Invasive meningococcal disease (IMD) completeness of reporting has never been assessed in New York City (NYC). We conducted a capture–recapture study to assess completeness of reporting, comparing IMD reports made to the NYC Department of Health and Mental Hygiene (DOHMH) and records identified in the New York State hospital discharge database [Statewide Planning and Research Cooperative System (SPARCS)] by ICD-9 codes from 1989 to 2010. Reporting completeness estimates were calculated for the entire study period, and stratified by year, age group, clinical syndrome, and reporting system. A chart review of hospital medical records from 2008 to 2010 was conducted to validate hospital coding and to adjust completeness estimates. Overall, 2194 unique patients were identified from DOHMH (n = 1300) and SPARCS (n = 1525); 631 (29%) were present in both. Completeness of IMD reporting was 41% [95% confidence interval (CI) 40–43]. Differences in completeness were found by age, clinical syndrome, and reporting system. The chart review found 33% of hospital records from 2008 to 2010 had no documentation of IMD. Removal of those records improved completeness of reporting to 51% (95% CI 49–53). Our data showed a low concordance between what is reported to DOHMH and what is coded by hospitals as IMD. Additional guidance to clinicians on IMD reporting criteria may improve completeness of IMD reporting.

We evaluated the incidence rate of culture-positive central nervous system tuberculosis (CNS TB) in France in 2007 and its time trend between 1990 and 2007. We used a capture–recapture analysis by using data recorded in 2007 by the mandatory notification system and the national network of the National Reference Centre (NRC). The 2007 sensitivity of the NRC was 79·4%. The previous sensitivity for 2000 (75·6%) and that for 2007 yielded a pooled estimate of 77·4% (95% confidence interval 64·8–88·0), which was used to extrapolate the number of culture-positive CNS TB cases from those reported in four surveys (1990, 1995, 2000, 2007). The extrapolated number of culture-positive CNS TB cases fell from 90 to 35 between 1990 and 2007, and the extrapolated incidence rates fell from 1·6 to 0·55 cases/million (P < 0·001). This favourable trend should be closely monitored following the change of the BCG vaccination policy in 2007.

In countries maintaining national hepatitis C virus (HCV) surveillance systems, a substantial proportion of individuals report no risk factors for infection. Our goal was to estimate the proportion of diagnosed HCV antibody-positive persons in Scotland (1991–2010) who probably acquired infection through injecting drug use (IDU), by combining data on IDU risk from four linked data sources using log-linear capture–recapture methods. Of 25 521 HCV-diagnosed individuals, 14 836 (58%) reported IDU risk with their HCV diagnosis. Log-linear modelling estimated a further 2484 HCV-diagnosed individuals with IDU risk, giving an estimated prevalence of 83. Stratified analyses indicated variation across birth cohort, with estimated prevalence as low as 49% in persons born before 1960 and greater than 90% for those born since 1960. These findings provide public-health professionals with a more complete profile of Scotland's HCV-infected population in terms of transmission route, which is essential for targeting educational, prevention and treatment interventions.