An increasing number of acute gastroenteritis outbreaks have been reported in the Netherlands since October 2004 to the Rijksinstituut voor Volksgezondheid en Milieu (RIVM, the Dutch National Institute of Public Health and the Environment) (Table). The early onset of the ‘winter vomiting disease’ season and the high number of reports are unusual. Outbreaks have been reported from different settings, most of which are institutional. So far, all of the outbreaks for which the diagnostic evaluation has been completed are caused by norovirus. This situation may be indicative of a wider trend as several countries have reported higher incidences recently through the global electronic reporting system ProMED-mail (http://www.promedmail.org).

Table. Number of norovirus outbreaks reported in the Netherlands in the winter seasons from 2000/2001 to 2004/2005

**: Number of norovirus outbreaks reported in the first 2 weeks of December. An additional five outbreaks are under investigation.

We would like to share this observation, because we suspect a repeat of the situation in 2002. In that year, the Food-borne viruses in Europe network (FBVE, http://www.eufoodborneviruses.co.uk/) saw a sharp increase in the number of norovirus outbreaks across Europe, and an increase was also reported in the United States. This had a major impact on hospitals and other settings such as nursing homes and cruise ships. The large increase in 2002 was associated with the introduction of a new variant norovirus within the GGII.4 genotype. This virus was first detected early in 2002, and had replaced the resident virus population by mid-summer in all the countries in Europe that were participating in the Food-borne Viruses in Europe network [1,2]. In the United Kingdom, the cost of the 2002 epidemic was calculated to be approximately US$184 million [3].

In the Netherlands outbreaks analysed so far in 2004, another new lineage (GGII.4-2004) within the GGII.4 genotype has been found. This variant is distinct from the 2002 variant strain (GGII.4-2002). Since the beginning of August 2004, 71 norovirus outbreaks have been reported in the Netherlands. Of these, viruses from 44 outbreaks were characterised by sequence analysis, and all 44 belong to the new GGII.4 lineage.

This variant has already been highly active in Australia during the 2004 southern hemisphere winter season (personal communication, Michael Lyon, Public Health Virology Laboratory, Queensland Health Scientific Services, Brisbane, Queensland, Australia, 2004). It caused many outbreaks in different settings and has now almost completely disappeared in the southern hemisphere with the onset of warmer weather.

Since the outbreak season for norovirus in the Netherlands normally starts in December and peaks in January, we believe that a warning that a worldwide increase of outbreaks comparable to 2002 might be on its way is appropriate.

Although data analysis needs to be finalised, we have indications from the FBVE surveillance that GGII.4 is more commonly associated with outbreaks in institutional settings than other norovirus variants; this suggests that the norovirus GGII.4 genotype has properties facilitating transmission, and thereby has the propensity to cause epidemics.

We are continuing to monitor the situation in Europe and are studying the difference in virulence between strains, biological background of the mechanism for its rapid dissemination, and insight into the micro-evolution of noroviruses. Details on the genetic background of these variant noroviruses can be obtained by sending an email to fbve@rivm.nl. We have used the polymerase gene primers for monitoring purposes, and sequence properties are given below. The FBVE network will monitor noroviruses as part of the activities in the EU-funded DIVINE project. We would also be interested to hear from parties outside the participating countries.

Prevention
There are some protocols for healthcare settings [4], hotels and cruise ships available in the literature. Guidelines in these protocols are partly evidence-based and partly common sense, and the effectiveness of some of these measures is subject to debate. Therefore, controlled intervention studies are needed in order to apply evidence-based practice during outbreaks in institutional settings, especially nursing homes and homes for the elderly. Until results of such studies are available, their effectiveness in controlling outbreaks is not clear for all settings.

With these caveats, the following common prevention measures are recommended:

Figure. Sequence alignment of norovirus GGII.4 lineages. GGII.4 cons is the consensus sequence of strains prevalent before 2002, GGII.4 2002 is the consensus sequence of the strain that was dominant in the 2002/2003 winter season, GGII.4 2004 is the consensus sequence of the strain that has become dominant during 2004. The sequence is from the RNA dependent RNA polymerase gene, the region upstream of the conserved YGDD motif. Eleven informative positions in the alignment have been highlighted with an asterisk above the sequence. In these positions one sequence is different from the other two.