Risk factors for asthma, allergy and eczema were studied in a stratified random sample of adults in Stockholm. In 2005, 472 multifamily buildings (10,506 dwellings) were invited (one subject/dwelling) and 7,554 participated (73%). Associations were analyzed by multiple logistic regression, adjusting for gender, age, smoking, country of birth, income and years in the dwelling. In total, 11% had doctor's diagnosed asthma, 22% doctor's diagnosed allergy, 23% pollen allergy and 23% eczema. Doctor's diagnosed asthma was more common in dwellings with humid air (OR = 1.74) and mould odour (OR = 1.79). Doctor's diagnosed allergy was more common in buildings with supply exhaust air ventilation as compared to exhaust air only (OR = 1.45) and was associated with redecoration (OR = 1.48) and mould odour (OR = 2.35). Pollen allergy was less common in buildings using more energy for heating (OR = 0.75) and was associated with humid air (OR = 1.76) and mould odour (OR = 2.36). Eczema was more common in larger buildings (OR 1.07) and less common in buildings using more energy for heating (OR = 0.85) and was associated with water damage (OR = 1.47), humid air (OR = 1.73) and mould odour (OR = 2.01). Doctor's diagnosed allergy was less common in buildings with management accessibility both in the neighbourhood and in larger administrative divisions, as compared to management in the neighbourhood only (OR = 0.49; 95% CI 0.29–0.82). Pollen allergy was less common if the building maintenance was outsourced (OR = 0.67; 95% CI 0.51–0.88). Eczema was more common when management accessibility was only at the division level (OR = 1.49; 95% CI 1.06–2.11). In conclusions, asthma, allergy or eczema were more common in buildings using less energy for heating, in larger buildings and in dwellings with redecorations, mould odour, dampness and humid air. There is a need to reduce indoor chemical emissions and to control dampness. Energy saving may have consequences for allergy and eczema. More epidemiological studies are needed on building management organization.
Citation: Norbäck D, Lampa E, Engvall K (2014) Asthma, Allergy and Eczema among Adults in Multifamily Houses in Stockholm (3-HE Study) - Associations with Building Characteristics, Home Environment and Energy Use for Heating. PLoS ONE 9(12): e112960. https://doi.org/10.1371/journal.pone.0112960
Editor: Maosheng Yao, Peking University, China
Received: June 19, 2014; Accepted: October 16, 2014; Published: December 5, 2014
Copyright: © 2014 Norbäck et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper.
Funding: The study was supported by grants from The Swedish Asthma and Allergy Association's Research Foundation (contract 22012041-K), The Swedish Research Council FORMAS (contracts 294-2008-80 and 244-2010-277), The Swedish Property Foundation (FORMAS contract 244-2008-80), The Property Federation in Stockholm (FORMAS contract 244-2008-80), The City of Stockholm (FORMAS contract 244-2008-80) and the Fortum corporation (FORMAS contract 244-2008-80). The contract 294-2008-80 with FORMAS included requested co-sponsorship by the organizations or companies mentioned above. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Asthma and allergies have been on the rise in many countries over the past few decades and indoor exposure has been suggested as a possible cause. The dwelling is where we spend most of our time. Studies on indoor risk factors for asthma and allergies in the home environment among adults have focused mainly on building dampness and indoor mould  and environmental tobacco smoke (ETS) –. Another sources of indoor microbial contamination in homes, associated with adult asthma, includes storage of household waste for several days . Chemical indoor factors at home associated with adult asthma include formaldehyde , , volatile organic compounds (VOC)  and emissions from indoor paint  and household cleaning sprays , leakage of gasoline emissions from garages , plastic wall materials and recent use of floor levelling plaster . Moreover, use of air cleaners and air humidifiers at home has been associated with adult asthma . Few studies are available on the association between ventilation flow in dwellings and asthma, but one study found a positive association between indoor levels of carbon dioxide (CO2), a proxy variable for ventilation flow, and asthmatic symptoms . Another study found associations between wheeze and window pane condensation in winter, a proxy variable for high air humidity and low ventilation flow . Other domestic risk factors for adult asthma include indoor allergens from cats, dogs, mice, house dust mites (HDM) and cockroaches . Moreover, indoor fuel combustion is associated with asthma, especially in developing countries . A few studies have investigated the associations between indoor factors at home and pollen allergy (hay fever) or furry pet allergy in adults. One found an association between dampness/indoor moulds at home and pollen allergy  and another found positive associations between recent indoor painting and furry pet allergy . As for eczema and dermal symptoms in adults, domestic factors include dampness/indoor mould – and storage of organic household waste .
Over the past decade there has been growing concern regarding energy consumption in buildings in many countries, as a part of “Low Carbon Cities” programmes  striving for reduced greenhouse-gas emissions . A recent review concluded that improvements in the physical fabric of the house, such as improved insulation, can lead to health improvements . One intervention study showed that green renovation of dwellings could reduced energy consumption by 45% and improved the respiratory health in children . Another intervention study found lower levels of asthma symptoms among asthmatic children living in homes that achieved better energy efficiency through retrofitted insulation and more effective heating . Another study reported lower levels of house dust mite allergens in buildings designed for low energy use, as compared to normal buildings . However, some studies indicate negative health effects of energy saving in dwellings. The potential health problems surrounding energy saving were addressed as early as in 1994 . In the former East Germany, energy saving caused health disturbances due to new tight windows and doors  and increased concentrations of house dust mite allergens and mould spores . One intervention study found that among asthmatic adults, some asthmatic symptoms were aggravated following energy retrofit work . Another recent study concluded that tightening the building envelope in homes could lead to 20% more childhood asthma events, provided no adequate improvement of ventilation .
In Sweden, energy conservation issues were first addressed in the building code of 1975, and the energy use in residential buildings has been constant for many decades. The country's gross heated area increased by 40% from the 70s to 90s  after which it has increased further, indicating gradual increased energy efficiency for building heating. In 2010, the total annual energy consumption in Sweden was 395 TWh with residential buildings and commercial premises using 85 TWh for space heating and domestic water production. Of this, 42% was used in detached houses, 32% in multi-family buildings (flats) and 26% in commercial premises. In multifamily buildings, distance heating was the main source of building heating (91% of energy use); electric heating (4%) and oil combustion (1%) were rare. In detached houses, 37% used electric heating, 36% burnt wood/wood chips, 18% used distance heating and only 3% used oil. A full 46% of the detached houses have a heat pump installed. The hot water used for distance heating is produced mainly through the combustion in large units of renewable fuels such as wood chips (38%) garbage (19%) and peat (4%) in the cities and to a lesser extent by heat pumps (3%) . Despite increased energy efficiency in Swedish buildings over past decades, we found no evaluation of the health consequences of this energy saving. Moreover, there are few large population studies exist where the associations between building characteristics, building management, indoor characteristics and energy consumption on the one hand and adult asthma and allergies on the other have been studied using mutual adjustment for various domestic environment factors, with additional adjustment for socioeconomic status.
The aim was to study associations between doctor's diagnosed asthma, doctor's diagnosed allergy, self-reported pollen allergy and self-reported eczema and selected building characteristics, building management routines and flat indoor characteristics. The hypothesis to be tested was whether the four health variables were associated with the following building data: building construction year, building size (no. of residents), type of ventilation, type of heat distribution, energy use for heating, type of ownership, type of maintenance management and management accessibility. Moreover, whether they were associated with the following flat/family data; crowdedness (no. of persons per room), family income, years living in the current dwelling, airing habits (airing index), recent redecoration, mouldy odour, water/dampness damage and indications of humid indoor air.
Material and Methods
Study design and study population
The study procedure and study protocol were approved by the Regional Ethical Review Board in Uppsala, Sweden. All participants gave their informed consent. An information letter sent together with the questionnaire stated that if the subjects answered and returned the questionnaire it meant they had given their informed consent.
In February-April 2005, 501 of a total of 14,465 (3.5%) multifamily buildings in Stockholm were selected from the central building register by stratified random sampling. Single-family houses (detached houses) and semi-detached houses were excluded, as well as 20 multifamily buildings used as nursing homes for the elderly. Data on building age and ownership was obtained from the building register. Strata were based on the construction year of the building classified into six classes (before 1961, 1961–75, 1976–84, 1985–1990, 1991–1997 and 1998–2003). The sampled buildings were selected from buildings with the registration code “exclusively multifamily buildings” which means that the buildings only contained flats. To obtain enough buildings in each class, the sampling proportion was lower for older buildings and higher for newer buildings. All dwellings in the selected 481 buildings were identified. The sampling proportions of selected flats were based on building size. In larger buildings, with more than 30 dwellings (N = 63), 30 dwellings were randomly selected into the study. In buildings up to 30 dwellings (N = 418) all dwellings were selected. In each selected dwelling, one randomly selected person (>18 years) who had lived in the flat for more than 6 months, was chosen by combining information in the building register with the civil registration register.
A self-administered postal questionnaire, the Stockholm Indoor Environment Questionnaire (SIEQ) , was sent to the selected subjects (N = 11,160 with the exception of 654 who did not actually live at the address. Responses came from 7,640 of 10,506 with correct addresses (73%). There were differences in response rate between the age classes of buildings, with the lowest response rate (66%) for buildings built in 1961–1975, and the highest (79%) for the newest buildings (1998–2003). Moreover, the response rate was lower in rented buildings (63%) as compared to buildings owned by the inhabitants (73%). A separate questionnaire was sent to building owners in April-May 2005, including questions about the building construction and maintenance organization. This information was received from 7,554 participants in dwellings in 472 buildings (98%).
Assessment of building data
Data on building construction year and type of ownership was obtained from the building register. The questionnaire to the building owners contained questions on the building, including number of floors and number of flats, type of ventilation system and type of heating system. By combining the building register with the civil registration register, the number of residents in each building was obtained. Number of residents was divided by number of apartments to get the mean number of residents per square metre (persons/m2) for heated areas of the building. Heated area (m2, Atemp) was defined as areas with more than +10 °C room temperature in the building. In addition, the building owner questionnaire contained questions on management, operation and maintenance of the building. One question asked if management was by contract with the building owner or by an external company, or both. The other question was whether management was accessible for tenants at the division level or local level or both.
Assessment of energy use for heating
In Sweden, it is compulsory for the building owners to provide information to the Government on energy consumption for their building. Information on the energy use for heating (kWh/m2, Atemp) was obtained mainly from this energy declaration form, available in the database Gripen at the National Board of Housing, Building and Planning . Energy use for heating was calculated for heated areas of the building (m2, Atemp), defined as areas with more than +10 °C room temperature, and is the value used as the standard for energy declarations in Sweden. This roughly corresponds to the area inside the outer walls. Energy used for domestic hot water or household electricity is not included in energy use for heating. The energy declaration forms were mainly collected in 2008–2009. Measured data on electricity consumption and energy for heating for the buildings for March 2004-March 2005 was also gathered from the utility company Fortum Corporation. Data on energy uses for heating was controlled by certified energy consultants in the project group, recalculated if needed, and finally added into a new database. Special attention was given to make sure that energy measurements were for the individual building and not for the whole property, since a property can consist of several buildings. The methods for quality assurance of the energy data for these building has reported previously .
Assessment of flat data
Information on the flat was obtained from the postal questionnaire, which included questions on size of the dwelling, number of rooms, environmental tobacco smoke at home (ETS), furry pet keeping, redecoration in the last 12 months, and water leakage, damage from damp, window pane condensation, slowly drying towels and airing habits (how much and how often windows were opened).
Assessment of demographic and medical data
Information on medical symptoms and personal factors was obtained from the postal questionnaire, including age, gender, smoking habits, residency duration and airing and cleaning habits. There were four questions on asthma, allergy and eczema: Have you ever received a diagnosis of asthma from a medical doctor? Have you ever received an allergy diagnosis from a medical doctor? Have you ever had an allergy to pollen (hay fever)? Have you ever had eczema? Socioeconomic data such as income, citizenship, native country and year of immigration to Sweden, if not borne in Sweden, was obtained for each individual from Statistics, Sweden.
The number of independent variables that could be included in the statistical models was chosen according to the rule of thumb for a binary outcome . To assess whether the set of included variables could be reduced we performed a hierarchical cluster analysis based on squared Spearman correlations as strongly correlated variables could create multicollinearity problems. Squared correlations were used since we were only interested in the magnitude of the correlations and not in whether they were positively or negatively correlated. From each cluster of variables we either created a summary score from the variables or chose one variable to represent the cluster. Window pane condensation and slow drying of towels in the bathroom were combined to one variable (1 = yes on any of the two variables, or else 0), and moisture damage and water damage during the past 5 years were combined into another variable (1 = yes on any of the two variables, or else 0).
Multiple logistic regression was used to relate the independent variables to the four dependent health variables. The standard errors were corrected for the clustering of individuals within buildings. One statistical model was used for each health variable, including all 14 selected building/flat variables, with additional adjustment for six potential confounders (gender, age, current smoker, passive smoking in the dwelling, annual income per family, foreign born yes/no), and number of years living in the current flat). One categorical variable was used combining active and passive smoking (coded zero if non-smoker with no ETS, coded 1 if non-smoker with ETS and coded 2 if current smoker). Continuous variables were modelled using restricted cubic splines with 3 knots. The non- linear terms were then either kept or removed based on a test of all non-linear terms equal to zero versus at least one non-linear term not equal to zero. Since there was some missing data on each independent variable, the main analysis was performed using multiple imputation. Imputations were done using the method of chained equations and 20 imputed data sets were created. Building variables were used in the imputation equations for the individual variables but not vice-versa. This was done to ensure that imputed building variables would be constant within buildings. The results were then compared with a complete case analysis. All analyses were made using R and STATA.
Among the participants, the proportion of females was 58%, the majority was 18–44 years old, and one-fifth was foreign-born and 15% were current smokers. In total, 11% had doctors' diagnosed asthma, 22% doctor's diagnosed allergy, 23% pollen allergy and 23% eczema (Table 1). Due to the stratified sample strategy, the number of buildings was almost equally distributed between the age classes in the stratified sample, with the highest proportion (25%) of the newest buildings (1998–2003). However, when weighing this against the total building stock in Stockholm, older buildings (built before 1976) were dominant and newer buildings (built after 1975) represented only 15% of all buildings in Stockholm (Table 2). In the sample, 52% of the buildings were owned by the inhabitant, 29% were publicly owned and 19% had a private landlord. Buildings owned by the inhabitants were more common in the newest buildings from 1998–2003, while publicly owned buildings were more common in buildings built in 1961–1975 and 1985–1990. Almost all buildings had district heating and were heated by water-borne radiators and the majority consumed 56–90 kWh/m2 and year for heating (Atemp). Nearly half of the building owners carried out their own management and a few had mixed work management. About half of the inhabitants had management accessibility in the neighbourhoods; a few had accessibility both in the neighbourhood and at the division level (Table 2). Data on family and flat level is reported in Table 3. The median number of persons per room was 0.67 (IQR range 0.5–1) and about half of the participants had lived more than five years in the current flat. In total, 30% had redecorated their flat during the past year, 7.8% reported damage from either moisture or water during the last five years and 9.2% reported humid air (either window condensation or long time to dry towels). Reports on mould odour were relatively uncommon (4.6%), and both pet keeping and environmental tobacco smoke at home (ETS) was relatively uncommon. Cats and dogs were the most common pets (Table 3).
Associations between building/home environmental factors and the four health variables were analysed by weighted multiple logistic regression with imputation. One statistical model was used for each health variable, including all 15 selected building/flat variables, with additional adjustment for six potential confounders (gender, age, current smoker, annual income per family, foreign born yes/no, and number of years living in the current flat). When evaluating associations between the six confounders and four health variables, females hade more eczema (OR = 1.47; 95% CI 1.08–2.00; p = 0.02), doctor's diagnosed asthma (OR = 1.72; 95% CI 1.07–2.78; p = 0.03) and doctor's diagnosed allergy (OR = 1.77; 95% CI 1.20–2.60; p = 0.004). Younger subjects had more eczema (OR = 0.99; 95% CI 0.98–1.00 per year; p = 0.05) and hay fever (OR = 0.99; 95% CI 0.98–1.00 per year; p<0.001). There was a positive association between family income and eczema only (OR = 1.07; 96% CI 1.00–1.14 per 1000 SEK/Year; p = 0.04). Moreover, there was a tendency for foreign-born subjects to have less eczema (OR = 0.71; 95% CI 0.49–1.03; p = 0.07). There were no associations between number of years in the current dwelling and any of the four health variables.
Reports on humid air (p = 0.01), mouldy odour (p = 0.05) and type of management accessibility (p = 0.03) were significantly associated with doctor's diagnosed asthma, with a tendency of more asthma when management accessibility was only at the division level and less asthma if the accessibility was both in the neighbourhood and at the division level (Table 4). Building construction year, number of flats in the building, type of ventilation, energy use for space heating, type of ownership, type of maintenance management, environmental tobacco smoke (ETS), number of persons/room, airing habits, recent redecoration, damage from dampness or furry pet keeping were not associated with doctor's diagnosed asthma.
Participants living in buildings with both supply and exhaust air ventilation systems had more doctor's diagnosed allergy than those living in buildings with only exhaust air ventilation (p = 0.01). Moreover, subject living in recently decorated flats (p = 0.02) and those reporting mould odour at home (p = 0.02) had more doctor's diagnosed allergy while those who had management accessibility both in the neighbourhood and at the division level had less doctor's diagnosed allergy (p = 0.007) (Table 5). Building construction year, number of flats in the building, energy use for heating, type of ownership, type of maintenance management, number of rooms in the flat, ETS, number of persons/room, damage from damp, humid air or furry pet keeping were not associated with doctor's diagnosed allergy.
Building construction years as a combined categorical variable was associated with pollen allergy (p = 0.04) with a tendency of less pollen allergy in building constructed 1961–1975. Buildings using more energy for space heating had less pollen allergy (p = 0.001) and those living in buildings with maintenance management by contract had less pollen allergy (p = 0.004) as compared to those whose owners took care of building maintenance themselves. Moreover, reports on humid air (p = 0.001) and mould odour (p = 0.001) were associated with pollen allergy while those living in flats with furry pets had less pollen allergy (p = 0.005) (Table 6). Number of flats, type of ventilation, type of ownership, type of management accessibility, and number of rooms in the flat, ETS, number of persons/room, airing habits, recent redecoration or damage from damp were not associated with pollen allergy.
Subjects living in larger buildings (number of flats) had more eczema (p = 0.02) and those living in buildings using more energy for space heating had less eczema (p = 0.05). Moreover those with management accessibility at the division level had more eczema (p = 0.02) as compared to those with management accessibility in the neighbourhood. Finally, those reporting water damage during the five years (p = 0.05), humid air in the flat (p = 0.02) or mould odour (p = 0.03) had more eczema (Table 7). Building construction year, type of ventilation, type of ownership, type of maintenance management, ETS, number of persons per room, airing habits, recent redecoration and furry pet keeping were not associated with eczema.
Eleven of the fifteen selected building/flat variables - building construction year, number of flats, type of ventilation, energy use for heating, type of maintenance management, type of management accessibility, recent redecoration, damage from damp, humid air, mouldy odour and pet keeping - were either positively or negatively associated with asthma, allergy or eczema. Energy use for heating, type of maintenance management, type of management accessibility, humid air and mouldy odour were the most important exposure variables associated with more than one health variable. Most studies on health associations between indoor home environment and asthma, allergies and eczema have focused on environmental factors on the flat level. Few have studied factors on the building level and we found no previous studies on health associations between building management and management accessibility.
Epidemiological studies can be influenced by selection bias, recall bias (information bias) and type of statistical model. The participation rate was relatively high (74%). There were some differences in the proportion of participants between age classes of buildings, but they were relatively small. The study was based on a stratified random sample of the total population in Stockholm, with a higher sampling proportion for newer buildings and a lower sampling proportion for older buildings. The stratified random sampling influenced the prevalence of certain building related factors, such as the number of flats in the building, type of ventilation and type of ownership since these factors were associated with construction year or size of the building. However, mean age or proportion of females, foreign born, smokers, pet keepers or mean family income per year and mean time living in the current dwelling and the prevalence of asthma, allergies or eczema were similar in the stratified random sample as compared to the estimated total population in Stockholm. Thus, selection bias is unlikely to have had any major influence on the results from the study. Most data was collected from independent sources such as the population register (age, gender, foreign born, family income per year), the building register (construction year, number of flats, type of ownership) or a building questionnaire sent to the building owner (type of ventilation, type of maintenance management, type of heating system, type of management accessibility). Energy use for heating was determined independently by energy experts. However, variables concerning the flat were collected in the same questionnaire as the medical information and could be influenced by information bias. To reduce the number of statistical models, only four were used, one for each health variable. The advantage of this approach is that it reduces the risk for mass significance and the use of mutually adjusted models gives a better estimate of the true association, when adjusting for all exposure variables and potential confounders. Since the response rate was not 100% for each independent variable, imputation was applied. When comparing models with and without imputation, similar associations and estimates of the ORs were found (data not shown). In conclusion, it is less likely that the study was seriously hampered by selection bias, information bias or the selection of a particular statistical model.
The prevalence of doctor's diagnosed asthma was 10.9% in our study, performed in 2005. This is somewhat higher than the 9.3% doctor's diagnosed asthma reported from another population study in adults (20–79 years), performed in the Stockholm area in 2007 . The prevalence of pollen allergy (hay fever) was 22.6% in our study, somewhat lower than the prevalence of adult allergic rhinitis (28%) in the Stockholm area 2006–2007 . This latter study's definition of allergic rhinitis is similar to our definition of hay fever (pollen allergy). The prevalence of eczema in our study in adults (more than 18 y) was 22.7%, much lower than the prevalence of lifetime eczema in adults (16–75 y) (40.7%) reported from a population study in western Sweden . Their question about eczema was somewhat different, since they asked about eczema or any kind of skin allergy, while we only asked about eczema.
The study confirms the conclusions from other studies suggesting that building dampness and moulds in homes are associated with asthma and to some extent allergies and eczema . Water/dampness damage over the last five years was associated with eczema only, while indications of humid air was associated with doctor's diagnosed asthma, pollen allergy and eczema, and reports on mould odour showed associations with doctor's diagnosed allergy, hay fever and eczema. Humid air is an indicator of high relative air humidity in indoor air and can be influenced by low ventilation flow as well as the evaporation of water in flat. Mouldy odour at home has been reported to be consistently associated with onset of asthma  but we found only one previous study reporting an association between pollen allergy and mouldy odour . The association in our study between dampness/moulds and eczema is in agreement with other population studies among adults in Sweden , ,  and China . One large multicenter study, mainly from Europe, found an association between adult eczema in the initial analysis, but the associations were not significant when adjusting for other indoor exposures and confounders .
New building material and indoor paints can emit formaldehyde and volatile organic compounds (VOC). The question about redecoration in our study revealed examples of redecoration, including change of floor materials, wall papers and indoor painting. Indoor painting is a common type of redecoration in dwellings . We found an association between recent indoor redecoration and doctor's diagnosed allergy. This is in agreement with a previous study among Japanese female university students reporting an association between recent indoor painting and furry pet allergy .
Furry pet keeping was negatively associated with pollen allergy (hay fever). The most likely explanation is a selection effect, since subjects with allergies may avoid keeping furry pets. However, it has also been demonstrated that furry pet keeping may be associated with increased levels of endotoxin in house dust  and endotoxin may be protective against allergy, but the protective effect is mainly considered to occur in early childhood .
We found that subjects living in multifamily buildings using more energy for heating reported more pollen allergy and more eczema. In Sweden, energy saving in buildings has been undertaken for decades, with gradual reduced energy consumption per square meter of heated area. However, due to the global concern about greenhouse emissions, energy saving in buildings is still of major concern in both Sweden and abroad. Our data suggests that negative health consequences linked to energy saving in buildings should be considered. We found associations between energy consumption and pollen allergy and eczema after adjusting for a large number of confounder and other building/flat factors. Despite no information in our study on ventilation flow, only type of ventilation (which was adjusted for), it can be assumed that the buildings with higher energy consumption have better natural ventilation through air leakage in the building construction. In our study most multifamily building had a mechanical ventilation system but the majority only had exhaust ventilation. The potential of energy-saving is largest for a combined supply/exhaust ventilation system since heat exchangers can be used to transfer heat from the exhaust to the supply air ducts during the heating season. However, we found more pollen allergy in buildings with supply/exhaust ventilation, as compared to those with exhaust ventilation systems only. This illustrates a potential risk of supply/exhaust ventilation systems. The reason for the observed association is unclear, and should be confirmed in further studies. However it is important to avoid transport of pollen allergens and other outdoor particle pollutants through the supply air system to the indoor environment through sufficient air filtration and regular filter exchange.
We found that subjects in larger buildings reported more eczema. We found no other studies on associations between building size and adult eczema, but an association between degree of urbanisation and adult eczema has been reported in one Swedish population study from 2012  as well as in a systematic review . One explanation for our finding could be that areas with larger buildings have a larger degree of urbanization within the Stockholm area.
Asthma, allergies or eczema among adults in Stockholm are more common in buildings with less use of energy for space heating, in larger buildings and in dwellings with redecoration, mould odour, dampness and humid air. There is still a need to reduce chemical emissions from new building materials and to control dampness and microbial growth in the dwellings. Increasing the energy efficiency in buildings may have consequences for health, and increased insulation of walls or roof should be combined with better air exchange, using an energy efficient mechanical ventilation system. More studies are needed on the role of building management for the inhabitants' asthma, allergies and eczema.
We would like to thank all project members in the 3H and 3HE projects for previous work performed to produce a high-quality database for our data analysis.
Conceived and designed the experiments: DN KE. Performed the experiments: KE. Analyzed the data: DN EL KE. Contributed reagents/materials/analysis tools: KE. Wrote the paper: DN EL KE.
- 1. WHO. Website for WHO Regional office for Europe. WHO Guidelines for Indoor Air Quality: Dampness and Mould. Copenhagen and Bonn: WHO Regional office for Europe. Available: http://www.euro.who.int/_data/assets/pdf_file/0017/43325/E92646.pdf (assessed 16 January 2014).
- 2. Delfino RJ (2002) Epidemiological evidence for asthma and exposure to air toxics: Linkages between occupational, indoor and community health air pollution research. Environ Health Perpect 110 (Suppl 4)573–589.
- 3. Thomson NC (2007) The role of environmental tobacco smoke in the origins and progression of asthma. Curr Allergy Asthma Rep 7:303–309.
- 4. Jie Y, Ismail NH, Jie X, Isa ZM (2011) Do indoor environments influence asthma and asthma-related symptoms among adults in homes? A review of the literature. J Formos Med Asssoc 110:555–563.
- 5. Herr CE, zur Nieden A, Stilianakis NI, Gieler U, Eikmann TF (2004) Health effects associated with indoor storage of organic waste. Int Arch Occup Environ Health 77:90–96.
- 6. Norbäck D, Björnsson E, Janson C, Widström J, Boman G (1995) Asthmatic symptoms and volatile organic compounds, formaldehyde, and carbon dioxide in dwellings. Occup Environ Med 52:388–395.
- 7. Wieslander G, Norbäck D, Björnsson E, Janson C, Boman G (1997) Asthma and the indoor environment: the significance of emission of formaldehyde and volatile organic compounds from newly painted indoor surfaces. Int Arch Occup Environ Health 69:115–124.
- 8. Zock JP, Plana E, Jarvis D, Anto JM, Kromhout H, et al. (2007) Household cleaning sprays and adult asthma: an international longitudinal study. Am J Respir Crit Care Med 176:735–741.
- 9. Gordian ME, Stewart AW, Morris SS (2010) Evaporative gasoline emissions and asthma symptoms. Int J Environ Res Public Health 7:3051–3062.
- 10. Jaakkola JJ, Leromnimon A, Jaakkola MS (2006) Interior surface materials and asthma in adults: a population-based incidence case-control study. Am J Epidemiol 164:742–749.
- 11. Nguyen T, Lurie M, Gomez M, Reddy A, Pandya K, et al. (2010) The national asthma survey-New York State: association of the home environment with current asthma status. Public health Rep 125:877–887.
- 12. Takaoka M, Norbäck D (2011) The home environment of Japanese female university students-association with respiratory health. Indoor Built Environ 20:369–376.
- 13. Peden D, Reed CE (2010) Environmental and occupational allergies. J Allergy Clin Immunol 125 (Suppl 2): (2): 150–160.
- 14. Engvall K, Norrby C, Norbäck D (2001) Asthma symptoms in relation to building dampness and odour in older multifamily houses in Stockholm. Int J Tubercl Lung Dis 5:468–477.
- 15. Sun Y, Zhang Y, Sundell J, Fan Z, Bao L (2009) Dampness at dorm and its associations with allergy and airways infections among college students in China: a cross-sectional study. Indoor Air 19:174–182.
- 16. Sahlberg B, Wieslander G, Norbäck D (2010) Sick building syndrome (SBS) in relation to domestic exposure in Sweden - A cohort study from 1991 to 2001. Scand J Public Health 38:232–238.
- 17. Sahlberg B, Norbäck D, Wieslander G, Gislason T, Janson C (2012) Onset of mucosal, dermal, and general symptoms in relation to biomarkers and exposures in the dwelling: a cohort study from 1992 to 2002. Indoor Air 22:331–338.
- 18. Li-Qun L, Chun-Xia L, Yun-Guang G (2014) Green and sustainable city will become the development objective of China's low carbon cities in future. J Environ Health Sci Eng 12:34.
- 19. Wilkinson P, Smith KR, Davies M, Adair H, Armstrong BG, et al. (2009) Public health benefits of strategies to reduce greenhouse-gas emissions: household energy. Lancet 374:1917–1929.
- 20. Thomson H, Thomas S, Sällsrom E, Petticrew M (2013) Housing improvements for health and associated socio-economic outcomes. Cochrane Database Syst Rev 2:CD008657.
- 21. Breysse J, Jacobs DE, Weber W, Dixon S, Kawecki C, et al. (2011) Health outcomes and green renovation of affordable housing. Public Health Rep 126 (suppl 1)64–75.
- 22. Howden-Chapman P, Crane J, Chapman R, Fougere G (2011) Improving health and energy efficiency through community-based housing interventions. Int J Public Health 56:583–588.
- 23. Spertini F, Berney M, Foradini F, Roulet CA (2010) Major mite allergen Der f 1 concentration is reduced in buildings with improved energy performance. Allergy 65:623–629.
- 24. Thornton EW (1994) Health risks of energy saving. Nature 372:327.
- 25. Sobottka A, Thriene B (1996) Sanitation programs for living spaces and health risks involved. Toxicol Letters 88:365–368.
- 26. Hirsch T, Hering M, Bürkner K, Hirsch D, Leupold W (2000) House-dust-mite allergen concentrations (Der f 1) and mold spores in apartment bedrooms before and after installation of insulated windows and central heating systems. Allergy 55:79–83.
- 27. Wilson J, Dixon SL, Jacobs DE, Breysse J, Akoto J, et al. (2014) Watts-to-wellbeing: does residential energy conservation improve health? Energy Efficiency 7:151–160.
- 28. Fabian MP, Adamkiewicz G, Stout NK, Sandel M, Levy J (2014) A simulation model of building intervention impacts on indoor environment quality, pediatric asthma and costs. J Allergy Clin Immunol 133:77–84.
- 29. The Swedish Council for Building Research (1996) The energy book-a resume of present knowledge and research. Stockholm: The Swedish Council for Building Research. 319 p.
- 30. The Swedish Energy Agency (Energimyndigheten) Energy in Sweden 2013 (2013). Available: https://energimyndigheten.a-w2m.se/FolderContents.mvc/Download?ResourceId=2918. Accessed 4 November 2014.
- 31. Engvall K, Norrby C, Sandstedt E (2014) The Stockholm indoor environment questionnaire: a sociologically based tool for the assessment of indoor environment and health in dwellings. Indoor Air 14:23–33.
- 32. The National Board of Housing, Building and Planning (Boverket) (2009) Register for data from the energy declaration form in Swedish Buildings (The Gripen database). Available: https://gripen.boverket.se. Accessed 4 November 2014.
- 33. Engvall K and Wickman P (2010) Methods for quality assurance of data on energy use in 472 residential buildings included in a multidiciplinary study about health sustainable houses in Stockholm (3HE-study) Report nr 1, 2011 (in Swedish). Available: http:/ammuppsala.se/3H. Accessed 4 November 2014.
- 34. Harrell FE (2001) Regression modelling strategies: with applications to linear model, logistic regression and survival analysis. New York: Springer Press. 568 p.
- 35. Ekerljung L, Andersson A, Sundblad BM, Rönmark E, Larsson K, et al. (2010) Has the increase in the prevalence of asthma and respiratory symptoms reached a plateau in Stockholm, Sweden? Int J Tubercl Lung Dis 14:764–771.
- 36. Eriksson J, Ekerljung L, Rönmark E, Dahlen B, Ahlstedt S, et al. (2012) Update of prevalence of self-reported allergic rhinitis and chronic nasal symptoms among adults in Sweden. Clin Respir J 6:159–168.
- 37. Rönmark EP, Ekerljung L, Lötvall J, Wennergren G, Rönmark E, et al. (2012) Eczema among adults: prevalence, risk factors and relation to airway diseases. Results from large-scale population survey in Sweden. Br J Dermatol 166:1301–1308.
- 38. Quansah R, Jaakkola MS, Hugg TT, Heikkinen SA, Jaakkola JJ (2012) Residential dampness and molds and the risk of developing asthma: a systematic review and meta-analysis. Plos One 7:e47526.
- 39. Harrop J, Chinn S, Verlato G, Olivieri M, Norbäck D, et al. (2007) Eczema, atopy and allergen exposure in adults: a population-based study. Clin Exp Allergy 37:526–535.
- 40. Moniruzzaman S, Hägersted Engman L, James P, Sigsgaard T, Thorne PS, et al. (2012) Levels of endotoxin in 390 Swedish homes: determinants and the risk for respiratory symptoms in children. Int J Environ Health Res 22:22–36.
- 41. Radon K (2006) The two sides of the “endotoxin coin”. Occup Environ Med 63:73–78.
- 42. Schram ME, Tedja AM, Spijker R, Bos JD, Williams HC, et al. (2010) Is there a rural/urban gradient in the prevalence of eczema? A systematic review. Br J Dermatol 162:964–973.