Three components of the surveillance system were required: laboratory surveillance, sentinel surveillance and mapping of circulating strains.

The laboratory surveillance element was operational for the 2003-2004 winter norovirus season. This report is concerned with the laboratory data from that period. Sentinel surveillance and mapping of circulating strains are planned for the 2004-2005 season.

The objectives of the laboratory surveillance were to identify spatial clustering, the demographic characteristics of laboratory confirmed cases, and early detection of any abnormal seasonal increase in cases and trends. This surveillance remit does not include information on the setting of cases, as this will be included in the sentinel surveillance. Nor does it include the reporting of outbreaks, which is covered by the Miljökontoret (Environmental Health Protection Board) and the County Medical Officers.

The surveillance method is a voluntary, laboratory based system, using all 12 of Sweden’s norovirus testing laboratories. The case definition is a norovirus positive result from ELISA, polymerase chain reaction or electron microscopy.

Data from individual cases are sent weekly to Smittskyddinstitutet (SMI, Swedish Institute of Infectious Disease Control). The SMI aggregated data are also sent weekly to the county medical officers, infection control nurses and laboratories (Figure 1)

Figure 1. Data flow for laboratory surveillance of norovirus infections.

All 12 laboratories participated in the surveillance. From week 43 of 2003 to week 25 of 2004, the laboratories transmitted 99% of all their weekly reports to SMI. A total of 4776 patients were tested, 692 of whom tested positive for norovirus infection (14.5%). Peak norovirus activity was around week 9 of 2004 (Figure 2).

Determination of the number of patients tested for norovirus and the proportion of positive results, has the added value of acting as a crude check on laboratory methods. It may also indicate the possible presence of a new strain. Viruses are characterised both in local laboratories and at the SMI.

Figure 2. Laboratory confirmed norovirus cases by week of diagnosis, Sweden, October 2003-July 2004. (Data source: SMI)

The two main age groups affected were those under five and over 70 years of age. The overall distribution of cases by gender was 42% male, 57% female and 1% unknown. There was an equal gender distribution in children under five years. There are more female cases than male in those over 70 (Figure 3). This probably reflects the age and gender differences in the Swedish population (7.5% of women are >70 years of age, compared with 5.2% of men). The age, sex and spatial distribution of laboratory confirmed cases may not reflect the true distribution of norovirus infection in the population at large.

Figure 3. Laboratory confirmed cases of norovirus by age and sex in Sweden 2003-2004. (Data source: SMI)

In the 2003-2004 season, laboratory confirmed norovirus cases occurred in areas of high and low population density (Figure 4). This could reflect the distribution of the laboratories and reflect local interest in obtaining samples, as there was a noticeable geographical absence of cases reported from other areas with less access to laboratory norovirus diagnostic capacity.

Figure 4. Map of cumulative laboratory confirmed cases of norovirus related to density of population, Sweden, October 2003 – July 2004. (Data source: SMI)

The laboratory surveillance system was introduced after a thorough consultation process and feedback from the laboratories and was well supported by the public and private sectors. Sentinel surveillance and mapping of circulating strains will improve the quality of the data.

Acknowledgements: Solveig Andersson, Kasia Grabowska, Benn Sartorious, all participating laboratory technicians.