Juan B Bellido-Blasco1, Francisco Pardo-Serrano2, Isabel Ballester-Rodríguez3, Alberto Arnedo-Pena3, M Dolores Tirado-Balaguer2, M Ángeles Romeu-García4, Ester Silvestre-Silvestre4, Noemí Meseguer-Ferrer4, Concha Herrero-Carot4, Joan A Caylà-Buqueres5. 1. Sección de Epidemiología, Centro Salud Pública de Castellón, Castellón, España; Facultad de Medicina, Universidad Cardenal Herrera - CEU, Castellón, España; CIBER de Epidemiología y Salud Pública, CIBERESP, España. Electronic address: bellido_jua@gva.es. 2. Servicio de Microbiología, Hospital General de Castellón, Castellón, España. 3. Sección de Epidemiología, Centro Salud Pública de Castellón, Castellón, España; CIBER de Epidemiología y Salud Pública, CIBERESP, España. 4. Sección de Epidemiología, Centro Salud Pública de Castellón, Castellón, España. 5. Servicio de Epidemiología, Agència de Salut Pública de Barcelona - ASPB, Barcelona, España; CIBER de Epidemiología y Salud Pública, CIBERESP, España.
Abstract
INTRODUCTION: The influenza pandemic of 2009 had a great social impact. Many health resources were devoted to the care, prevention and surveillance of this disease. Epidemiological surveillance is based on the reporting of cases of influenza-like illness (ILI) and confirmed influenza cases. The objective was to estimate the true incidence of ILI during the influenza pandemic of 2009. METHODS: The capture-recapture method was applied during the month of highest influenza incidence in Castellón. Two notification systems were used: (i)electronic reporting of Notifiable Diseases (ND), and (ii)laboratory-based (LAB) data collection. Estimates were made by stratifying by age group and week. Independence coefficients were calculated for those strata. RESULTS: No dependence was found between stratification variables and the reporting system. A total of 7,181 ND cases and 524 LAB cases were identified, of which 211 were recorded in both systems. The estimated total of cases was 17,785 in a single month. In the study period, almost 4% of people in the area suffered flu symptoms (cumulative incidence), with 1% being affected each day (daily prevalence). The sensitivity of the ND system was 40%, i.e., the percentage of patients seeking primary care. CONCLUSIONS: To obtain an estimate of the actual incidence of influenza-like illness in the population during a pandemic period, the number of medical consultations should be multiplied by a factor of 2.5. This factor is lower than that estimated for periods without pandemic alert.
INTRODUCTION: The influenza pandemic of 2009 had a great social impact. Many health resources were devoted to the care, prevention and surveillance of this disease. Epidemiological surveillance is based on the reporting of cases of influenza-like illness (ILI) and confirmed influenza cases. The objective was to estimate the true incidence of ILI during the influenza pandemic of 2009. METHODS: The capture-recapture method was applied during the month of highest influenza incidence in Castellón. Two notification systems were used: (i)electronic reporting of Notifiable Diseases (ND), and (ii)laboratory-based (LAB) data collection. Estimates were made by stratifying by age group and week. Independence coefficients were calculated for those strata. RESULTS: No dependence was found between stratification variables and the reporting system. A total of 7,181 ND cases and 524 LAB cases were identified, of which 211 were recorded in both systems. The estimated total of cases was 17,785 in a single month. In the study period, almost 4% of people in the area suffered flu symptoms (cumulative incidence), with 1% being affected each day (daily prevalence). The sensitivity of the ND system was 40%, i.e., the percentage of patients seeking primary care. CONCLUSIONS: To obtain an estimate of the actual incidence of influenza-like illness in the population during a pandemic period, the number of medical consultations should be multiplied by a factor of 2.5. This factor is lower than that estimated for periods without pandemic alert.