Estimating seroprevalence of vaccine-preventable infections: is it worth standardizing the serological outcomes to adjust for different assays and laboratories?

Kafatos, G.; Andrews, N.; McConway, K.; Anastassopoulou, C.; Barbara, C.; De Ory, F.; Johansen, K.; Mossong, J.; Prosenc, K.; Vranckx, R.; Nardone, A.; Pebody, R. and Farrington, P. (2015). Estimating seroprevalence of vaccine-preventable infections: is it worth standardizing the serological outcomes to adjust for different assays and laboratories? Epidemiology and Infection, 143(11) pp. 2269–2278.

DOI: https://doi.org/10.1017/S095026881400301X

Abstract

The aim of the European Sero-Epidemiology Network 2 (ESEN2) project was to estimate age-specific seroprevalence for a number of vaccine-preventable diseases in Europe. To achieve this serosurveys were collected by 22 national laboratories. To adjust for a variety of laboratory methods and assays, all quantitative results were transformed to a reference laboratory’s units and were then classified as positive or negative to obtain age-specific seroprevalence. The aim of this study was to assess the value of standardization by comparing the crude and standardized seroprevalence estimates. Seroprevalence was estimated for measles, mumps, rubella, diphtheria, varicella zoster and hepatitis A virus (HAV) and compared before and after serological results had been standardized. The results showed that if no such adjustment had taken place, seroprevalence would have differed by an average of 3·2% (95% bootstrap interval 2·9–3·6) although this percentage varied substantially by antigen. These differences were as high as 16% for some serosurveys (HAV) which means that standardization could have a considerable impact on seroprevalence estimates and should be considered when comparing serosurveys performed in different laboratories using different assay methods.

Viewing alternatives

Metrics

Item Actions

Export

You can export this page using these formats Digital Object Identifier - DOI