Laboratory policies on testing for rotavirus affect surveillance data. PHLS East Epidemiology and Virology...

The effect of different laboratory testing policies on the surveillance of rotavirus was assessed in eight laboratories between 1995 and 1998. In 1995, five laboratories tested all faecal specimens from children aged 5 years and under all year, two tested all specimens from children aged 4 years and under all year, and one tested all specimens from children aged 3 years and under between November and May only. Five laboratories changed their testing policy between 1995 and 1998. By 1998, three tested all specimens from children aged 5 years and under all year and two from the same age group during the 'season' only. Three laboratories had unique policies: one tested all specimens from children aged 2 years and under between January and June, one tested all specimens from children aged 4 years and under all year, and one tested specimens only on clinical request. The onset date of the rotavirus infection 'season' as determined by retrospective scrutiny of reported cases varied by up to 15 weeks between laboratories, starting as early as week 45 (November) and as late as week 13 (March). Laboratories with more restrictive testing policies yielded fewer reports of rotavirus and changes in policy within a particular laboratory affected the number of reports. Temporal and geographic trends were visible, even within the relatively small area covered by this study, and showed how laboratory testing policies affect surveillance data.