organic compounds of possible microbialorigin (MVOCs), and airborne levels of bacteria, molds, formaldehyde, and two plasticizers in dwellingswere associated with the prevalence of SBS, and to study associations between MVOCs and reports ondampness and mold.The study included homes from three centers included in ECRHS II. A total of 159 adults (57% females) par-ticipated (19% from Reykjavik, 40% from Uppsala, and 41% from Tartu). A random sample and additionalhomes with a history of dampness were included. Exposure measurements were performed in the 159homes of the participants. MVOCs were analyzed by GCMS with selective ion monitoring (SIM). Symptomswere reported in a standardized questionnaire. Associations were analyzed by multiple logistic regression.In total 30.8% reported any SBS (20%mucosal, 10% general, and 8% dermal symptoms) and 41% of the homes hada history of dampness and molds There were positive associations between any SBS and levels of 2-pentanol(P=0.002), 2-hexanone (P=0.0002), 2-pentylfuran (P=0.009), 1-octen-3-ol (P=0.002), formaldehyde (P=0.05), and 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate (Texanol) (P=0.05). 1-octen-3-ol (P=0.009)and 3-methylfuran (P=0.002) were associated with mucosal symptoms. In dwellings with dampness andmolds, the levels of total bacteria (P=0.02), total mold (P=0.04), viable mold (P=0.02), 3-methylfuran (P=0.008) and ethyl-isobutyrate (P=0.02) were higher.In conclusion, some MVOCs like 1-octen-3-ol, formaldehyde and the plasticizer Texanol, may be a risk factorfor sick building syndrome. Moreover, concentrations of airborne molds, bacteria and some other MVOCswere slightly higher in homes with reported dampness and mold.

© 2012 Elsevier B.V. All rights reserved.

ciences, Occupational and Environmental Medicine, Uppsala University, 75185 University Hospital, Uppsala, Sweden.rg).

rights reserved.

434 B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

1. Introduction

There is increased concern about the health effects of the indoor en-vironment (Norback, 2009) and the home is the indoor environmentwhere both adults and children spend most of their time. Several stud-ies have shown that building dampness and indoor mold growth arecommon in dwellings (Bornehag et al., 2001; Brunekreef, 1992; Fisk etal., 2007; Husman, 1996; WHO, 2009). Building dampness and indoormold growth is associated with an increased prevalence of both asth-matic symptoms and symptoms compatible with the sick building syn-drome (SBS) (Bornehag et al., 2001; Brunekreef, 1992; Fisk et al., 2007;Husman, 1996; Peat et al., 1998). Dampness approximately doubles therisk of health effects (Bornehag et al., 2001; Peat et al., 1998). One largestudy in Swedishmulti-family residential buildings found a strong asso-ciation between building dampness and SBS symptoms (Engvall et al.,2001).

SBS is a set of non-specific symptoms (from eyes, upper airways,facial skin, headache, tiredness and nausea) occurring in a particularbuilding (Norback and Edling, 1991). Various indoor factors, such asa low supply rate of outdoor air, high room temperature and low in-door air humidity have been shown to influence the prevalence ofSBS-symptoms (Apter et al., 1994; Hodgson, 1995; Mendell, 1993;Norback, 2009). Moreover, personal factors such as female genderand history of allergic disorder have been associated with SBS(Apter et al., 1994; Bjornsson et al., 1998; Hodgson, 1995; Mendell,1993).

Building dampness may lead to an increased exposure to varioustypes of molds, bacteria, microbial compounds and chemical compounds(WHO, 2009). It is well known thatmolds and bacteria emit certain VOCs,so-called microbial volatile organic compounds (MVOCs), when growingon building materials. Examples of compounds that can be producedby microorganisms include certain ketones (e.g., 2-heptanone), alcohols(e.g., 1-octen-3-ol), terpenes and terpene derivatives (e.g., geosmin) andsulfur compounds (e.g., dimethyl disulfide) (Claeson et al., 2002; Korpiet al., 1998; Sunesson et al., 1996; Wilkins et al., 2000).

One early study, which analyzed MVOCs by gas chromatography–mass spectrometry (GC–MS) with selective ion monitoring (SIM) inair samples from damp and moldy buildings, reported data on con-centrations of 26 different MVOCs. The concentrations were usuallyhigher in indoor air than outdoor air (Wessen and Schoeps, 1996)One intervention study found that MVOC concentration was higherin a hospital building with dampness and SBS symptoms as comparedto a dry control building (Wieslander et al., 2007). Another studyreported that children living in dwellings with higher MVOC levelshad a higher prevalence of asthma, hay fever, wheezing, and eye irri-tation, although the difference was not statistically significant (Elke etal., 1999). In a school environment study it was found that exposureto several MVOCs at school was associated with asthmatic symptomsin the pupils (Kim et al., 2007), while an earlier school environmentstudy reported an association between asthma in school teachers andlevels of MVOCs in the classrooms (Smedje et al., 1996). We foundonly one previous study on associations between MVOCs and SBS indwellings. In that study, some MVOCs, especially 1-octen-3-ol, wereassociated with SBS (Araki et al., 2010).

The use of MVOCs as a marker of fungal or microbial exposure hasbeen criticized, since these compounds are also emitted from othersources such as tobacco smoke, plants, furniture, furnishing, and build-ing materials (Korpi et al., 1999, 1998; Schleibinger et al., 2008). Onestudy comparing MVOC levels in moldy and mold-free dwellingsfound no significant association between most MVOCs and the moldstatus. Only 2-methyl-1-butanol and 1-octen-3-ol showed significantbut weak associations with mold status (Schleibinger et al., 2008).

Dampness may also influence the levels of volatile organic com-pounds which are not produced by microorganisms. Formaldehydeis a reactive volatile compound that may induce airway irritation atlow concentrations, and the emission may be influenced by dampness

in building materials (Sarigiannis et al., 2011). Indoor sources includecigarette smoke, insulating materials, particle board and plywood fur-niture containing formaldehyde-based resins, water based paints,fabrics, and various other consumer products. Plasticizers are anothergroup of compounds that are commonly found in dwellings, and usedin plastic material. Associations between phthalates, and asthma and rhi-nitis in pre-school children have been reported (Bornehag et al., 2004;Hsu et al., 2011; Naydenov et al., 2008). One study showed that buildingdampness in dwellings was associated with increased dust levels ofphthalates, a common plasticizer in poly-vinyl-chloride (PVC) materials(Bornehag et al., 2005). Moreover, one school study found an associationbetween the indoor air concentration of two common plasticizers(TMPD-MIB (2,2,4-trimethyl-1,3-pentanediolmonoisobutyrate, Texanol)and TMPD-DIB (2,2,4-trimethyl-1,3-pentanediol diisobutyrate, TXIB),and asthmatic symptoms in school children (Kim et al., 2007). Finally,the compound 2-ethyl-1-hexanol has been reported to be produced bymolds (Claeson et al., 2002) as well as degradation of the plasticizerdi-ethyl-hexyl phthalate (DEHP) or acrylate polymers in water-basedcarpet glue. This degradation is related to increased dampness in floorconstruction (Norback et al., 2000).

The objective of this study was to examine whether air bornelevels of MVOCs, bacteria, molds, formaldehyde, Texanol and TXIB indwellings in northern Europe are associated with the prevalence ofsick building syndrome (SBS) symptoms. In addition, we investigatedwhether there was an association between levels of MVOCs and re-ports on dampness and molds in the home.

2. Materials and methods

2.1. Study design and population

Our study is based on data from the Uppsala, Reykjavik and Tartucenters of the European Community Respiratory Health Survey(ECRHS) (Janson et al., 2001). ECRHS is a multi-center populationstudy carried out on a random sample of subjects aged 20–44 years in1990–1994 (Burney et al., 1994)with a follow-up in 2002 (ECRHSII)(The European Community Respiratory Health Survey II, 2002). Of atotal of 1238 subjects from ECRHS II in these centers, 510 included inthe random sample had not moved (from one dwelling to another)since the ECRHS I survey. Sixty of these were randomly selected fromeach center and invited to participate in the present study by providingair measurements from their home environment, as previously de-scribed (Gunnbjornsdottir et al., 2009). In addition all subjects whohad reported having dampness or indoor molds in the screening ques-tionnaire, but were not included in the 60 randomly selected, were in-vited as well. A total of 129 subjects from the random sample (42 inReykjavik, 46 in Uppsala and 41 in Tartu) and subjects reportingwater leakage/flooding, signs of dampness in the floor construction, vis-ible indoor molds or moldy odor in the dwelling during the last12 monthswere invited. Finally, 159 subjectswithMVOCdata availablewere included (28 random samples and 3 additional dampness samplesfrom Reykjavik, 43 random samples and 20 additional dampness sam-ples fromUppsala, and 40 random samples and 25 additional dampnesssamples from Tartu) (Fig. 1).

The subjects' homes were visited between March 2001 and Janu-ary 2002 in Iceland, between February 2001 and December 2001 inUppsala and between April 2001 and June 2002 in Tartu. Mosthome visits were performed during the winter months, early spring,or late autumn.

All subjects participating in this study also participated in theECRHSII Indoor study (Zock et al., 2006). We used questionnairedata and blood sample (specific IgE) data from that study. In additionwe used an additional questionnaire on SBS symptoms and five questionson dampness and indoor molds not included in the ECRHS questionnaire(Sahlberg et al., 2012). Local ethics committees at each center approvedthe study protocols.

R=460U=519T=259

R=150U=196T=164

R=60U=60T=60

R=28U=43T=40

R=42U=46T=41

R=3U=20T=25

Sample ofECRHS II

Selected

Participants

MVOC-dataavailable

Not moved

Randoma

Additionalb

R=31U=63T=65

Total participants

Fig. 1. Selections of participants. The numbers of participants are given in the boxes.R=Reykjavik, U=Uppsala, and T=Tartu. a. A random sample of subject who hadnot moved where selected. b. Subjects who had dampness in the dwelling and werenot in the random samples.

435B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

3. Measurement of indoor microbial and chemical exposure

Indoor measurements included temperature, relative humidity,airborne molds, bacteria, and the 16 MVOCs, Texanol and TXIB. Tem-perature and relative air humidity were measured with an AssmanPsychrometer type SK-RHG (Sato Keiryoki Mfg. Co., Tokyo, Japan).Airborne microorganisms were sampled on 25 mm nucleopore filters(pore size 0.4 μm) (2.0 l/min; 3 h). The nucleopore filters werestored in a refrigerator (+4 to +10 °C) after sampling, and sent bypriority mail to our department within 7 days and then immediatelytransported to Pegasus Lab AB, Uppsala, Sweden, for analysis. Thetotal concentration of airborne molds and bacteria were determinedby the CAMNEA method (Palmgren et al., 1986). Selected viablemolds and bacterial species were determined by incubation in twodifferent media, tryptone glucose agar and malt extract agar (DG18)at 22 °C (±1 °C). The incubation time was 7 days for both mediaand all microorganisms, with the exception of Streptomyces sp.,where the incubation time was 21 days. The detection limit for viableorganisms was 30 colony-forming units per m3 of air and 11,000/m3

for total bacteria and molds.Airborne MVOCs and the two plasticizers were sampled in a

charcoal tube (Anasorb 747; SKC Inc., Eighty Four, PA, USA) with apump with a flow rate of 0.4 l/min for 3 h. The tubes were desorbedwith 2 ml of methylene chloride and analyzed by selective ionmonitoring (SIM) gas chromatography–mass spectrometry (GC–MS) by Pegasus Lab AB, Uppsala, Sweden (Wessen and Schoeps,1996). The following compounds were measured: 3-methylfuran,3-methyl-1-butanol, dimethyldisulfide, 2-hexanone, 2-heptanone,1-octen-3-ol, 3-octanone, 2-metyl-1-butanol, ethyl-2-methylbutyrate,

2-pentylfuran, isobutylacetate, isobutanol, 1-butanol, 2-pentanol,ethylisobutyrate, 2-ethyl-1-hexanol, Texanol and TXIB. The detectionlimit was 1 ng/m3 for all MVOCs and 0.1 μg/m3 for 2-ethyl-1-hexanolTexanol and TXIB. The total concentration of the selected MVOC (totalMVOC) was defined as mass summation excluding the butanols and2-ethyl-1-hexanol. Indoor concentrations of formaldehyde weremeasured with glass fiber filters impregnated with 2,4-dinitro-phenylhydrazine (Andersson et al., 1981) with an air sampling flowof 0.4 l/min during 3 h. The filters were analyzed by liquid chroma-tography at the Department of Occupational and EnvironmentalMedicine, Örebro, Sweden. The detection limit for formaldehydewas 6 μg/m3. The charcoal tubes and formaldehyde filters werestored and sent by the same procedure as the nucleopore filters.

4. Assessment of symptoms and personal factors

Information on age, sex, and smoking habits was collected fromthe screening and interview questionnaire. A current smoker was de-fined as one who, during the interview, reported current smoking, orwho had ceased smoking less than a year ago.

Information on 16 symptoms compatible with SBS was obtainedfrom the self-administered questionnaire (Bjornsson et al., 1998;Sahlberg et al., 2012). The recall period was 3 months. For each symp-tom, an answer could be given according to one of three options: ‘no,never,’ ‘yes, sometimes’, and ‘yes, often’ — often meaning every week.There was one question asking whether the respondents attributedthe symptoms to their work environment, and another asking ifthey considered the symptoms to be related to the home environ-ment. This information, however, was not used in this study, whichcovers symptoms regardless of the subject's opinion on causes. Thereason was that very few subjects addressed their symptoms to thehomeenvironment. Theprevalence ofweekly symptomswas calculatedfor each symptom, as well as the prevalence of at least one weeklysymptom, and at least one mucosal, general and dermal symptom.

5. Blood samples

Blood and serum samples were collected and stored at −20 °C forthe measurement of total serum IgE, specific serum immunoglobulin E(IgE) at a central laboratory, and additional biomarkers locally. Specificserum levels against cat, timothy grass, the Cladosporium herbarium,mold and the house dust mite Dermatophagoides pteronyssinus, weredetermined by using Pharmacia CAP system (Pharmacia Diagnostics,Uppsala, Sweden). A level above 0.35 kU/Lwas considered a positive re-action. Atopywas defined as having a positive reaction to at least one ofthe allergens.

6. Data analysis

A geometric mean (GM) with a 95% confidence interval and max-imum value for MVOCs, bacteria, molds, formaldehyde, temperatureand relative air humidity, were calculated for both the total, and ran-dom sample. Association between weekly occurrence of any SBSsymptoms, or mucosal symptoms and each exposure variable wascalculated by multiple logistic regression analysis keeping age, gen-der, atopy, smoking and the particular home-exposure variable themodel. Mucosal symptoms was defined as if a subjects reported atleast one of the following symptoms; eye irritation, swollen eyelids,nasal catarrh, nasal obstruction, dryness in throat, sore throat or irri-tative cough. Each environmental variable was log-transformed be-fore the analysis, since the distributions of the measured valueswere skewed. Significant associations for any SBS, and mucosal symp-toms were analyzed, including the same independent variables. Der-mal symptoms and general symptoms were uncommon and were notevaluated separately. If the value for a compound was below the

436 B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

detection limit, half the detection limit was addressed before logtransformation.

Differences in levels of MVOCs, bacteria, molds, formaldehyde, andthe plasticizers among buildings with and without a history of damp-ness and molds, according to the participant, were analyzed by theMann Whitney U test.

Lastly, we used the Kruskal–Wallis test to see if there was any dif-ference in total MVOC levels between different centers. Chi-squareanalysis (2∗3 contingency tables) was used to see if there was anycenter difference in weekly SBS symptoms. In all statistical analyses,two-tailed tests and a 5% level of significance were used. Calculationswere performed using SAS® system version 9.2, SAS Institute Inc.,Cary, NC, USA.

7. Results

The distribution of subjects by center was 19% from Reykjavik, 40%Uppsala, and 41% from Tartu, in total 159 subjects. Of these 57% werefemales. The prevalence of any weekly SBS symptom was 19% forReykjavik, 33% for Uppsala and 29% for Tartu, which was a significantdifference. Moreover, the total MVOC level in the total sample dif-fered between the centers' geometric means (GM): 9810 ng (95% CI7800–12,340) in Reykjavik, 10,070 ng (95% CI 8370–12,120) in Upp-sala, and 8100 ng (95% CI 6680–9790) in Tartu.

The level of specific MVOC, formaldehyde, and the plasticizers isgiven in Table 1. Data on temperature, relative humidity, and formal-dehyde was available from all homes, but data on airborne molds andbacteria was missing in one dwelling due to pump failure. The levelsof temperature, relative air humidity, mold, and bacteria in the dwell-ings are presented in (Table 2).

Prevalence of weekly symptoms is shown in Table 3. The mostcommon SBS symptomwas nasal catarrh. Prevalence of asthma, aller-gies, chronic bronchitis, and smoking habits among subjects in thisstudy is shown in Table 4. The proportion of current smokers washigh, 31%.

Table 1Concentrations of Indoor MVOC and other specific compounds.

(N=159) DR (%)a GMb

MVOC (ng/m3)Sum of MVOC 12,3303-Methylfuran 99 20Isobutanol 100 15781-Butanol 100 60622-Pentanol 99 133-Methyl-1-butanol 98 269Dimethyl disulphide 99 242-Hexanone 100 612-Heptanone 100 2941-Octen-3-ol 100 523-Octanone 100 392-Methyl-1-butanol 99 73Ethyl isobutyrate 68 1Isobutyl acetate 84 48Ethyl-2-methylbutyrate 93 202-Pentylfuran 100 47

Other specific compounds (μg/m3)2-Ethyl-1-hexanol 100 2.4Formaldehyde 99 20Texanol 99 1.2TXIB 100 1.5

a DR=detection rate.b Geometric mean random sample.c Geometric mean total sample.d 95% confidence interval of geometric mean.e Maximum value of analyzed MVOC.

Associations between each exposure variable and anyweekly SBS ormucosal symptoms were analyzed by multiple linear regression analy-sis, keeping age, gender, smoking and any symptoms in themodels. Ini-tially we analyzed a logistic regression model including only theconfounders (e.g., gender, smoking and atopy). The OR was 1.04 (95%CI 0.99–1.10) for age, 1.11 (95% CI 0.58–2.14 for gender, 1.27 (95% CI0.66–2.43) for smoking and 1.21 (95% CI 0.56–2.60) for atopy.

Significant positive associations were found for six compounds(formaldehyde, 2-pentanol, 2-hexanone, 1-octen-3-ol, 2-pentylfuraneand Texanol). A significant negative association was found betweenlevel of etyl-2-metylbutyrate and any SBS symptoms. The sum ofMVOCs (total MVOC) or total or viable mold or bacteria was not signif-icantly associated with any SBS symptom (Table 5). Significant positiveassociations were found between 1-octen-3-ol, 3-methylfuran and via-ble bacteria and mucosal symptoms. A significant negative associationwas found between isobutyl acetate and mucosal symptoms (Table 5).

As a next step, we analyzed associations between the exposure vari-ables and anymucosal symptoms.We found that three compoundswerepositively associated with mucosal symptoms, and one was negativelyassociated. When comparing levels of the six significant variables be-tween centers, there were significant differences for all six compounds(Pb0.05), with lowest levels in Reykjavik and highest levels in Uppsala.

Finally, we compared levels of exposure variables between thosewith and without a history of dampness and molds (either duringthe last 10 years or the last 12 months). The levels of total bacteria,total molds, viable molds, 3-methylfurane and ethyl-isobutyratewere significantly higher in dwellings with a history of dampnessand microbial growth (Table 6).

If the dampness variable is restricted to reports of dampness ormolds during the last 12 months (current dampness or molds), only2-ethyl-1-hexanol and 2-methylfuran were significantly higher indamp buildings. Levels of 2-etyl-1-hexanol (GM with 95% CI) were(2.80; 95% CI 2.22–3.69) in damp dwellings and (1.85; 95% CI 1.40–2.44) in dry dwellings (P=0.03). Levels of 3-methylfuran (GM with95% CI) were (35 ng/m3; 95% CI 20–50) in damp dwellings and(17 ng/m3; 95% CI 11–26) in dry dwellings (P=0.03).

95% CId GMc 95% CId Maxe

9160 8160–10,270 88,00016–25 20 20–30 690

1371–1816 1560 1370–1790 18,0005214–7047 5840 5110–6660 78,000

9–16 10 9–10 220187–388 270 200–360 532017–6 30 20–40 578043–63 50 50–60 320

254–341 320 280–360 317043–3 50 40–60 233035–44 40 40–50 35054–98 70 60–90 9401–2 1 1–2 70

23–100 60 30–100 17,25013–29 30 20–40 239037–4 40 30–50 46,000

2.1–2.8 2.4 2.1–2.7 1418–23 20 18–23 2330.9–1.7 1.1 0.8–1.4 2471.2–1.8 1.5 1.3–1.7 23

Table 2Concentrations of bacteria, molds, temperature and relative air humidity.

(N=159) DR (%)a GMb 95% CId GMc 95% CId Maxe

MicroorganismsViable bacteriaf 99 225 178–285 227 299–1500 42,000Total bacteriaf 99 16,160 13,400–19,500 16,358 14,045–19,050 610,000Viable moldsf 99 209 174–250 216 186–251 20,000Total moldsf 99 12,420 11,010–14,000 12,830 11,573–14,224 110,000

Temperature (C°) 100 20.9 20.5–21.3 20.7 20.4–21.1 25Relative air humidity (%) 100 40 38–43 41 39–43 76

a DR=detection rate.b Geometric mean random sample.c Geometric mean total sample.d 95% confidence interval of geometric mean.e Maximum value of analyzed, bacteria or mold.f Colony forming unit, cfu/m3.

437B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

As a next step, correlations between the eight compounds whichwere significantly associated were analyzed. Four of the compounds(2-pentanol, 2-hexanone, 1-octen-3-ol and 2-pentylfuran) were closelycorrelated to each other (Kendal–Tau beta 0.5–0.7, give details here foreach significant correlation)while the other four compounds (formalde-hyde, ethylisobutyrate, ethyl-2-methylbutyrate and Texanol) were notsignificantly inter-correlated. Since 2-pentanol, 2-hexanone, 1-octen-3-ol and 2-pentylfuran could not be kept simultaneously in the re-gression models, four different models for partial mutual adjustmentwere created, keeping one of the closely correlated four compoundsin the models together with formaldehyde, ethylisobutyrate, ethyl-2-methylbutyrate and Texanol, as well as age, sex, atopy and smokinghabits.

Only 1-octen-3-ol remained significant, while 2-pentanol, 2-hexanone, and 2-pentylfuran formaldehyde, ethyl isobutyrate, ethyl-2-methylbutyrate and Texanol were not significantly associated with SBSin any of the models.

Finally we made further analysis, using the same four regressionmodels for partiallymutual adjustment, butwith additional adjustment

Table 3Prevalence of weekly symptoms.

Type of symptom Prevalence (%)

(n=111)a (n=159)b

Eye irritation 4.5 4.5Swollen eyelids 4.7 4.6Nasal catarrh 7.2 7.0Nasal obstruction 9.2 10.3Dryness in the throat 5.4 5.2Sore throat 1.8 1.3Irritative cough 3.6 3.8Any mucosalc (7 symptoms) 20.0 20.1Headache 6.4 5.1Tiredness 5.5 7.7Sensation of getting a cold 1.9 1.3Nausea 1.8 2.0Any generald (4 symptoms) 10.5 11.3Facial itching 0.9 3.3Facial rash 2.8 3.3Itching on the hands 2.8 2.6Rashes on the hands 0.9 1.9Eczema 6.2 5.8Any skine (5 symptoms) 6.3 7.6Any symptoms 30.4 30.8

a Prevalence of subjects with at least one symptom classified as mucosal (eye irrita-tion, swollen eyelids, nasal catarrh, nasal obstruction, dryness in the throat, sore throat,irritative cough).

b Prevalence of subjects with at least one symptom classified as general (headache,tiredness, sensation of getting a cold, nausea).

c The prevalence of subjects with at least one symptom classified as skin (facialitching, facial rash, itching on the hands, rashes on the hands, eczema).

d Random sample.e Total sample.

for reports on any dampness the last 10 years. The association for1-octen-3-ol remained significant with odds ratios, OR=1.64 (95% CI)1.00–2.75.

8. Discussion

In this cross-sectional study we found that levels of three specificMVOCs, formaldehyde and the plasticizer Texanol in dwellings werepositively associated with any SBS symptoms. Regarding mucosalsymptoms two specific MVOCs and viable bacteria were positively as-sociated. Moreover, we found that certain MVOCs, as well as airbornemolds and bacteria, were measured at higher levels in dwellings witha history of dampness and molds, as compared with dwellings with-out such reports. However, the compounds that were associatedwith SBS were not the same compounds that were associated withdampness and indoor molds. The sum of MVOCs, which is used byindoor consultant companies to evaluate the dampness status of abuilding, was neither associated with SBS symptoms nor reports ondampness or molds except for 3-methylfuran.

The lowest prevalence of any SBS symptoms was found inReykjavik, and the highest prevalence was found in Uppsala. Thesame pattern was found for total concentration of MVOCs, as well asfor specific MVOCs associated with SBS symptoms. In a previousstudy from the same centers associations between asthma, atopy,bronchial responsiveness and molds, bacteria and allergens in mat-tress dust, and differences in the indoor environment between thecenters were investigated. Total level of airborne molds was lowest inReykjavik and highest in Tartu,while indoor relative humiditywas lowestin Reykjavik and highest in Uppsala. The room temperaturewas the same

Table 4Prevalence of asthma, allergies, chronic bronchitis and smoking habits among subjects.

Prevalence (%)

(N=111)a (N=159)b

Females 58 57Asthmac 9 8Atopyd 26 24Hay fevere 37 35Chronic bronchitis 8 8Current smoker 32 31Any type of building dampnessf 23 4112 month dampness 14 20Indoor paintingg 50 56Any wall to wall carpet 39 37

a Self reported asthma (ever).b Defined as having either allergy to grass or tree pollen.c Defined as positive specific IgE, or positive skin-prick test.d Subjects reporting at least one factor regarded as dampness after 12 month or after

10 years.e 12-Month recall period.f Random sample.g Total sample.

Table 5Associations between measured exposures levels and mucosal symptom or any SBS symptom.

ORa mucosal symptom 95% CIb P-value ORa any SBS symptom 95% CIb P-value

Total bacteria 1.25 0.84–1.86 0.27 1.08 0.78–1.52 0.62Viable bacteria 1.36 1.00–1.87 0.05 0.92 0.70–1.23 0.59Total molds 1.59 0.87–2.87 0.12 1.52 0.92–2.50 0.10Viable molds 1.22 0.80–1.88 0.36 0.86 0.60–1.23 0.42Formaldehyde 1.31 0-72–2.38 0.37 1.61 1.00–2.62 0.053-Methylfuran 1.68 1.10–2.59 0.02 1.03 0.77–1.37 0.84Isobutanol 1.03 0.65–1.66 0.88 1.09 0.74–1.59 0.671-Butanol 1.22 0.75–1.98 0.41 1.07 0.73–1.58 0.722-Pentanol 0.94 0.70–1.27 0.69 1.49 1.16–1.92 0.0023-Methyl-1-butanol 1.02 0.55–1.86 0.96 0.69 0.43–1.11 0.13Dimethyl disulphide 1.15 0.93–1.42 0.19 0.90 0.76–1.07 0.222-Hexanone 1.32 0.78–2.23 0.31 2.30 1.47–3.59 0.00022-Heptanone 1.42 0.84–2.40 0.19 0.91 0.60–1.38 0.651-Octen-3-ol 1.91 1.18–3.10 0.009 1.91 1.28–2.86 0.00153-Octanone 1.17 0.55–2.44 0.68 0.75 0.40–1.42 0.382-Methyl-1-butanol 1.14 0.54–2.41 0.73 0.73 0.42–1.25 0.13Ethyl isobutyrate 1.09 0.80–1.50 0.58 0.74 0.57–0.97 0.03Isobutyl acetate 0.55 0.37–0.82 0.004 0.93 0.74–1.17 0.55Ethyl-2-methylbutyrate 1.10 0.89–1.38 0.36 0.80 0.67–0.96 0.022-Penthylfuran 1.12 0.78–1.59 0.53 1.43 1.09–1.88 0.0092-Ethyl-1-hexanol 0.97 0.25–1.82 0.92 0.85 0.52–1.38 0.52Texanol 0.93 0.71–1.23 0.63 1.23 1.00–1.52 0.05TXIB 0.80 0.51–1.25 0.32 0.92 0.66–1.27 0.62

Multiple linear regression analysis was used. All analyses were controlled for age, sex, atopy and smoking habits.a Odds ratio (OR).cb 95% confidence interval.c Odds ratio expressed as change of coefficient by 1. Odds ratios were calculated using log-transformed variables.

438 B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

in Reykjavik and Uppsala, and lower in Tartu. There were no signifi-cant differences in indoor levels of total bacteria, viable bacteria orviable molds between centers (Gunnbjornsdottir et al., 2009).

We analyzed MVOCs with GCMS and SIM mode, which is a sensi-tive and specific method. The detection rate of almost all MVOCs wasgood (93–100%) except for ethyl isobutyrate (68%) and isobutylacetat

Table 6Level of specific MVOC, formaldehyde, plasticizers, bacteria, molds, temperature, and relati

Any dampness

YES (GMb, 95% CIc) n=64

Formaldehyde 18.30 (15.44–21.68)Viable bacteria 249.90 (185.81–336.09)Total bacteria 19,156 (14,809–24,780)Viable molds 263.80 (200.40–347.28)Total molds 13,712 (11,641–16,151)3-Methylfuran 0.029 (0.022–0.038)Isobutanol 1.41 (1.15–1.75)1-Butanol 5.60 (4.54–6.90)2-Pentanol 0.012 (0.008–0.017)3-Methyl-1-butanol 0.30 (0.18–0.51)Dimethyl disulphide 0.046 (0.027–0.079)2-Hexanone 0.053 (0.043–0.066)2-Heptanone 0.33 (0.28–0.40)1-Octen-3-ol 0.06 (0.046–0.078)3-Octanone 0.041 (0.035–0.049)2-Methyl-1-butanol 0.08 (0.052–0.12)Ethyl isobutyrate 0.002 (0.0016–0.0033)Isobutyl acetate 0.061 (0.021–0.17)Ethyl-2-methylbutyrate 0.04 (0.025–0.064)2-Penthylfuran 0.044 (0.031–0.061)2-Ethyl-1-hexanol 2.31 (1.87–2.84)Texanol 1.0 (0.69–1.45)TXIB 1.37 (1.08–1.74)Sum of MVOC 1.07 (0.86–1.34)Temperature (C°) 20.20 (19.65–20.76)Relative humidity (%) 43.16 (40.24–46.29)

a Any dampness or molds last 12 months or last 10 years.b Geometric mean.c 95% confidence interval of geometric mean.

(84%). The concentrations of indoor MVOCs in the dwellings inthis study were a bit lower compared with other previous studies(Araki et al., 2010; Elke et al., 1999). The study by Elke et al.(1999) reported higher levels of MVOC. Median values of measuredMVOC concentrations range from 1 ng/m3 to 5 μg/m3 in residentialindoor air.

ve air humidity in relation to presence or absence of dampness in the home.a

NO (GMb, 95% CIc)n=92 P-value

21.27 (18.18–24.89) 0.11212.15 (166.27–270.71) 0.1314,605 (12,123–17,595) 0.02186.58 (158.36–219.84) 0.0212,232 (10,695–13,992) 0.040.018 (0.015–0.024) 0.0081.69 (1.42–2.01) 0.1786.02 (5.06–7.17) 0.72

0.011 (0.008–0.015) 0.660.26 (0.017–0.037) 0.07

0.025 (0.017–0.040) 0.180.059 (0.050–0.070) 0.730.31 (0.26–0.28) 0.33

0.046 (0.038–0.057) 0.070.039 (0.034–0.044) 0.360.067 (0.050–0.089) 0.05

0.0013 (0.0010–0.0016) 0.020.053 (0.026–0.11) 0.530.022 (0.014–0.034) 0.050.041 (0.029–0.056) 0.452.43 (2.12–2.78) 0.621.12 (0.80–1.57) 0.711.55 (1.25–1.93) 0.820.86 (0.45–0.80) 0.08

21.14 (20.73–21.55) 0.0839.67 (37.18–42.34) 0.008

439B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

We found significant positive associations between SBS and fiveMVOC compounds, 2-pentanol, 2-hexanon, 1-octen-3-ol, 2-pentylfuranand 3-methylfuran. One previous epidemiological study suggested anassociation between low levels of MVOCs in dwellings and non-specifichealth symptoms, but the associations were not significant (Elke et al.,1999). However, a recent MVOC study from Japan, using GCMS in SIMmode, showed that SBS symptoms were associated with levels of2-pentanol and 1-octen-3-ol in dwellings. In particular, levels of1-octen-3-ol were associated with the prevalence of home-related mu-cous membrane symptoms. The other MVOCs had no consistent rela-tionship to symptoms (Araki et al., 2010). We also found that1-octen-3ol and 3-methylfuranwas associatedwithmucosal symptoms.In addition, association between viable molds and mucosal symptomswas found. This indicates that molds can cause illness when people areexposed to mold growth indoors.

The arithmetic mean of 1-octen-3-ol was 0.09 μg/m3, with a maxi-mum value of 2.33 μg/m3 in our study. This is lower than the levelsreported from Japanese homes, which had a geometric mean of0.19 μg/m3 and a maximum value of 8.58 μg/m3 (Araki et al., 2010).One experimental exposure chamber study showed that 1-octen-3-ol,at much higher levels than found in our study, can cause an increasein inflammatory biomarkers in nasal lavage and an increase of mucosaland general symptoms (Wålinder et al., 2005). Associations betweenSBS and levels of 1-octen-3-ol and 2-pentylfuran have not been demon-strated before, but an association between these two compounds inschools and nocturnal attacks of breathlessness in the pupils have pre-viously been reported (Kim et al., 2007).

We found that total bacteria, total molds, viable molds, and threechemical compounds (2-ethyl-1-hexanol, 2-methylfuran and ethylisobutyrate) were associated with a history of building dampnessand indoor molds. The associations were stronger if we includedretrospective information about dampness and molds during thelast 10 years, as compared with reports on dampness and molds thelast 12 months. This shows the importance of asking retrospectivelyfor the history of dampness and molds over several years indwellings, not just last year, if the exposure is to be assessed in thebest way.

Texanol is a common plasticizer in water based paints and someplastic material. We found an association between indoor levels ofTexanol and prevalence of any SBS symptoms. We found no otherstudies on associations between Texanol in dwellings and SBS. How-ever, one school study found associations between the level ofTexanol in schools and nocturnal attacks of breathlessness (Kim etal., 2007). Moreover, in an exposure chamber study, experimental ex-posure to Texanol, at much higher levels than in our dwellings, wasdemonstrated to cause mild irritation in eyes, nose, and airways(Ernstgard et al., 2007).

In addition we found an association between the indoor level offormaldehyde and prevalence of any SBS symptom. The geometricmean level was 20 μg/m3, with a maximum level of 233 μg/m3. Form-aldehyde has been measured in dwellings throughout Europe, mostlyin the Nordic countries and also in southern Europe. Airborne levelsof formaldehyde currently present in dwellings in various nationsgenerally range between 7.3 and 50.4 μg/m3, although concentrationsexceeding 120 μg/m3 have sometimes been observed (Lovreglio et al.,2009). A recent study measured the formaldehyde concentrations inresidential environments in 12 European cities, demonstrating atotal mean level of 23.8 μg/m3 across Europe, which are similar levelsto those in our study (Bruinen de Bruin et al., 2008). Studies fromChina have measured formaldehyde in dwellings. The formaldehydeconcentrations in homes in Hong Kong ranged from 46 μg/m3 to1425 μg/m3 (Guo et al., 2009). The median indoor level of formalde-hyde in six Hong Kong dwellings was 16.1 μg/m3 in the warm seasonand 85.7 μg/m3 in the cold season (Lee et al., 2002). We found fewprevious studies on associations between formaldehyde levels inhomes and SBS. One study of newly built houses in Japan found a

dose-dependent association between indoor levels of formaldehydeand SBS symptoms (Takigawa et al., 2010).

We found higher levels of molds and bacteria on the one hand,and 3-methylfuran, ethyl-isobutyrate on the other in dwellings witha history of dampness and molds. Moreover, levels of 3-methylfuranand 2-ethyl-1-hexanol were higher in dwellings with current reporton dampness and molds. However, the difference in geometricmean values were small; typically, damp dwellings had 10–60%higher indoor concentrations than dry dwellings. Some previousstudies have reported about two times higher concentrations inmold dwellings as compared with dry dwellings (Verhoeff et al.,1992), but other studies have failed to find any association betweenindoor viable molds in homes and moldy odor or visible dampness(Holme et al., 2010). One hospital study reported that total levels ofMVOCs were higher in a flooded building as compared with a normalbuilding (149 ng/m3 vs. 94 ng/m3). Some individual MVOCs, forinstance 1-octen-3-ol and 2-pentanol, were found in higher concentra-tions in damp buildings (Wieslander et al., 2007). One study on 40dwellings with evident mold damage and 44 where mold damage wasexcluded after thorough investigation found that, 2-methyl-butanoland 1-octen-3-ol were significantly associated with dampness statusbut only 10% of theMVOCvariabilitywas explained bydampness status.Levels of 2-methylfuran and 3-methylfuran were associated with envi-ronmental tobacco smoke, and these compounds were also associatedwith humidity and air exchange rate (Schleibinger et al., 2008).

9. Conclusion

The indoor levels of 2-pentanol, 2-hexanon, 1-octen-3-ol, and2-pentylfuran were positively associated with any SBS symptoms. Fur-thermore, 3-methylfuranwas positively associated withmucosal symp-toms. However, since none of these compounds, except 3-methylfuran,were associated with reports on dampness or molds, we found little ev-idence that they were emitted frommicrobial growth or damp buildingmaterials. On the other hand, 3-methylfuran has been shown to beformed by a broad spectrum of fungi (Borjesson et al., 1992) and hasalso been used as a marker for the active growth of microorganisms inwater-damaged buildings (Wessen and Schoeps, 1996). 3-Methylfuranis also related to symptoms of airway obstruction (Smedje et al., 1996).

Moreover, formaldehyde and the plasticizer Texanolmay be a risk fac-tor for sick building syndrome. Indoor concentration of airborne molds,bacteria, 2-ethyl-1-hexanol, 3-methylfuran, and ethyl isobutyrate wereassociated with reports on dampness and molds, which illustrates asignificant but weak correlation between questionnaire data ondampness and molds in dwellings and measured levels of somemicrobial pollutants.

References

Andersson K, Hallgren C, Levin JO, Nilsson CA. Chemosorption sampling and analysis offormaldehyde in air. Influence on recovery during the simultaneous sampling offormaldehyde, phenol, furfural and furfuryl alcohol. Scand J Work Environ Health1981;7(4):282–9. [Dec].

Apter A, Bracker A, Hodgson M, Sidman J, Leung WY. Epidemiology of the sick buildingsyndrome. J Allergy Clin Immunol 1994;94:277–88.

Araki A, Kawai T, Eitaki Y, Kanazawa A, Morimoto K, Nakayama K, et al. Relationshipbetween selected indoor volatile organic compounds, so-called microbial VOC,and the prevalence of mucous membrane symptoms in single family homes. SciTotal Environ 2010;408(10):2208–15. [Apr 15, Epub 2010 Feb 25. 2010].

Bjornsson E, Janson C, Norback D, Boman G. Symptoms related to the sick building syn-drome in a general population sample: associations with atopy, bronchial hyper-responsiveness and anxiety. Int J Tuberc Lung Dis 1998;2:1023–8.

Borjesson T, Stollman U, Schnurer J. Volatile metabolites produced by six fungal speciescompared with other indicators of fungal growth on cereal grains. Appl EnvironMicrobiol 1992;58(8):2599–605. [Aug].

Bornehag CG, Blomquist G, Gyntelberg F, Jarvholm B, Malmberg P, Nordvall L, et al.Dampness in buildings and health. Nordic interdisciplinary review of the scientificevidence on associations between exposure to “dampness” in buildings and healtheffects (NORDDAMP). . [see comment]Indoor Air 2001;11:72–86.

440 B. Sahlberg et al. / Science of the Total Environment 444 (2013) 433–440

Bornehag CG, Lundgren B, Weschler CJ, Sigsgaard T, Hagerhed-Engman L, Sundell J.Phthalates in indoor dust and their association with building characteristics. EnvironHealth Perspect 2005;113(10):1399–404. [Oct].

Bornehag CG, Sundell J, Weschler CJ, Sigsgaard T, Lundgren B, Hasselgren M, et al. Theassociation between asthma and allergic symptoms in children and phthalates inhouse dust: a nested case–control study. Environ Health Perspect 2004;112(14):1393–7. [Oct].

Bruinen de Bruin Y, Koistinen K, Kephalopoulos S, Geiss O, Tirendi S, Kotzias D. Charac-terisation of urban inhalation exposures to benzene, formaldehyde and acetalde-hyde in the European Union: comparison of measured and modelled exposuredata. Environ Sci Pollut Res Int 2008;15(5):417–30. [Jul, Epub 2008 May 20. 2008].

Brunekreef B. Damp housing and adult respiratory symptoms. Allergy 1992;47:498–502.Burney PG, Luczynska C, Chinn S, Jarvis D. The European Community Respiratory

Health Survey. Eur Respir J 1994;7:954–60.Claeson AS, Levin JO, Blomquist G, Sunesson AL. Volatile metabolites from microorgan-

isms grown on humid building materials and synthetic media. J Environ Monit2002;4(5):667–72. [Oct].

Elke K, Begerow J, OppermannH, Kramer U, Jermann E, Dunemann L. Determination of se-lected microbial volatile organic compounds by diffusive sampling and dual-columncapillary GC-FID—a new feasible approach for the detection of an exposure toindoor mould fungi? J Environ Monit 1999;1(5):445–52. [Oct].

Engvall K, Norrby C, Norback D. Sick building syndrome in relation to building damp-ness in multi-family residential buildings in Stockholm. Int Arch Occup EnvironHealth 2001;74:270–8.

Ernstgard L, Lof A, Wieslander G, Norback D, Johanson G. Acute effects of some volatileorganic compounds emitted from water-based paints. J Occup Environ Med2007;49(8):880–9. [Aug].

Fisk WJ, Lei-Gomez Q, Mendell MJ. Meta-analyses of the associations of respiratoryhealth effects with dampness and mold in homes. Indoor Air 2007;17:284–96.

Gunnbjornsdottir MI, Norback D, Bjornsson E, Soon A, Jarvis D, Jogi R, et al. Indoor en-vironment in three North European cities in relationship to atopy and respiratorysymptoms. Clin Respir J 2009;3(2):85–94. [Apr].

Guo H, Kwok NH, Cheng HR, Lee SC, Hung WT, Li YS. Formaldehyde and volatile organiccompounds in Hong Kong homes: concentrations and impact factors. Indoor Air2009;19(3):206–17. http://dx.doi.org/10.1111/j.1600-0668.2008.00580.x. [Electronicpublication ahead of print 2009 Feb 7].

Hodgson M. The sick-building syndrome. Occup Med 1995;10:167–75.Holme J, Hägerhed-Engman L, Mattsson J, Sundell J, Bornehag CG. Culturable mold in in-

door air and its association with moisture-related problems and asthma and allergyamong Swedish children. Indoor Air 2010;20(4):329–40. http://dx.doi.org/10.1111/j.1600-0668.2010.00658.x. [Electronic publication ahead of print 2010 Apr 16].

Hsu NY, Lee CC, Wang JY, Li YC, Chang HW, Chen CY, et al. Predicted risk of childhood al-lergy, asthma, and reported symptoms using measured phthalate exposure in dustand urine. Indoor Air 2011. http://dx.doi.org/10.1111/j.1600-0668.2011.00753.x.[Oct. 13].

Husman T. Health effects of indoor-air microorganisms. [see comment]Scand J WorkEnviron Health 1996;22:5-13.

Janson C, Anto J, Burney P, Chinn S, de Marco R, Heinrich J, et al. The European Commu-nity Respiratory Health Survey: what are the main results so far? European Com-munity Respiratory Health Survey II. Eur Respir J 2001;18(3):598–611. [Sep].

Kim JL, Elfman L, Mi Y, Wieslander G, Smedje G, Norback D. Indoor molds, bacteria, mi-crobial volatile organic compounds and plasticizers in schools—associations withasthma and respiratory symptoms in pupils. Indoor Air 2007;17(2):153–63. [Apr].

Korpi A, Kasanen JP, Alarie Y, Kosma VM, Pasanen AL. Sensory irritating potency ofsome microbial volatile organic compounds (MVOCs) and a mixture of fiveMVOCs. Arch Environ Health 1999;54(5):347–52. [Sep–Oct].

Korpi A, Pasanen AL, Pasanen P. Volatile compounds originating from mixed microbialcultures on building materials under various humidity conditions. Appl EnvironMicrobiol 1998;64(8):2914–9. [Aug].

Lee SC, Li W-M, Ao C-H. Investigation of indoor air quality at residential homes inHong Kong—case study. Atmos Environ 2002;36:225–37.

Lovreglio P, Carrus A, Iavicoli S, Drago I, Persechino B, Soleo L. Indoor formaldehydeand acetaldehyde levels in the province of Bari, South Italy, and estimated healthrisk. J Environ Monit 2009;11(5):955–61. [May, Epub 2009 Feb 13. 2009].

Mendell MJ. Non-specific symptoms in office workers: a review and summary of theepidemiological literature. Indoor Air 1993:3.

Naydenov K, Popov T, Mustakov T, Melikov A, Bornehag CG, Sundell J. The associationof pet keeping at home with symptoms in airways, nose and skin among Bulgarianchildren. Pediatr Allergy Immunol 2008;19(8):702–8. [Dec, Epub 2008 Feb 11.2008].

Norback D. An update on sick building syndrome. Curr Opin Allergy Clin Immunol2009;9:55–9.

Norback D, Edling C. Environmental, occupational, and personal factors related to theprevalence of sick building syndrome in the general population. Br J Ind Med1991;48:451–62.

Norback D, Wieslander G, Nordstrom K, Walinder R. Asthma symptoms in relation tomeasured building dampness in upper concrete floor construction, and 2-ethyl-1-hexanol in indoor air. Int J Tuberc Lung Dis 2000;4(11):1016–25. Nov.

Palmgren U, Strom G, Blomquist G, Malmberg P. Collection of airborne micro-organismson Nuclepore filters, estimation and analysis—CAMNEA method. J Appl Bacteriol1986;61(5):401–6. [Nov].

Peat JK, Dickerson J, Li J. Effects of damp and mould in the home on respiratory health:a review of the literature. Allergy 1998;53:120–8.

Sahlberg B, Norback D, Wieslander G, Gislason T, Janson C. Onset of mucosal, dermaland general symptoms in relation to biomarkers and exposures in the dwelling.A cohort study from 1992 to 2002. Indoor Air 2012. http://dx.doi.org/10.1111/j.1600-0668.2012.00766.x. [Jan. 19].

Sarigiannis DA, Karakitsios SP, Gotti A, Liakos IL, Katsoyiannis A. Exposure to major vol-atile organic compounds and carbonyls in European indoor environments and as-sociated health risk. Environ Int 2011;37(4):743–65. [May, Epub 2011 Feb 26].

Schleibinger H, Laussmann D, Bornehag CG, Eis D, Rueden H. Microbial volatile organiccompounds in the air of moldy and mold-free indoor environments. Indoor Air2008;18(2):113–24. [Apr].

Smedje GND, Wessen B, Edling C. Asthma among school employees in relation to theschool environment. Indoor Air '96, the 7th International Conference on IndoorAir Quality and Climate, 21–26 July 1996. Nagoya, Japan. Tokyo: Institute of PublicHealth; 1996. p. 611–6.

Sunesson AL, Nilsson CA, Andersson B, Blomquist G. Volatile metabolites produced bytwo fungal species cultivated on building materials. Ann Occup Hyg 1996;40(4):397–410. [Aug].

Takigawa T, Wang BL, Saijo Y, Morimoto K, Nakayama K, Tanaka M, Shibata E,Yoshimura T, Chikara H, Ogino K, Kishi R. Relationship between indoor chemicalconcentrations and subjective symptoms associated with sick building syndromein newly built houses in Japan. Int Arch Occup Environ Health 2010;83(2):225–35. http://dx.doi.org/10.1007/s00420-009-0475-9. [Feb].

Verhoeff AP, van Wijnen JH, Brunekreef B, Fischer P, van Reenen-Hoekstra ES, SamsonRA. Presence of viable mould propagules in indoor air in relation to house dampand outdoor air. Allergy 1992;47(2 Pt 1):83–91. [Apr].

Wålinder R, Ernstgård L, Johanson G, Norbäck D, Venge P,Wieslander G. Acute effects of afungal volatile compound. Environ Health Perspect 2005;113(12):1775–8. [Dec].

Wessen B, Schoeps KO. Microbial volatile organic compounds—what substances can befound in sick buildings? Analyst 1996;121(9):1203–5. [Sep].

WHO. WHO Guidelines for Indoor Air Quality: Dampness and Mould; 2009.Wieslander G, Norback D, Venge P. Changes of symptoms, tear film stability and eosin-

ophilic cationic protein in nasal lavage fluid after re-exposure to a damp officebuilding with a history of flooding. Indoor Air 2007;17(1):19–27. [Feb].

Wilkins K, Larsen K, Simkus M. Volatile metabolites from mold growth on building ma-terials and synthetic media. Chemosphere 2000;41(3):437–46. [Aug].

Zock JP, Heinrich J, Jarvis D, Verlato G, Norback D, Plana E, et al. Distribution and deter-minants of house dust mite allergens in Europe: the European Community Respi-ratory Health Survey II. J Allergy Clin Immunol 2006;118(3):682–90. Sep, Epub2006 Jul 27.