Using GIS and dispersion modelling tools to assess the effect of the environment on respiratory health

Using GIS and dispersion modelling tools to assess the effect of the environment on respiratory health:

linking lung function and personal exposure to air pollution

GIS/EM4 No. 177

Helen Crabbe
Ron Hamilton
Nuria Machin

Extended Abstract: [See special issue of Transactions in GIS for full paper]

A feasibility study in progress to integrate health and air quality information is presented. The methods of using integrated GIS and air quality dispersion modelling tools to assess the effect of the environment on health are discussed. The main aim is to model human personal exposure to air pollutants and compare to respiratory health data for asthmatics in a health telematics project.
The MEDICATE project develops and tests the feasibility of using a health telematic system for asthmatics. Key to this is the determination of the real-time health (lung function) response to the ambient environment. For this, air quality information is related to respiratory measurements by modelling personal exposure through a GIS. The methods of integrating environmental modelling and assessment tools (GIS) in this case are examined. ESRI’s ArcView is used to locate and compile environmental information about the patients’ locations and lifestyles in the study areas (London and Barcelona). A dispersion modelling extension to ArcView, ADMS Urban, is used to interrogate the spatial environmental databases (e.g. emission inventories) to model air pollution concentrations. Patients’ personal exposure is modelled by time-weighted estimates of their exposure to ambient air quality at each defined location. Lung function data can be compared on a time-wise basis with these air quality indicators to see if there is a relationship on an hourly, daily or lagged-day basis. The aim of this paper is to propose a methodology and discuss the integration and practicalities of using air quality assessment tools (inventories, dispersion modelling and use of a GIS) in assessing the impacts of the environment on human health.

Keywords

Air quality, respiratory health, lung function, dispersion modelling, personal exposure, health telematics.

Data collection exercises:

The MEDICATE (Medical Diagnosis, Communication and Analysis Throughout Europe) project is an EC TEN-TELECOM funded project for the development and feasibility of using a health telematic system for asthmatics. Two hospitals are currently in the trial, with study areas in Manresa (Barcelona, Spain) and North London, U.K. Asthmatics admitted into hospital are recruited onto the feasibility trial, subsequently discharged and instructed to use a remote lung function monitoring device to record lung function measurements electronically. They then send this data to the hospital using a modem base station at home. Clinical consultants have access to this data and thus have knowledge about the patient’s symptoms and wellbeing after discharge. The monitor is able to record all lung function parameters, symptoms, medication use and location of readings. One aim of the project is to assess the impact of environmental factors on respiratory health, namely lung function response.

A number of data gathering exercises are conducted to collect environmental information on each patient on the trial. The project attempts to use a more representative value of personal exposure than ambient air quality concentrations to represent dose-response relationships. Dispersion modelling tools linked to spatial environmental databases within the GIS are used to develop a methodology to predict a representative measure of human personal exposure. Progress to date is discussed, presenting no results but emphasising the development of a methodology of using such tools. The intention is to integrate respiratory health and air quality information, including personal exposure, on a temporal and spatial scale. (Crabbe et al. 2000a,b).

To compute each individual’s exposure to air pollution and other environmental factors (allergens and irritants), a series of data gathering exercises are necessary. Questionnaires were devised to collect this data and are executed or given to each patient on the MEDICATE trial. Three different data gathering steps were devised;
a. An environmental factors questionnaire. In order to identify the patient’s exposure to allergens and irritants, an ‘environmental factors’ questionnaire was designed. Its primary aim is to identify patient lifestyles, activity patterns and collect information for the GIS for personal exposure modelling such as indoor and outdoor conditions in a spatial context.
b. Daily health diaries and use of a location button. An electronic record of the patient’s location is also recorded at the time of each lung function measurement. Patients have the choice of pressing a button to record one of four categories of location to describe where they are. The categories are: home, work, commuting/travelling or outdoors/other. As there can be some overlap between the classifications, a written daily health diary allows for an explanation of the precise location of the reading, for personal exposure analysis.
c. An asthma quality of life questionnaire (AQLQ) is also administered to the patients on the trial at the end of the two-week period to collect psychosocial and perception data.

Integrating air quality, personal exposure and respiratory health information.

For each patient enrolled on the trial various data gathering techniques are carried out, involving conducting the questionnaires already described. The following steps integrate the various types of information.

The patients home and work locations are spatially identified in the GIS for preparation for use in dispersion modelling scenarios and to identify proximity to air quality monitoring stations. ESRI’s ArcView is used to locate the spatial information needed to integrate environmental factors for both the London and Manresa (Barcelona) study areas.
An emission inventory database within the GIS is compiled for each study area. For London, data from the London Research Centre’s (LRC) database on the London Atmospheric Emissions Inventory (LAEI) (LRC, 1997) was utilised to compile information on point sources (industries), road sources and area sources. Information on the industries polluting to air in the Manresa area is based on data from the Ajuntament de Manresa (1999) using the Catáleg d’Activitats Potencialment Contaminadores de l’Atmosfera (CAPCA) database, and the Generalitat de Catalunya (2000). Background emissions data are also added at an appropriate spatial level using the EC EMEP European emission inventory (EMEP, 1999).
The ‘new generation’ ADMS Urban dispersion model (Carruthers, et.al. 1997) is utilised to model concentrations of air quality at patient locations. The model is a custom built, commercially available UK extension to ArcView and is integrated with an emissions inventory relational database within the GIS. This allows for the inclusion of sources within a particular area and the resulting air quality impacts can be modelled within a spatial framework, where air quality monitoring is absent. The windows based multi-source dispersion model can handle up to 1000 line (i.e. traffic) sources as well as point, area and volume sources. The model provides maps of concentration estimates using the Spatial Analyst extension to ArcView, that are used in the estimation of time weighted exposures. The GIS based emissions inventory database is interrogated directly for inclusion of input data into the model.
Model scenarios are prepared using the above emissions data and location information for receptors placed at patient’s locations and air quality monitoring sites. Site specific information on street canyon morphology (road and building dimensions) is also added into the GIS at this stage. The model is run under different meteorological conditions to obtain hourly averages, daily averages and maximum levels that are likely to be experienced at the defined locations.
Patient’s total exposure on a daily basis are calculated using equations of integrated exposure of concentrations over time (Crabbe et al, 2000b). This uses weighted modelled concentrations to represent air quality at each receptor from the scenario runs. Patients’ total calculated hourly, daily and lagged-day personal exposure levels are compared to the various lung function variables measured by the monitor. This is represented by time series graphs and statistical analysis involving multiple correlation analysis of individual exposure to pollutants against respiratory function parameters, including a relationship to local pollen levels. The results of the comparison are displayed on an intranet/www when analysis is complete.

Although many factors will influence respiratory health, including other environmental parameters, this project tries to determine the effect of air pollution on measured lung function and respiratory symptom parameters. The environmental emphasis in the project focuses on air quality and the short-term impacts on respiratory health. The results from this project and its suitability for use will be available towards the end of 2000. It is envisaged that the project will be extended in the future to several other European countries and the U.S. involving several thousand patients.

References

Ajuntament de Manresa (1999) Auditoria Ambiental Municipal de Manresa, Document 1. Memória descriptiva i diagnosi, (and personal communications). Ajuntament de Manresa, Barcelona.

Carruthers, D.J., Edmunds, H.A., McHugh, C.A., Riches, P.J. and Shingles, R.J. (1997) ADMS Urban - an integrated air quality modelling system for local government. In Power, H.., Tirabassi, T., Brebbia, C.A. (Eds.) Air Pollution V, Modelling, Monitoring and Management. Southampton, Computational Mechanics Publications, pp 45-58.

Crabbe, H., Hamilton, R. and Machin, N. (2000a) Integrating health and air quality information for use in a health telematics project. In Brebbia, C.A., Longhurst, J.W.S. and Power, H. (eds) Air Pollution VIII, WIT press, Southampton (in press).

Crabbe, H., Hamilton, R. and Machin, N. (2000b) Using GIS and dispersion modelling tools to assess the effect of the environment on health. Transactions in GIS (in press).

EMEP (1999) Co-operative programme for monitoring and evaluation of the long range transmission of air pollutants in Europe. WWW document, http://www.emep.int/index.html

Generalitat de Catalunya (2000) personal communication, Barcelona, January 2000.

London Research Centre (LRC) (1997) The London Atmospheric Emissions Inventory, LRC, London.

Authors

Helen Crabbe, Research Fellow, Urban Pollution Research Centre,
School of Health, Biological and Environmental Sciences,
Middlesex University, Bounds Green Road, London. N11 2NQ. U.K.
E mail: h.crabbe@mdx.ac.uk. Tel: +44 (0)20 8362 6361 Fax: +44 (0)20 8362 6580

Professor Ron Hamilton, Dean of School and Nuria Machin, Research Fellow,
School of Health, Biological and Environmental Sciences,
Middlesex University, Furnival Building,10 Highgate Hill, London. U.K. N19 3UA.
Email: r.hamilton@mdx.ac.uk or n.machin@mdx.ac.uk
Tel: +44 20 8362 6803, Fax: +44 20 8352 5232