We also conducted several sensitivity analyses to evaluate the
robustness of our results for the association between air pollution
and OHCD. First, we changed the df per year in the smoothness
function for time trend from 4 to 10. Second, we controlled for
longer temperature lags using the moving averages of the 0e3, 0e7
and 0e14 days. Third, we re-analyzed the data after excluding the
days with extremely high air pollution levels (above 95th
percentile)