Features of 2006 cholera outbreak in Pemba island, Tanzania

Since the first outbreak in 1978, Pemba experienced recurrent outbreaks in the period 1983-98, then annually since 2000, except for 2005. In 2006 the island Cholera Committee - composed by the Medical Core Unit of Chake hospital and the staff of both Zanzibar Water Authority (ZAWA) and Ministry of Education - worked in strict co-operation with the technicians of the Public Health Laboratory Ivo de Carneri (PHL-IdC) and the Department of Public Health-Microbiology-Virology (Milan) in order to follow the outbreak. Date of admission, sex, age-range (younger than 5 years of age and others) and living area for each patient admitted in the Primary Health Care Units (P) and District Hospitals (H) have been got from the registration forms; the demographic data for each affected area have been obtained by a population survey. Patients' rectal swabs and water sources used for drinking and/or household practices have been collected in and around the cholera treatment centres, respectively. At the PHL-IdC the samples have been analysed for the detection of V. cholerae, following the traditional protocols: the suspected colonies have been confirmed by serotyping with a polyvalent O1 antiserum. The 2006 outbreak started in the south-eastern cost (Mkoani District) on 13 March; later on, the outbreak moved further north, affecting Wete, M’weni and Chake Districts. This figure has confirmed that there are four high risk areas along the east coast, involved also during each of the previous outbreaks and characterized by particularly poor latrine coverage, overcrowding as well a limited access to safe water. As the 2006 outbreak ended on 31 October, a total of 464 cases including 10 deaths (case-fatality rate 2%) were reported from the cholera treatment centres in Pemba. The index case was a mobile fisherman, travelling between islands and Tanzania mainland: fishermen have always played a critical role in spreading the disease. High peaks have reported in heavy rainy season (March-June), accounting for 71% of all the cases and showing a weekly case-fatality rate up to 25%. The overall incidence in Pemba island was 1‰ and much lower with respect to the affected areas, where it ranged from 8‰ (Kojani island in Wete) to 61‰ (Shamiani island in Mkoani). There were not differences between males and females; the incidence was higher among people older than 5 years, except for Kojani (25‰ children vs. 6‰ other age groups). Available data are most likely an underestimate of the real situation, as they do not account for outpatients diagnosed with cholera: surveillance activities remain an important challenge in Pemba. The PHL-IdC confirmed 65% samples positive to V. cholerae O1 on a total of 109 analysed specimens. In 6/9 affected areas, 45 on 56 (80%) water samples were found positive to Vibrio cholerae. The majority of the water is provided by unprotected shallow wells and springs, followed by piped water systems: the first are easily contaminated through ropes and buckets, the last due to poor maintenance and leakages. In south-eastern cost the water sampling was done both before and after chlorination of the sources: the analysis results showed that direct chlorination at the source was not a proper treatment, so that ZAWA operators pushed the population to prepare a stock disinfectant solution for the water buckets used at household level. As evaluated by the WHO Cholera Global Task Force, the response provided to the 2006 outbreak has been efficient and well organized compared to outbreaks before 2002. Nonetheless, it is still a challenge to ensure proper surveillance, health education activities, and proper environmental management (safe water and proper excreta disposal). A 3 phase proposal has been planned for a new approach for cholera control in Pemba: it includes improved surveillance for cholera and other epidemic diarrhoeal diseases, validation of the rapid immunochromatographic diagnostic test for cholera (CrystalV), and a mass vaccination campaign using oral cholera vaccines as an additional means for control.