flubase {flubase} | R Documentation |
This function estimates the mortality (or other indicator) baseline free of
influenza epidemics for one or more than one time series (age groups, gender, regions, etc)
using one of four methods made available in the package, implemented respectively in
baseRM
, baseSA
, baseIt_RM
and baseIt_SA
.
flubase(dat, groups, per, pe = 0, method, indicator = "mortality", g_label)
dat |
a data.frame with all the variables needed: dados$group indicating the group where each time series
belongs, dados$year indicating the civil year, dados$todeath is
the time unit index that can be week or month, dados$nod are the number of
deaths observed at that time and dados$epi is an indicator variable of the epidemic period (dados$epi=1 if
the week or month belongs to the epidemic period and dados$epi =0 otherwise). |
groups |
number of groups considered, e.g. number of age groups, regions, etc. |
per |
per =52 for weekly data or per =12 for monthly data |
pe |
pe = 0 if the user provides the epidemic periods in dados$epi and pe = 1
otherwise, in case which the function uses a fixed period from week 48 to week 17 or
from month 12 to month 4. |
method |
the method used to estimate the baseline, method =c("nrm","nsa","irm","isa").
method=nrm : non iterative multiple regression, method =nsa: non iterative seasonal ARIMA,
method=irm : iterative cmultiple regression and method =isa: iterative seasonal ARIMA |
indicator |
a text string indicating the name of the indicator. By default indicator ="mortality" |
g_label |
a vector string of length the number of groups, containing the labels for the
groups, e.g. g_label =c("male","female") or g_label =c("0-14 yrs","15-44 yrs","45-64 yrs","65+ yrs") |
In order to use the methods irm
and isa
the time series must initiate in the week 27
or month 7 and end at week 26 or month 6 of the next year.
The function will return a list flubase
, with the original data set plus
flubase$beta0 |
containing the mortality (or other indicator) baseline of without the effect of influenza epidemics |
flubase$beta_up |
containg the upper 95 CI for the baseline |
flubase$da |
a dummy variable indicating the periods with excess deaths |
flubase$ex |
the excess deaths |
This function delivers the following outputs: a text file with the original data set plus four new variables,
the baseline (base0), the upper 95 confidence limit (dados$base_up
), the periods whith
excess deaths (dados$da
) and the excess deaths, estimated for each week or month (dados$ex
); a time series
graph for each group (age group, gender, etc) where
the blue line represents the observed mortality, the black line the estimated baseline, the red line the upper
95 confidence limit of the baseline, the gray rectangles the epidemic periods and the yellow rectangles
the periods with excess deaths.
Nunes B, Natario I and Carvalho L.
Nunes B, Natario I, Carvalho L. Time series methods for obtaining excess mortality. Submitted to Statistical Methods in Medical Research (2009).
Serfling RE Methods for Current Statistical Analysis of Excess Mortality Pneumonia-Influenza Deaths Public Heath Reports 1963; 78 6:494 506.
K. Choi and S.B. Thacker An evaluation of influenza mortality surveillance 1962-1979. American Journal of Epidemiology 1981; 113 3: 215 216.
Lui K-J and Kendal A.P. Impact of influenza epidemics on mortality in the united states from October 1972 to May 1985. American Journal of Public Health 1987; 77(6):712 716.
L. Simonsen, M.J. Clarke, D. Williamson, D.F. Stroup, N.H. Arden and L.B. Schonberger The impact of influenza vaccination on seasonal mortality in the US elderly population. American Journal of Public Health 1997; 87(12):1994 1950.
baseIt_RM
, baseSA
, baseIt_SA
,
baseRM