Using self-testing to augment syndromic surveillance--United Kingdom, December 2003-January 2004

• Published in: MMWR. Morbidity and mortality weekly report 2005, Vol.54 (33), p.S186
• Main Authors: Cooper, Duncan L, Smith, G, Loveridge, P, Chinemana, F, Gerard, E, Joseph, C, Baker, M, Mant, D, Watson, J, Griffiths, R, Zambon, M
• Format: Newsletterarticle
• Language: English
• Subjects: Diseases; Epidemics; Epidemiologic methods; Influenza; Information management; Methods; Reporting; United Kingdom
• ISSN: 0149-2195; 1545-861X

Introduction: Effective community surveillance is needed for the rapid identification of outbreaks of serious illness. In the United Kingdom, surveillance of calls to NHS Direct, a national telephone health helpline, are used as a method of rapid outbreak detection. A limitation of this syndromic surveillance approach is the lack of laboratory confirmation of diagnosis after surveillance signals have been generated. Objective: This report presents a pilot study to investigate the feasibility of virologic sampling conducted by NHS Direct callers. Methods: During December 2003--January 2004, NHS Direct nurses in two regions of England were asked to recruit NHS Direct callers aged >12 years who had reported cold or influenza symptoms (i.e., "cold/flu" callers). Persons who agreed to participate in the study were mailed a specimen kit. Callers were asked to take a swab from each nostril and return the swabs by mail to the United Kingdom national influenza reference laboratory. Swabs were tested by multiplex polymerase chain reaction (PCR) for influenza viruses and, if identified as positive, were cultured for viable virus isolation. Results: During the study period, 686 cold/flu callers were eligible for the study. Although 67 of these were recruited for the study, determining how many cold/flu callers were asked to participate but refused was impossible. Of the 67 specimen kits sent, 36 (54%) were returned to the laboratory. The mean time between the call to NHS Direct and laboratory analysis of the specimen was 7.5 days. This period was shorter for positive samples (mean time: 6.12 days) than negative samples (mean time: 7.9 days), although the difference was not significant (p = 0.13). Eight specimens (22%) were positive on PCR for influenza virus. Five were antigenically characterized as Fuijian/411/2002-like influenza A H3N2. Higher positivity rates might have been achieved if the sampling study had begun earlier in the year before the peak of the influenza season. Conclusion: This study demonstrates the possibility of community-based clinical surveillance that does not require sampling by a health-care worker. This methodology will allow novel approaches to be developed to integrate syndromic surveillance with virologic sampling. Therefore, the rapid follow up of syndromic surveillance signals might provide confirmation of a specific common infection or provide evidence of a potentially more sinister cause.