Piscirickettsiosis (SRS) is an endemic bacterial disease of high economic importance and is the primary reason for antibiotic usage in the aquaculture industry in Chile. Understanding the epidemiology of this disease is important in order to develop better control strategies for the Chilean aquaculture industry. The objectives of this project were to 1) describe the epidemiology of SRS on Atlantic salmon and rainbow trout farms, and 2) to identify factors that impact the severity of SRS outbreaks. Special attention was given to vaccine strategies currently used by the industry. Production data from 14 Atlantic salmon farms (252 cages) and 11 rainbow trout farms (216 cages) that had completed their production cycles between 2010 and 2012 were investigated. Regression models were used to evaluate time-to-first outbreak of SRS and total mortality attributed to SRS mortality. The factors evaluated in our models were: vaccine type (control and 5 vaccines), smolt weight and cumulative mortality within the first 4 weeks in saltwater, season of smolt introduction, infection with other pathogens during the production cycle, SRS treatments, and total number of sea lice treatments. The prevalence of SRS-affected cages on infected farms was high for both species; however, outbreaks appeared more severe (i.e. higher mortalities) on rainbow trout farms. Onset of SRS outbreaks was, on average, for fish in different vaccine groups, between 2480 degree-days (dd) and 3829 dd for Atlantic salmon, and between 1696 dd and 2241 dd for rainbow trout. For both species, none of the vaccines evaluated completely prevented SRS, although there were significant variations in the time-to-first outbreak and the severity of SRS outbreaks associated with vaccines after controlling for farm effect and other predictors. Specifically, a booster vaccine strategy in Atlantic salmon had significantly lower mortalities associated with SRS and a delay in the onset of disease compared to several of the other vaccines evaluated. Whether the differences observed between vaccines are economically significant is unknown. In rainbow trout, time-to-first outbreak was significantly delayed for vaccinated fish compared to the unvaccinated fish after we controlled for other factors in our model; however, total SRS mortality of vaccinated rainbow trout was not significantly different than unvaccinated rainbow trout. Consistent for both species was that mortality during the first 4 weeks post-salt water entry was associated with time-to-outbreak of SRS, and this effect was dependent on the vaccine used.