Extensively Drug-resistant Tuberculosis, Italy and Germany

780 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 part of this study, all samples from malaria case-patients identifi ed through clinical diagnosis were subject to a Paracheck-Pf immunoassay test (Orchid Biomedical Systems, Verna, Goa, India). Results indicated that, at the peak of the apparent malaria outbreak, the percentage of samples from clinically diagnosed cases that produced a positive diagnostic test was as low as 4% (Figure, panel B). These results are unlikely to refl ect poor diagnostic performance of the testing (6); febrile illness other than malaria was likely the cause of the outbreak. Recent experiences in Kabale also highlight the potentially unwieldy nature of indoor residual spraying campaigns in the absence of spatial targeting. In Kabale, a district-wide spraying campaign supported by the US President’s Malaria Initiative (7) was planned for the 2006 transmission season. However, shortages of trained personnel and other institutional delays meant that spraying could not begin until the third week of June, by which time the epidemic had peaked (and densities of vector mosquitoes had presumably begun to fall). By July 17, <50% of the targeted structures had been sprayed. In the future, careful targeting of spraying to areas of highest epidemic risk might lead to more timely completion of spraying activities. It might also be benefi cial to create special spray teams that can respond quickly to specifi c alerts. Recent experiences in Kabale have underlined the potential value of simple monitoring tools for early detection of epidemics but have also shown potential barriers to effective epidemic control. Our fi ndings highlight the need to build systems that improve routine collection of data on parasitologically confi rmed cases of malaria and allow rapid investigation of anomalies in incoming clinical data. It is equally important to develop procedures that translate early warning information into timely decisions concerning which epidemic control measures to use and how best to target them (8). Without these procedures, the value of early detection will be seriously undermined.