R/sir_byfutime.R
sir_byfutime.RdCalculate standardized incidence ratios with custom grouping variables stratified by follow-up time
sir_byfutime(
df,
dattype = NULL,
ybreak_vars = "none",
xbreak_var = "none",
futime_breaks = c(0, 0.5, 1, 5, 10, Inf),
count_var,
refrates_df = rates,
calc_total_row = TRUE,
calc_total_fu = TRUE,
region_var = NULL,
age_var = NULL,
sex_var = NULL,
year_var = NULL,
race_var = NULL,
site_var = NULL,
futime_var = NULL,
expect_missing_refstrata_df = NULL,
alpha = 0.05,
quiet = FALSE
)dataframe in wide format
can be "zfkd" or "seer" or NULL. Will set default variable names if dattype is "seer" or "zfkd". Default is NULL.
variables from df by which SIRs should be stratified in result df. Multiple variables will result in appended rows in result df. Careful: do not chose any variables that are dependent on occurrence of count_var (e.g. Histology of second cancer). If y_break_vars = "none", no stratification is performed. Default is "none".
One variable from df by which SIRs should be stratified as a second dimension in result df. This variable will be added as a second stratification dimension to ybreak_vars and all variables will be calculated for subpopulations of x and y combinations. Careful: do not chose any variables that are dependent on occurrence of count_var (e.g. Year of second cancer). If y_break_vars = "none", no stratification is performed. Default is "none".
vector that indicates split points for follow-up time groups (in years) that will be used as xbreak_var. Default is c(0, .5, 1, 5, 10, Inf) that will result in 5 groups (up to 6 months, 6-12 months, 1-5 years, 5-10 years, 10+ years). If you don't want to split by follow-up time, use futime_breaks = "none".
variable to be counted as observed case. Cases are usually the second cancers. Should be 1 for case to be counted.
df where reference rate from general population are defined. It is assumed that refrates_df has the columns "region" for region, "sex" for biological sex, "age" for age-groups (can be single ages or 5-year brackets), "year" for time period (can be single year or 5-year brackets), "incidence_crude_rate" for incidence rate in the respective age/sex/year cohort.The variable "race" is additionally required if the option "race_var" is used. refrates_df must use the same category coding of age, sex, region, year and t_site as age_var, sex_var, region_var, year_var and site_var.
option to calculate a row of totals. Can be either FALSE for not adding such a row or TRUE for adding it at the first row. Default is TRUE.
option to calculate totals for follow-up time. Can be either FALSE for not adding such a column or TRUE for adding. Default is TRUE.
variable in df that contains information on region where case was incident. Default is set if dattype is given.
variable in df that contains information on age-group. Default is set if dattype is given.
variable in df that contains information on sex. Default is set if dattype is given.
variable in df that contains information on year or year-period when case was incident. Default is set if dattype is given.
optional argument, if SIR should be calculated stratified by race. If you want to use this option, provide variable name of df that contains race information. If race_var is provided refrates_df needs to contain the variable "race".
variable in df that contains information on site or subsite (e.g. ICD code, SEER site code or others that matches t_site in refrates_df) of case diagnosis. Cases are usually the second cancers. Default is set if dattype is given.
variable in df that contains follow-up time per person between date of first cancer and any of death, date of event (case), end of FU date (in years; whatever event comes first). Default is set if dattype is given.
optional argument, if strata with missing refrates are expected, because incidence rates of value 0 are not explicit, but missing from refrates_df. It is assumed that expect_missing_refstrata_df is a data.frame has the columns "region" for region, "sex" for biological sex, "age" for age-groups (can be single ages or 5-year brackets), "year" for time period (can be single year or 5-year brackets), and "t_site" for The variable "race" is additionally required if the option "race_var" is used. refrates_df must use the same category coding of age, sex, region, year and t_site as age_var, sex_var, region_var, year_var and site_var.
significance level for confidence interval calculations. Default is alpha = 0.05 which will give 95 percent confidence intervals.
If TRUE, warnings and messages will be suppressed. Default is FALSE.
#There are various preparation steps required, before you can run this function.
#Please refer to the Introduction vignette to see how to prepare your data
if (FALSE) {
usdata_wide %>%
sir_byfutime(
dattype = "seer",
ybreak_vars = c("race.1", "t_dco.1"),
xbreak_var = "none",
futime_breaks = c(0, 1/12, 2/12, 1, 5, 10, Inf),
count_var = "count_spc",
refrates_df = us_refrates_icd2,
calc_total_row = TRUE,
calc_total_fu = TRUE,
region_var = "registry.1",
age_var = "fc_agegroup.1",
sex_var = "sex.1",
year_var = "t_yeardiag.1",
site_var = "t_site_icd.1", #using grouping by second cancer incidence
futime_var = "p_futimeyrs",
alpha = 0.05)
}