In most instances, flu surveillance data are incomplete. Official figures often extrapolate from reported data to estimate the total number of cases, hospitalizations, and/or deaths for the total population or sub-populations or age groups of interest. These estimates inherently include a degree of uncertainty, depending on the volume and quality of the underlying data. A confidence interval (CI) is one tool for expressing that uncertainty, providing range of values that are expected to contain the true value (e.g., mean) a certain percentage of the time, commonly at the 90% or 95% level. If you choose to do so, you can provide 95% CIs for your data, and the Flu Tool will adjust the outputs accordingly.