Second, we expanded the unadjusted zero-truncated negative binomial models to include multiple environmental factors that might influence inhaler use simultaneously, including air pollution, pollen, and meteorological data (Equation 3):
In our zero-truncated negative binomial modeling process, the exposure values from all rescue inhaler events occurring in a single day were averaged to daily means.
We also conducted a third sensitivity analysis, which implemented the zero-truncated negative binomial regression models using only the inhaler use data that had geolocation information.
We also identified the feasibility of using sensor-collected data to detect possible associations with environmental triggers, including air pollution, pollen, and mold, in the unadjusted zero-truncated negative binomial models (see Table S2).
In addressing the first goal, the study estimated the rural tourism demand function for the self-drive travel market by using four statistical models: the Ordinary Least Square regression (OLS) model, the Zero-truncated Poisson regression model, the Zero-truncated Negative Binomial
regression model, and the duration model with Log-logistic regression.