Figure 6.
Expected minimal standard error (SE) of symptomatic case fatality ratio (sCFR) estimators in the presence of uncertainty around severity parameters, in a mild (2009-like) influenza pandemic scenario. A recruitment capacity of 10,000 cases is assumed. Parameters pM|S, pH|M, and pD|H (the probabilities of medical attention upon symptoms, hospitalization upon medical attention, and death upon hospitalization, respectively) are assumed to be uncertain at the start of the pandemic. The true values of pM|S, pH|M, and pD|H are 0.3500, 0.0157, and 0.0455, respectively, with uncertainty bounds of 0.2–0.5, 0.005–0.03, and 0.01–0.1, respectively. The minimal SE of each sCFR estimator is plotted for the true pandemic scenario (bold line) and for the 8 anticipation scenarios constructed by combining the uncertainty bounds. Those minimal SEs are obtained by optimally allocating the 10,000 recruited cases between surveillance levels and are calculated using equation 6.