10.06.2013 Views

Auditorium - Associazione Italiana di Epidemiologia

Auditorium - Associazione Italiana di Epidemiologia

Auditorium - Associazione Italiana di Epidemiologia

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

poster 7 settembre<br />

AGREEMENT BETWEEN NO2 PASSIVE SAMPLERS AND URBAN MONITORING<br />

STATIONS<br />

Samantha Sartori 1 , Cristian Pattaro 1 , Marta Rava 1 , Roberto Bono 2 , Roberto De Marco 1<br />

1 Division of Epidemiology & Me<strong>di</strong>cal Statistics, University of Verona; 2 Department of public<br />

health and microbiology, University of Turin<br />

The interest of environmental epidemiology towards the effects of air pollution on<br />

health, requires reliable data on in<strong>di</strong>vidual exposure to pollutants.<br />

The measurements from one or more monitoring stations (MSs) may not represent the<br />

exposure of the whole population.<br />

The aim of this study was to investigate the most appropriate measurement of in<strong>di</strong>vidual<br />

exposure when air pollution can only be assessed from data collected by a number of<br />

MS. Single and combined measurements are evaluated and compared.<br />

In the frame of the European Community Respiratory Health Survey II (2001-03), 342<br />

outdoor passive samplers (PSs) were used to measure the 14-day NO2 exposure in the<br />

urban area of two Italian cities. In<strong>di</strong>vidual measurements were compared with the sameperiod<br />

NO2 mean concentrations obtained from local (background and traffic) MSs.<br />

Correlation and Concordance Correlation Coefficients (CCC) were estimated to assess<br />

the agreement between MSs and PSs.<br />

For this purpose, traffic MSs, background MSs, the best MS* and the MS average for<br />

all the MSs of each centre were compared to PS concentrations. The results are shown<br />

in the table.<br />

Centre<br />

Verona<br />

(nr.=222)<br />

3 traffic<br />

3 background<br />

Torino<br />

(nr.=104)<br />

4 traffic<br />

1 background<br />

Overall<br />

(nr.=326)<br />

Corr<br />

95% C.I.<br />

.61<br />

[.52-.69]<br />

.63<br />

[.49-.73]<br />

.62<br />

[.55-.68]<br />

Traffic Background Best monitoring station* MSs Average<br />

CCC<br />

95% C.I.<br />

.46<br />

[.39-.53]<br />

.20<br />

[.14-.26]<br />

.23<br />

[.16-.30]<br />

Corr<br />

95% C.I.<br />

.64<br />

[.56-.72]<br />

.58<br />

[.43-.69]<br />

.62<br />

[.55-.69]<br />

CCC<br />

95% C.I.<br />

.41<br />

[.35-.49]<br />

.15<br />

[.10-.21]<br />

.22<br />

[.17-.28]<br />

Corr<br />

95% C.I.<br />

.66<br />

[.58-.73]<br />

Urban-Traffic<br />

.54<br />

[.34-.68]<br />

Urban-<br />

Background<br />

.63<br />

[.56-.69]<br />

CCC<br />

95% C.I.<br />

0.58<br />

[.50-.67]<br />

Urban-<br />

Background<br />

.31<br />

[.21-.41]<br />

Urban-Traffic<br />

.47<br />

[.39-.55]<br />

Corr<br />

95% C.I.<br />

.66<br />

[.58-.73]<br />

.64<br />

[.51-.74]<br />

.65<br />

[.58-.71]<br />

CCC<br />

95% C.I.<br />

.49<br />

[.42-.56]<br />

.19<br />

[.13-.25]<br />

.29<br />

[.23-.35]<br />

* We define the best MS, as the one with the best correlate with he PS (column Corr), or the one with the<br />

best concordance (column CCC).<br />

When data on NO2 concentration coming from multiple MSs were available, the MS<br />

average appeared to be the statistic summary representing and correlating the exposure<br />

of the population.<br />

101

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!