We estimated secondary attack rates, based on the proportion of household contacts in whom influenza developed (as determined by RT-PCR assay), as well as rates of acute respiratory illness and influenza-like illness. We calculated 95% confidence intervals for the crude secondary attack rates using a cluster bootstrap method with 1000 resamples.15 Households in which one or more household contacts had RT-PCR–confirmed influenza at the baseline home visit (i.e., households with one or more potential co-index patients) were excluded from the analysis of secondary attack rates.
We calculated standardized daily scores for three groups of signs and symptoms — systemic signs and symptoms (temperature ≥37.8°C, headache, and myalgia), upper respiratory symptoms (sore throat and runny nose), and lower respiratory symptoms (cough and phlegm) — by adding up the total number of signs and symptoms that were present and dividing by the highest possible score (3, 2, and 2, respectively).16 (link),17 (link) We plotted average symptom scores according to the time since the onset of acute respiratory illness, which was defined as the first day when the subject reported at least two of the seven signs or symptoms listed above.16 (link),18 (link)We defined the serial interval as the time between the onset of illness in an index patient and the onset of illness in a household contact. We estimated serial interval distributions on the basis of an underlying Weibull distribution, using methods that have been described previously.18 (link) We estimated 95% confidence intervals for the mean serial interval, using a parametric bootstrap approach with 1000 resamples.18 (link),19 We estimated geometric mean antibody titers and used Wilcoxon signed-rank tests to compare groups. Antibody titers below the lower limit of 1:10 were estimated to be 1:5 for calculations of geometric mean titers. Statistical analyses were performed with the use of R software, version 2.8.1 (R Development Core Team). Raw data from the study and R syntax to permit additional statistical analyses are available by contacting the corresponding author.
We calculated standardized daily scores for three groups of signs and symptoms — systemic signs and symptoms (temperature ≥37.8°C, headache, and myalgia), upper respiratory symptoms (sore throat and runny nose), and lower respiratory symptoms (cough and phlegm) — by adding up the total number of signs and symptoms that were present and dividing by the highest possible score (3, 2, and 2, respectively).16 (link),17 (link) We plotted average symptom scores according to the time since the onset of acute respiratory illness, which was defined as the first day when the subject reported at least two of the seven signs or symptoms listed above.16 (link),18 (link)We defined the serial interval as the time between the onset of illness in an index patient and the onset of illness in a household contact. We estimated serial interval distributions on the basis of an underlying Weibull distribution, using methods that have been described previously.18 (link) We estimated 95% confidence intervals for the mean serial interval, using a parametric bootstrap approach with 1000 resamples.18 (link),19 We estimated geometric mean antibody titers and used Wilcoxon signed-rank tests to compare groups. Antibody titers below the lower limit of 1:10 were estimated to be 1:5 for calculations of geometric mean titers. Statistical analyses were performed with the use of R software, version 2.8.1 (R Development Core Team). Raw data from the study and R syntax to permit additional statistical analyses are available by contacting the corresponding author.
