msSPChelpR (version 0.9.1)

ir_crosstab_byfutime: Calculate crude incidence rates and cross-tabulate results by break variables; cumulative FU-times as are used as xbreak_var

Description

Calculate crude incidence rates and cross-tabulate results by break variables; cumulative FU-times as are used as xbreak_var

Usage

ir_crosstab_byfutime(
  df,
  dattype = NULL,
  count_var,
  futime_breaks = c(0, 0.5, 1, 5, 10, Inf),
  ybreak_vars,
  collapse_ci = FALSE,
  add_total = "no",
  futime_var = NULL,
  alpha = 0.05
)

Value

df

Arguments

df

dataframe in wide format

dattype

can be "zfkd" or "seer" or NULL. Will set default variable names if dattype is "seer" or "zfkd". Default is NULL.

count_var

variable to be counted as observed case. Should be 1 for case to be counted.

futime_breaks

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).

ybreak_vars

variables from df by which rates should be stratified in rows of result df. Multiple variables will result in appended rows in result df. y_break_vars is required.

collapse_ci

If TRUE upper and lower confidence interval will be collapsed into one column separated by "-". Default is FALSE.

add_total

option to add a row of totals. Can be either "no" for not adding such a row or "top" or "bottom" for adding it at the first or last row. Default is "no".

futime_var

variable in df that contains follow-up time per person (in years). Default is set if dattype is given.

alpha

significance level for confidence interval calculations. Default is alpha = 0.05 which will give 95 percent confidence intervals.

Examples

#load sample data
data("us_second_cancer")

#prep step - make wide data as this is the required format
usdata_wide <- us_second_cancer %>%
                    #only use sample
                    dplyr::filter(as.numeric(fake_id) < 200000) %>%
                    msSPChelpR::reshape_wide_tidyr(case_id_var = "fake_id", 
                    time_id_var = "SEQ_NUM", timevar_max = 2)
                    
#prep step - calculate p_spc variable
usdata_wide <- usdata_wide %>%
                 dplyr::mutate(p_spc = dplyr::case_when(is.na(t_site_icd.2)   ~ "No SPC",
                                                       !is.na(t_site_icd.2)   ~ "SPC developed",
                                                       TRUE ~ NA_character_)) %>%
                 dplyr::mutate(count_spc = dplyr::case_when(is.na(t_site_icd.2)   ~ 1,
                                                              TRUE ~ 0))
                                                              
#prep step - create patient status variable
usdata_wide <- usdata_wide %>%
                  msSPChelpR::pat_status(., fu_end = "2017-12-31", dattype = "seer",
                                         status_var = "p_status", life_var = "p_alive.1",
                                         birthdat_var = "datebirth.1", lifedat_var = "datedeath.1")
 
#now we can run the function
usdata_wide <- usdata_wide %>%
                 msSPChelpR::calc_futime(., 
                        futime_var_new = "p_futimeyrs", 
                        fu_end = "2017-12-31",
                        dattype = "seer", 
                        time_unit = "years",
                        status_var = "p_status",
                        lifedat_var = "datedeath.1", 
                        fcdat_var = "t_datediag.1", 
                        spcdat_var = "t_datediag.2")
                    
#for example, you can calculate incidence and summarize by sex and registry
msSPChelpR::ir_crosstab_byfutime(usdata_wide,
      dattype = "seer",
      count_var = "count_spc",
      futime_breaks = c(0, .5, 1, 5, 10, Inf),
      ybreak_vars = c("sex.1", "registry.1"),
      collapse_ci = FALSE,
      add_total = "no",
      futime_var = "p_futimeyrs",
      alpha = 0.05)