File S1 - The Effects of School Holidays on Transmission of Varicella Zoster Virus, England and Wales, 1967–2008

Supporting information, figures and tables. This file contains: Calculation of the deviance, Alternative TSIR modelling analyses, Additional analyses using the simple mass action model, and Figure S1-Figure S12. Figure S1, Weekly overall and age-specific GP consultation rates for chickenpox, England and Wales 1967–2008 (RCGP). Figure S2, Plot of the cumulative birth rate per 100,000 against the cumulative reported rate of chickenpox per 100,000, 1967–2008. Figure S3, Estimated percentage difference between the contact parameter (per fortnight) for chickenpox during termtime and that during A) all school holidays (fortnights 1, 8, 16–18 and 26) and B) summer holidays (fortnights 16–18), as estimated from the TSIR model. Figure S4, Fortnightly estimates of the contact parameter for chickenpox for 1967–76 (column 1), 1977–97 (column 2) and 1998–2008 (column 3), assuming α = 1. Figure S5, Assessment of model fit for chickenpox data, assuming that α = 1. A) Relationship between the RCGP chickenpox consultation rates and the fitted values from the regression for the periods 1967–76, 1977–1997 and 1998–2008 separately; B) Time series of RCGP data and predictions from difference equations for the periods 1967–76, 1977–97 and 1998–2008 separately; C) and D) as A) and B) but based on the full period 1967–2008. Figure S6, Fortnightly estimates of the contact parameter for chickenpox (1967–2008), for α = 0.717. Figure S7, Assessment of model fit for chickenpox data, using the full period 1967–2008 and estimating α as 0.717. A) Relationship between the RCGP chickenpox consultation rates and the fitted values from the regression; B) Relationship between the RCGP data and the values predicted by the difference equations using the estimated contact parameters. Figure S8, Estimated percentage reductions in the contact parameter (per fortnight) for chickenpox during school holidays, based on the full period 1967–2008. Figure S9, Fortnightly estimates of the contact parameter for chickenpox, based on the full time period 1967–2008 and assuming α = 1. Figure S10, Fortnightly estimates of the contact parameter for chickenpox based on the simple mass action model and RCGP consultation data, assuming that 20% of the population was susceptible to infection at the beginning of the time series. Figure S11, Fortnightly estimates of the contact parameter for chickenpox based on the simple mass action model and RCGP consultation data and assuming the proportion susceptible at the start of the time series to be 12% (the mean of the mean percentage susceptible during 1967–76, 1977–97 and 1998–2008); 13%. Figure S12, Percentage reduction in the contact parameter for chickenpox during school holidays, by year, 1967–2008 (simple mass action model), based on alternative assumptions about the dates of school holidays.
(DOCX)