ANNOUNCEMENTS
This time series study has been conducted to analyse the association between ambient particulate matter pollution and mortality (total non-accidental and cardiopulmonary) in the Indian capital city of Delhi using a generalized additive model. The data from 2013 to 2019 has been used to conduct this study with monthly level pollution and mortality data. The resolution at the monthly level was chosen as availability of mortality data was the limiting factor and only monthly data was easily accessible for the city. Two pollutants were considered for the study namely PM2.5 and PM10 which fall in the category of ambient particulate matter pollutants. To account for the confounding effects on the variation in the mortality numbers, two temporal variables- month (to account for the intra-annual or seasonal variation) and year (to account for the inter-annual variation) have been used in the model. The results that were obtained indicate a highly non-linear relationship between ambient particulate matter pollution and mortality at the monthly level and hence the model coefficients and consequently the relative risk of mortality is indeterminable at the monthly level of resolution for the city. Based on the available literature, the possible change in pathway (using daily level mortality data) to conduct the time series study has been conjectured to yield different (and maybe more linear) results. If the aforementioned pathway leads to quasi-linear results that can be approximated by linearity and which consequently will result in determinable model coefficients and relative risk estimates, then our results could serve as an indicator of the importance of the temporal resolution of the available data.
Keywords: ambient particulate matter pollution, mortality, time series, generalized additive model, Delhi.
