OBJECTIVES To assess the prevalence of varicella zoster virus (VZV) antibodies in Flemish (Belgian) healthcare workers, to investigate the association between seronegativity and selected variables, and to assess the reliability of recall about disease as a predictor of immunity. METHODS A seroprevalence study of VZV antibodies (IgG) was conducted among a systematic sample of 4923 employees in various professional groups, employed in 22 hospitals in Flanders and Brussels (Belgium). Information about sex, age, department, job, and years of employment, the country of origin, and history of varicella was obtained. The presence of VZV antibodies was investigated with the enzyme linked immunosorbent assay (ELISA), Enzygnost anti VZV / IgG (Dade Behring, Marburg, Germany). Statistical analysis was performed by calculating prevalences and prevalence ratios (PRs) and their 95% confidence intervals (95% CIs). Sensitivity, specificity, and positive and negative predictive values (PPV and NPV) of the recalled history were determined. RESULTS The prevalence of VZV seropositivity in Flemish healthcare workers was 98.5% (95% CI 98.1 to 98.8). Seronegativity was significantly associated with age and job, increasing with both older and younger age. The prevalence of seronegative workers was significantly less in nursing staff than non-nursing staff. There was no significant difference for sex and years of employment. The PPV and NPV of recalled history were 98.9% and 3.4%. Sensitivity and specificity were 83% and 38.9%. CONCLUSION The prevalence of VZV seropositivity was very high in this sample of Flemish healthcare workers. Because of this low overall susceptibility, VZV infection seems not to be an important occupational risk among healthcare workers in Flanders. The increasing seronegativity above the age of 45 is possibly due to a loss of detectable antibodies. A positive history of varicella was a good predictor of immunity, but a negative history had no value as a predictor of susceptibility in adults.