The associations between PM10 or PM2.5 concentrations and the
decreases of lung function have been found repeatedly in various studies, but it is not the case in our study as we discovered
in opposite direction concerning PM10. The reasons might, however,
rely on the fact that the causal components and susceptible
subgroups of particulate matter have been unclear (Delfino et al.,
2003b). Moreover, the major sources of air pollutions in our study
location – northern Thailand – come from open burnings and forest
fires, unlike most studies that sources of air pollutions are trafficrelated.
In fact, surrogate measures for organic compounds in PM10
and diesel exhaust particles (Cass and Gray, 1995) showed more
robust associations than did PM10 (Delfino et al., 2003b). Nevertheless,
our finding must be interpreted cautiously due to the fact
that, as discussed above, it might be confounded by uncontrolled
factors.