Objectives: To assess the validity of self-reported hepatitis B virus vaccination status in young heroin users.
Study design: Cross-sectional study among 949 street-recruited young injection heroin users (IHUs) and non-injection heroin users (NIHUs) in Madrid, Barcelona and Seville. Face-to-face interviews and dried blood spot tested for anti-HBc and anti-HBs. The validity of self-reported vaccination status was assessed comparing with the serological status. The percentage of agreement with the kappa (kappa) statistic and the positive predictive value were calculated.
Results: The percentage of agreement between self-reported and serologic vaccination status was 51.9% overall, with little difference by city (53.1% in Barcelona, 49.0% in Madrid and 51.5% in Seville) or between IHUs (51.3%) and NIHUs (53.0%). All the kappa scores were lower than 0.1. The positive predictive value of self-reports was less than 25% in all categories analysed, except in the city of Barcelona (37%). Among those who indicated that they had been vaccinated, 31% were actually susceptible.
Conclusions: In areas with a high prevalence of infection and rising coverage of vaccination the policy "don't ask, take a blood sample, give a dose of vaccine and try to schedule another visit" should be recommended in clinical practice.