Investigation of the cluster has been carried out by Public Health Surveillance of Flanders, in collaboration with the Scientific Institute of Public Health (IPH). Data have been collected through school health services in Antwerp and by contact tracing (questioning parents and doctors who treated the patients). The investigation is still ongoing and information presented in this paper is preliminary.

To date, at least 15 other clinical measles cases have been identified in this cluster, bringing the total to 23 cases. Many more cases are believed to have gone unreported, since measles is not a disease with mandatory notification in Belgium, and some doctors refused to collaborate in the investigation. Additionally, not all ill children were taken to a doctor, especially when several children within a family fell ill.

The cases occurred in children aged 2 to 17 years. According to the information provided by general practitioners (GPs), nine cases (39%) had received at least one dose of measles vaccine. The index case of the cluster was thought to be a 17-year-old American studying in the United Kingdom, who had visited Jewish relatives in Antwerp and was diagnosed as a clinical measles case by a Belgian GP at the time of his visit. Transmission of the virus occurred within families, schools and neighborhoods. The virus circulating in Antwerp (D4) has identical genotype to the strain responsible for the outbreak in Jewish communities in the United Kingdom [1].

The investigation of this cluster led to identifying a previously unreported cluster of at least 10 clinical measles cases in the same Jewish communities, starting in August. Samples for confirmation of the diagnosis and genotyping were not collected at the time, but a link with the United Kingdom was established. Belgian children from a Jewish community participated in a summer camp in London for Jewish schools from Antwerp, London and Manchester. Some English children at the camp had fever at the time. Two Belgian children developed fever and rash upon returning home. They spread the infection to at least eight other children. At the time, the diagnosis of measles was made for all the cases identified so far (n=10), based on clinical signs, by three different doctors. The 10 cases of this first cluster were aged 8 months to 12 years. All were unvaccinated. Information on hospitalisation and complications for both clusters is not yet available.

The age distribution for all the measles cases identified so far since mid-August (n=33) is presented in Figure 1. Figure 2 shows the epidemiological curve for both clusters. The date of onset of rash is known for 16 cases (48%) and uncertain for 17 cases.
The last reported case fell ill on 31 October, although more recent cases might have been unreported.

So far, the virus has not spread to the wider community. Vaccine coverage for the first dose of measles vaccine (MMR) in Antwerp is 94% (vaccine coverage study in children aged 18-24 months, 2005) [2]. Separate information on measles vaccine coverage in Jewish communities in Belgium is not available. Apparently, reasons for non-vaccination in these communities are diverse (lack of knowledge about vaccination, misconceptions about possible side effects, etc.).

Information on the outbreak and on the importance of vaccination will be distributed to the Jewish communities through local papers (in Yiddish and in Flemish), with the help of Jewish doctors, rabbis and Jewish organisations. Letters have already been sent to general practitioners and pediatricians in Antwerp to inform them of the outbreak and to invite them to perform laboratory testing (on saliva) for confirmation of the diagnosis for suspected measles cases and to report the cases to the division of control of infectious diseases of Public Health Surveillance of Flanders.

Notification of measles in Belgium is only mandatory within the framework of control of infectious diseases in school medicine. A voluntary surveillance network of 40% of Belgian pediatricians also reports measles cases. It is reassuring that both sources reported some of the cases of the second (larger) cluster, which allowed for the confirmation of the outbreak. However, collection of information would probably be easier if notification of measles cases was mandatory.

Acknowledgements
Thanks to Dr. J. Kremer (WHO Reference Laboratory for Measles and Rubella in Luxemburg) for genotyping the virus, and to the different partners who collaborate to the collection of data: the school health services from Antwerp (CLB), Professor Dr Pierre Van Damme (UA), pediatricians and GPs, and the parents of the children.