In response to the high incidence of infantile asthma in Puerto Rico (PR), this study aimed to predict infantile asthma based on fungal spore, pollen, pollutant concentrations, and/or meteorological factors. A predictive model would allow for the creation of an alert to inform the general public about the risk of infantile asthma on a daily basis. Simulating dynamic linear models with discount factors using OpenBUGS, R, and R2OpenBUGS we constructed models which explained pediatric asthma cases at San Jorge Children's Hospital and the University of PR's Carolina Hospital by atmospheric ozone concentration or by a combination of total airborne fungal spore and ozone concentrations with and without interaction. High autocorrelation of residuals led us deseasonalize using a Fourier model for San Jorge Children's Hospital. By minimizing the deviance information criterion (DIC), and analyzing model coe cients and residuals we chose model yt = β 0,t+ β1,ttotalSporest + β,tozonet+ ϵ t where ϵ t represents errors. Residuals seemed to follow right-skewed distributions and we did not manage to approximate normality for any model. This result undoubtedly serves as a dynamically predictive model for the monthly data and might serve as starting points for future research with more complete, hopefully daily, data. Further analysis should attempt to better clarify outliers and to t a regressive model explaining seasonality at San Jorge. Furthermore we would like to compare our model with a non-regressive one that combines linear growth and seasonality. Future research should also work on establishing a network of daily and island-wide asthma case registry from hospitals and physicians. This type of information would greatly assist in creating a model with much better predictive capacity than the present one.