Indoor Air Quality

3. How can scientists determine whether indoor air pollutants pose a health risk?

The SCHER opinion states:

3.1.1.2. Dose-response assessment

Information about exposure-response relationship is crucial for risk assessment. Regarding local effects such as irritation, air concentrations of pollutants are directly relevant for effects assessment. For evaluation of possible systemic effects, air concentrations of pollutants may be transformed to internal doses using established values for breathing volumes and alveolar retention. When data from occupational exposures are used, they must be adjusted for differences in exposure duration (general population 20-21 hours indoors vs. a work shift of 8 hours, and a longer exposure period of the life span) and the limitations of the occupational observations need to be considered for the exposed population; occupational exposures, in general, do not include infants, children and the elderly. Both acute and chronic effects should be taken into account in the dose-response assessment (INDEX 2005).

Biomarkers, when available, may be used to establish the dose-response relationships.

3.1.1.3. Exposure assessment

More than 900 different organic compounds have been detected in indoor air (SCALE, 2004a) in addition to fine and ultrafine particles and biological material (microbes, allergens). Concentrations of pollutants in indoor air are determined by a number of factors including type and emission rates from sources, ventilation rate, adsorption/absorption of compounds on/in materials (sinks and possible secondary sources).

Differences in cultural habits of people throughout Europe may result in large qualitative and quantitative differences in indoor air quality. The results of the EXPOLIS-INDEX study concerning exposure to European indoor environments in Athens, Basel, Grenoble, Milan, Helsinki, Oxford and Prague (e.g. VOCs and PM2.5) showed larger within city variation in exposure than average between city variation (EXPOLIS-INDEX, 2004). The health consequence of these variations is insufficiently understood.

Emissions of chemicals may occur from building materials (e.g. Bornehag et al., 2005a), cooking activities (e.g. Afshari et al., 2005), cleaning activities (Nazaroff and Weschler 2004, Singer et al., 2006), heating and combustion of biomass fuels in general (e.g. the opinion on air fresheners, SCHER 2006) but the exposure patterns are not sufficiently known.

Models have been developed to predict the emissions and distribution patterns of pollutants. Such models are essential for the development indoor exposure and risk assessment (ECA 2006, Kephalopoulos et al., 2007). However, no comprehensive general model has yet emerged and been validated.

Exposures to indoor air pollutants may occur directly by inhalation, or indirectly by ingestion of e.g., dust, while volatile compounds such as formaldehyde and benzene are mainly present in the gas phase. Less volatile substances are also to some extent bound to particles and dust, and exposure via those media may contribute to the total exposure. Many semivolatiles such as phthalates, flame retardants, PAHs, chlorophenols, pesticides, organotins and metals may adsorb to house dust (Butte and Heinzow, 2002). Particles may abrade from materials containing the chemicals of interest, e.g. PVC particles containing phthalates. House dust forms a long-term sink, may be regularly resuspended and represent a secondary source for pollutants. The particle size of house dust is, however, typically large and only a fraction of the dust is respirable (Butte and Heinzow, 2002, Maertens et al., 2004). Ingestion is likely the main exposure route for house dust (Butte and Heinzow, 2002) in small children due to licking and biting on articles and “hand to mouth” pattern. Small children spend a considerable part of the time on the floor and may thus be more exposed to resuspended dust than adults. The particle/dust pathway has been shown to be an important exposure route e.g. for polybrominated diphenyl ethers (PBDEs). Exposure to dust may account for about 14% of the external exposure of those substances for adults, but over 90% for children in the age of 1 to 4 years (Shoeib et al, 2004, Stapleton et al, 2005; Wilford et al, 2005). However, there is insufficient information in general to what extent the exposure to dust actually leads to uptake of pollutants in children.

In the absence of exactly measured data, the exposure should be estimated using validated models and information on local conditions. Modelling of chemical substances has the longest tradition (WHO 2005a). At present, there are no reliable models for exposure to indoor air microbial exposure.

SCHER states that it is important to evaluate effects of inhalation exposure but also underlines the importance to assess the contribution of indoor air exposures to total exposure to a chemical by all routes (ingestion, inhalation and dermal) for assessment of systemic effects to obtain an overall risk assessment and to estimate the contribution of indoor air pollution. The frequent focus on VOCs and other selected compounds in measurement campaigns may neglect a possible impact of compounds with high toxicity present in low concentrations or of compounds which are difficult to detect by the procedures applied. Moreover, the SCHER recommends to use realistic exposure scenarios and to avoid conclusions on exposures based on the content of potentially toxic chemicals in materials such as wallpaper or furnishings. Biomonitoring will make a very valuable contribution to the exposure assessment.

3.1.1.4. Risk characterisation

For some pollutants occurring in indoor air, international (WHO 2000) and national (EPA, OEHHA, ATSDR, UBA, Health Canada, cited e.g. in the INDEX-report, INDEX 2005) health- based exposure limit/guidelines exist. They may not be aimed specifically for indoor air but may be used considering the specific exposure situation.

In risk characterisation, the whole exposure range should be included, not only average or median exposures. The variations in exposures may expand over several orders of magnitude. Analysis of EXPOLIS study data on VOCs has shown that the group at the upper 95% end of the distribution may get exposed significantly more than the median group (Edwards et al., 2005). On the other hand, the most sensitive subgroups may react at notably lower exposures. Accordingly, the range of plausible risk estimates (not only the central estimate) is useful. The precision and the uncertainty related to risk estimates should also be given in assessments.

Indication of the association between exposure and health effects may be difficult when the primary causing factor/agent is not known though the association is evident. This is typical for indoor air problems caused by microbes. For those cases, exclusion of other contributing factors may strengthen the association.

Source & ©: SCHER,  Opinion on risk assessment on indoor air quality (2007),
3.1.1. Risk Assessment for indoor air, p.9-11