Exposure Through Soil and Dust Ingestion - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 275

Hand-to-mouth frequency data are limited and difficult to collect. Few published
studies report hand-to-mouth frequency data. Different data collection approaches
are used, such as videotaping versus real-time observation, data analysis and report-
ing methods. Studies differ also by ages of children, locations and definitions of
mouthing. Xue et al. ( 2007 ) pooled and reanalysed indoor and outdoor mouthing
frequency data. The results of this meta-analysis are consistent with the results from
other studies reported in the literature. Most authors conclude that age and location
(indoor or outdoor) are important for hand-to-mouth frequency: as age increases,
frequencies decrease (Freeman et al. 2001 ; Tulve et al. 2002 ). Hand-to mouth fre-
quency is also higher indoors than outdoors (AuYeung et al. 2004 ; Freeman et al.
2001 ) and some other authors (AuYeung et al. 2004 ; Freeman et al. 2001 ) observe
higher frequencies for girls than for boys.
Most authors assume that each hand-to-mouth event equals a contact area of 3
fingers (Kuusisto et al. 2007 ;US-EPA 2001 ). Depending on the age, the contact area
ranges between 11 and 33 cm 2 from infant to 12 year old children (Kuusisto et al.
2007 ).
The results from hand-to-mouth transfer studies typically provide hourly inges-
tion rates of soil and/or house dust which can be compared with the daily ingestion
rates derived from tracer element studies when information exists on how children
have spent their time indoors and outdoors. As the location and the type of activ-
ity children are involved in (indoor or outdoor activities; passive or dynamic play)
determine the results from hand-loading studies, care should be taken when compar-
isons with tracer studies are made. Hand-loading studies can theoretically provide
the most accurate soil and dust ingestion values for humans and allow distinction
between soil (outdoor) and dust (indoor) ingestion. However, since many variables
have to be taken into account, the resulting soil and dust ingestion rates show a
rather high variation.
The Biokinetic model comparison methodology is based on the validation of
exposure models for contaminants which have soil or dust ingestion as one of the
major exposure routes and in which modelled and measured internal exposure can
be compared (Van Holderbeke et al. 2008 ;US-EPA 2008 ). Only studies that meet
the following criteria can be used to estimate soil/dust ingestion rates:

•

the contaminant should contribute significantly and preferably dominantly to the
exposure via the soil/dust ingestion pathway;

• the biomarker for model validation should relate to actual exposure;

• an adequate pharmacokinetic model, linking external or absorbed dose to the

biomarker, should be available.

Case studies that meet these criteria are mostly available for lead and children's
exposure. Some studies for arsenic are also available. The methodology is at present
limited to rough estimates, due to the uncertainty with regard to oral bioavail-
ability in the human body (if not measured), contribution of soil versus dust and
the representativeness of the soil and dust concentrations measured. Simulations
can be performed to obtain an optimal fit based on modifying the soil/dust inges-
tion rates and the contribution of soil and dust. However, in order to be able