Package 'Rcan'

Cancer Registry Data Analysis and Visualisation

Description

Tools for basic and advance cancer statistics and graphics. Groups individual data, merges registry data and population data, calculates age-specific rate, age-standardized rate, cumulative risk, estimated annual percentage rate with standards error. Creates graphics across variable and time, such as age-specific trends, bar chart and period-cohort trends.

Author(s)

Mathieu Laversanne [aut, cre], Jerome Vignat [aut], Cancer Surveillance Unit [cph]

Maintainer: Mathieu Laversanne <[email protected]>

References

https://publications.iarc.fr/_publications/media/download/3753/609d0d7711047dd76d7f3dbaa25d7f041fcd013e.pdf

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

data(ICD_group_GLOBOCAN)
data(data_individual_file)


#group individual data by 
#	5 year age group 
#	ICD grouping from dataframe ICD_group_GLOBOCAN
df_data_year <- csu_group_cases(data_individual_file,
  var_age="age",
  group_by=c("sex", "regcode", "reglabel"),
  df_ICD = ICD_group_GLOBOCAN,
  var_ICD  ="site",
  var_year = "doi")     

head(df_data_year)
# individual cases grouped by ICD and 5 years age group and year.
readline(prompt="Press [enter] to continue to merge cases and population)")	


data(data_population_file)	

df_data <- csu_merge_cases_pop(
	df_data_year, 
	data_population_file, 
	var_age = "age_group",
	var_cases = "cases",
	var_py = "pop",
	group_by = c("sex"))

head(df_data)
#Merge 5-years age grouped data with population by year (automatic) and sex
readline(prompt="Press [enter] to continue to merge cases and population)")	


data(csu_registry_data_2)
				  
# ASR with standard error with missing age.
df_asr <- csu_asr(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_age_group = c("registry_label"), 
                  var_st_err = "st_err")

df_asr[1:4,]	



# ASR with standard error with missing age.		  
readline(prompt="Press [enter] to continue to EAPC")


# EAPC with standard error		
	  
df_eapc <- csu_eapc(df_asr,
					"asr", "year",
					group_by=c("registry", "registry_label", "sex", "ethnic" ))
					
df_eapc[1:4,]	

# EAPC with standard error				  
readline(prompt="Press [enter] to continue to age specific graph")

data(csu_registry_data_1)

# plot age specific rate for 1 population.
df_colombia <- subset(csu_registry_data_1, registry_label=="Colombia, Cali")

csu_ageSpecific(df_colombia,
			plot_title = "Colombia, Liver, male")
				
# plot age specific rate for 1 population, and comparison with CI5XII data.
csu_ageSpecific(df_colombia,
				plot_title = "Colombia, Liver, male",
				CI5_comparison = "Liver")
			
# plot age specific rate for 4 population, legend at the bottom and comparison with CI5XII data.
csu_ageSpecific(csu_registry_data_1,
				group_by="registry_label",
				legend=csu_trend_legend(position="bottom", nrow = 2),
				plot_title = "Liver, male",
				CI5_comparison = 16)

---

csu_ageSpecific

Description

csu_ageSpecific calculate and plot Age-Specific Rate.

Usage

csu_ageSpecific(df_data,
		var_age="age",
        var_cases="cases",
        var_py="py",
        group_by = NULL, 
        missing_age = NULL,
        db_rate = 100000,
		logscale=FALSE,
        plot_title=NULL,
        legend=csu_trend_legend(),
        color_trend = NULL,
        CI5_comparison=NULL,
        var_rate="rate")

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_age	Age variable. Several format are accepted 1 "0-4" 0 2 "5-9" 5 3 "10-14" 10 ... ... ... 17 "80-84" 80 18 "85+" 85 Missing age value must be precise in the option missing_age. Last age group will always be considere without size (ie: 80+,85+, etc..).
var_cases	Number of event (cases, deaths, ...) variable.
var_py	Population year variable.
group_by	Variable to compare different age specific rate (sex, country, cancer ...). Only one variable can be chosen.
missing_age	Age value representing the missing age cases.
db_rate	The denominator population. Default is 100000.
logscale	Logical value: if TRUE Y-axis use logscale.
plot_title	Title of the plot.
legend	Legend option: see csu_trend_legend. Use only if group_by is not NULL.
color_trend	Vector of color for the trend. The color codes are hexadecimal (e.g. "#FF0000") or predefined R color names (e.g. "red").
CI5_comparison	Add a dotted line representing the CI5XII for a specific cancer. Value can be a cancer_label (example: "Liver"), or a cancer_code (example = 16). See csu_ci5_mean to get the list of possible values.
var_rate	Name of the age specific variable if a dataframe is return.

Details

This function calculate and plot the age specific rate. The group_by option allow to compare different population or cancer. The CI5_comparison option allow to compare with the CI5XII and therefore test the quality of the data. If the population data stops before 85+ (75+ for instance), the population data must be 0 when the population data is unknown so, the program can detect automatically the last age group (70+,75+,80+ or 85+) for population.

Value

Return a plot and a data.frame.

Author(s)

Mathieu Laversanne

References

https://publications.iarc.fr/_publications/media/download/3753/609d0d7711047dd76d7f3dbaa25d7f041fcd013e.pdf

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

data(csu_registry_data_1)
data(csu_registry_data_2)

# you can import your data from csv file using read.csv:
# mydata <-  read.csv("mydata.csv", sep=",")

# to select only 1 population.
test <- subset(csu_registry_data_1 , registry_label == "Colombia, Cali")


# plot age specific rate for 1 population.
csu_ageSpecific(test,
				plot_title = "Colombia, Liver, male")

# plot age specific rate for 1 population, and comparison with CI5XII data.
csu_ageSpecific(test,
				plot_title = "Colombia, Liver, male",
				CI5_comparison = "Liver")

# plot age specific rate for 4 population, 
# legend at the bottom and comparison with CI5XII data using cancer code.
csu_ageSpecific(
	csu_registry_data_1,
	group_by="registry_label",
	legend=csu_trend_legend(position="bottom", nrow = 1),
	plot_title = "Liver, male",
	CI5_comparison = 16
	)

	

# plot age specific rate for 4 population, legend at the right.
csu_ageSpecific(
	csu_registry_data_1,
	group_by="registry_label",
	legend=csu_trend_legend(
		position="right", right_space_margin = 6.5
	),
	plot_title = "Liver, male")	



# Plot embedded in a graphic device
pdf(paste0(tempdir(),"/test.pdf"),width = 11.692 , height =  8.267) 


csu_ageSpecific(
	csu_registry_data_1,
	group_by="registry_label",
	legend=csu_trend_legend(position="bottom", nrow = 2),
	plot_title = "Liver, male",
	CI5_comparison = 16)

plot.new()

csu_ageSpecific(
	csu_registry_data_1,
	group_by="registry_label",
	legend=csu_trend_legend(
		position="right", right_space_margin = 6.5
	),
	plot_title = "Liver, male")	



dev.off()

---

csu_ageSpecific_top

Description

csu_ageSpecific_top calculate and plot Age-Specific Rate for the top X cancer or top X population.

Usage

csu_ageSpecific_top(df_data,
           var_age="age",
       	   var_cases="cases",
           var_py="py",
           var_top, 
           group_by=NULL,
           missing_age=NULL,
           db_rate = 100000,
           logscale = FALSE,
           nb_top = 5,
           plot_title=NULL,
           plot_subtitle=NULL,
           var_color=NULL)

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_age	Age variable. Several format are accepted 1 "0-4" 0 2 "5-9" 5 3 "10-14" 10 ... ... ... 17 "80-84" 80 18 "85+" 85 Missing age value must be precise in the option missing_age. Last age group will always be considere without size (ie: 80+,85+, etc..).
var_cases	Number of event (cases, deaths, ...) variable.
var_py	Population year variable.
var_top	Cancer label variable or country/registry variable for example.
group_by	Variable to compare different age specific rate (sex, registry ...). Only one variable can be chosen. One graph will be produce per variable. If the variable is a factor, the labels associate to the variable will be the subtitle of the plot. See factor.
missing_age	Age value representing the missing age cases.
db_rate	The denominator population. Default is 100000.
logscale	Logical value: if TRUE Y-axis use logscale.
nb_top	Lowest Rank include. Default is 5.
plot_title	Title of the plot.
plot_subtitle	Subtitle of the plot. (For example, "Top 5 cancer").
var_color	Variable with a color associate to each cancer. The color codes are hexadecimal (e.g. "#FF0000") or predefined R color names (e.g. "red"). This allow to keep the same color coding for each graph.

Details

This function keep only the top X cancer and plot their age specific rate. The group_by option allow to compare different population. If the population data stops before 85+ (75+ for instance), the population data must be 0 when the population data is unknown so, the program can detect automatically the last age group (70+,75+,80+ or 85+) for population.

Value

Return plots and a data.frame.

Author(s)

Mathieu Laversanne

References

https://publications.iarc.fr/_publications/media/download/3753/609d0d7711047dd76d7f3dbaa25d7f041fcd013e.pdf

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

library(Rcan)
data("csu_CI5XII_data")

#get the id_code asssociate to id_label
#print(unique(csu_CI5XII_data[,c("id_label", "id_code")]),nrows = 1000)

#get the cancer code asssociate to cancer_label
#print(unique(csu_CI5XII_data[,c("cancer_label", "cancer_code")]),nrows = 1000)

#remove all cancers:
df_data <- subset(csu_CI5XII_data ,cancer_code < 62)
df_data$sex <- factor(df_data$sex, levels=c(1,2), labels=c("Male", "Female"))

#select Thailand changmai
df_data_1 <- subset(df_data, id_code==476400199)


# plot for Thailand Changmai
dt_result_1 <- 
	csu_ageSpecific_top(df_data_1,
		var_age="age", 
		var_cases="cases", 
		var_py="py",
		var_top="cancer_label", 
		group_by="sex", 
		plot_title= "Thailand, Chiangmai",
		plot_subtitle = "Top 5 cancer",
		missing_age = 19)



	#select USAm NPCR
	df_data_2 <- subset(df_data,id_code== 384008099)


	# plot for USA NPCR
	dt_result_2 <- 
		csu_ageSpecific_top(
			df_data_2,
			var_age="age", 
			var_cases="cases", 
			var_py="py",
			var_top="cancer_label", 
			group_by="sex", 
			plot_title= "USA",
			plot_subtitle = "Top 5 cancer",
			missing_age = 19
			)

---

csu_asr

Description

csu_asr calculate Age-Standardized Rate (ASR) and Truncated Age-Standardized Rate (TASR) across different population (Registry, year, sex...)

Usage

csu_asr(df_data,
		var_age = "age",
		var_cases = "cases",
		var_py ="py",
		group_by=NULL,
		var_age_group=NULL,
		missing_age = NULL,
		db_rate = 100000,
		first_age = 1,
		last_age = 18,
		pop_base = "SEGI",
		crude_rate = FALSE,
		var_st_err = NULL,
		correction_info = FALSE,
		var_asr = "asr",
		age_dropped = FALSE)

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_age	Age variable. Several format are accepted 1 "0-4" 0 2 "5-9" 5 3 "10-14" 10 ... ... ... 17 "80-84" 80 18 "85+" 85 Missing age value must be precise in the option missing_age. Last age group will always be considere without size (ie: 80+,85+, etc..).
var_cases	Number of event (cases, deaths, ...) variable.
var_py	Population year variable.
group_by	A vector of variables to compare different ASR (sex, country, cancer ...). Any variable included in the next option var_age_group must be also include here.
var_age_group	Variables over which the number of population age-group might change. For example the last age group (70+, 75+, 80+ or 85+) can vary across registries, or year, but not across sex.
missing_age	Age value representing the missing age cases. If not precise, missing age value will not affect the calculation.
db_rate	The denominator population. Default is 100000.
first_age	First age group included (for Truncated ASR), must be between 1 and 17. 1 represents 0-4, 2 represents 5-9, ... 5 represents 20-24 etc.
last_age	Last age group included (for Truncated ASR), must be between 2 and 18. 2 represents 5-9, ... 5 represents 20-24, ... 18 represents 85+ etc. To calculate ASR for 15-69 years, use: first_age(4), last_age(14).
pop_base	Select the standard population: "SEGI" SEGI world standard population (1960). "EURO" EURO is the European standard population (1976). "EURO2" EURO2 is the European standard population (2013). "WHO" WHO is the WHO standard population (2001).
crude_rate	Logical value. if TRUE calculate the crude rate instead of the ASR. Variables var_age_group, missing_age, pop_base, var_st_err, correction_info are not used. Default name of the new variable is "crude_rate".
var_st_err	Calculate the Standard error and name of the new variable.
correction_info	Logical value. if TRUE compute the variable "correction" : percentage of missing age cases.
var_asr	Name of the new variable for the ASR.
age_dropped	Only for truncated ASR. Logical value: if TRUE, Assume the age group not used are already dropped from the data. First_age and last_age option must still be filled.

Details

This function take automatically account the number of age group (last age group can be 70+,75+,80+,85+):
If the population data stops before 85+ (75+ for instance), the population data must be 0 when the population data is unknown so, the program can detect automatically the last age group (70+,75+,80+ or 85+) for population.

Value

Give the list of population with less than 18 age group (last age group: 70+, 75+, 80+).
Return a data.frame.

Author(s)

Mathieu Laversanne

References

https://publications.iarc.fr/_publications/media/download/3753/609d0d7711047dd76d7f3dbaa25d7f041fcd013e.pdf

See Also

csu_group_cases csu_merge_cases_pop csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

data(csu_registry_data_1)
data(csu_registry_data_2)

# you can import your data from csv file using read.csv:
# mydata <-  read.csv("mydata.csv", sep=",")

# Age standardized rate (ASR) with no missing age cases.
result <- csu_asr(csu_registry_data_1, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label" ),
                  var_age_group = c("registry_label"))
				  
# you can export your result as csv file using write.csv:
# write.csv(result, file="result.csv")
				  
# ASR,  with the percentage of correction due to missing age cases. 
result <- csu_asr(csu_registry_data_1, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label" ),
                  var_age_group = c("registry_label"),
				  missing_age = 19,				  
				  correction_info = TRUE)	
				  
# ASR and standard error with missing age.
result <- csu_asr(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_age_group = c("registry_label"), 
                  var_st_err = "st_err",
				  missing_age = 99)
				  
# Truncated ASR, 25-69 years.
result <- csu_asr(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_age_group = c("registry_label"), 
                  var_st_err = "st_err",
				  first_age = 6, 
				  last_age = 14,
				  missing_age = 99)	

# Truncated ASR, 0-15 with denominator population = 1000000. 
result <- csu_asr(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_age_group = c("registry_label"), 
                  var_st_err = "st_err",
				  first_age = 1, 
				  last_age = 3, 
				  missing_age = 99,	
				  db_rate = 1000000)		
				  
# ASR with EURO population as reference (instead of SEGI)
result <- csu_asr(csu_registry_data_1, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label" ),
                  var_age_group = c("registry_label"),
				  missing_age = 19,
                  pop_base = "EURO")

---

csu_bar_top

Description

csu_bar_top plots top X single-sided or double-sided bar chart.

Usage

csu_bar_top(df_data,
	var_value, 
	var_bar,
	group_by=NULL,
	nb_top = 10,
	plot_title=NULL,
	plot_subtitle=NULL,
	xtitle= NULL,
	label_by=NULL,
	color=NULL,
	digits = 1)

Arguments

df_data	Data (need to be R data.frame format, see examples to import csv file).
var_value	Value variable. There must be only 1 value for each bar.
var_bar	Bar label variable.
group_by	Single-sided bar chart. NULL (default) Double-sided bar chart. Variable name with exactly 2 values. (For example, "sex"). Must be filled if label_by argument is defined.
nb_top	Lowest Rank included. Default is 10.
plot_title	Title of the plot. (For example, "Top 10 cancer sites").
plot_subtitle	Subtitle of the plot. (For example, "Males").
xtitle	x-axe title. (For example, "Number of cases").
label_by	2 values vector. Will overwrite the legend label for double-sided bar chart. (See group_by). For example: c("Male", "Female").
color	The color codes are hexadecimal (e.g. "#FF0000") or predefined R color names (e.g. "red"). Single-sided bar chart. 1 hexadecimal color code (same color for each bar) or variable name with a color associated to each bar label variable. Double-sided bar chart. 2 values vector. For example: c("#2c7bb6","#b62ca1").
digits	Number of decimal digits. Default: 1

Details

This function plots a top X (default is top 10) bar chart, single-sided or double sided.

Value

Return plots and a data.frame.

Author(s)

Mathieu Laversanne

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_time_trend csu_trendCohortPeriod

Examples

data(data_individual_file)
data(data_population_file)
data(ICD_group_GLOBOCAN)

#Group individual data by:
#5 year age group
#ICD grouping from dataframe ICD_group_GLOBOCAN
#year extract from date of incidence

df_data_year <- csu_group_cases(data_individual_file,
  var_age="age",
  group_by=c("sex", "regcode", "reglabel"),
  df_ICD = ICD_group_GLOBOCAN,
  var_ICD  ="site",
  var_year = "doi")     

#Merge 5-years age grouped data with population by year (automatic) and sex

df_data <- csu_merge_cases_pop(
  df_data_year, 
  data_population_file, 
  var_age = "age_group",
  var_cases = "cases",
  var_py = "pop",
  group_by = c("sex"))


#prepare for calculate ASR
df_data$age_group_label <- NULL # to avoid warning
df_data <- subset(df_data , year == 2012) # to keep only 2012 data
df_data$year <- NULL # to avoid warning

# calculate asr
df_asr <- csu_asr(df_data,
  "age_group", 
  "cases",
  "pop",
  group_by=c("sex", "ICD_group", "LABEL", "reglabel", "regcode"),
  missing_age =19)

#remove Other cancer
df_asr <- subset(df_asr , LABEL != "Other") 
df_asr <- subset(df_asr , LABEL != "Other skin")


#keep male
df_asr_M <- subset(df_asr , sex==1)

#Single sided bar plot 
data1 <- csu_bar_top(
   df_asr_M,
   var_value="cases",
   var_bar="LABEL",
   nb_top = 10,
   plot_title = "Top 10 cancer sites",
   xtitle= "Number of cases",
   color= c("#2c7bb6"),
   digits=0) 

#Double sided bar plot example 1
data2 <- csu_bar_top(
   df_asr,
   var_value="cases",
   var_bar="LABEL",
   group_by="sex",
   nb_top = 15,
   plot_title = "Top 15 cancer sites",
   xtitle= "Number of cases",
   label_by=c("Male", "Female"),
   color = c("#2c7bb6","#b62ca1"),
   digits=0) 

#Double sided bar plot example 2
data3 <- csu_bar_top(
   df_asr,
   var_value="asr",
   var_bar="LABEL",
   group_by="sex",
   nb_top = 10,
   plot_title = "Top 10 cancer sites",
   xtitle= "Age-standardized rate per 100,000",
   label_by=c("Male", "Female"),
   color = c("#2c7bb6","#b62ca1"),
   digits=1)

---

cancer registry data

Description

CI5 XII data, all population grouped.

Usage

data("csu_ci5_mean")

Format

A data frame with 1026 observations on the following 5 variables.

ci5_cancer_code

cancer code

ci5_cancer_label

cancer label

CSU_age_factor

age variable from 1 to 18. 1 is 0-4 years, 2 is 5-9, etc..., 17 is 80-84, 18 is 85+.

CSU_C

Number of cases (incidence)

CSU_P

Population-year: Reference population

Details

This dataset have been created in order to compare age specific rate with the CI5XII mean directly on the age specific rate graph. Here is the table of cancer_label and cancer_code available:

ci5_cancer_code	ci5_cancer_label
1	Lip
2	Tongue
3	Mouth
4	Salivary glands
5	Tonsil
6	Other oropharynx
7	Nasopharynx
8	Hypopharynx
9	Pharynx unspecified
10	Oesophagus
11	Stomach
12	Small intestine
13	Colon
14	Rectum
15	Anus
16	Liver
17	Gallbladder
18	Pancreas
19	Nose, sinuses
20	Larynx
21	Lung
23	Bone
24	Melanoma of skin
25	Other skin
26	Mesothelioma
27	Kaposi sarcoma
28	Connective and soft tissue
29	Breast
30	Vulva
31	Vagina
32	Cervix uteri
33	Corpus uteri
35	Ovary
37	Placenta
38	Penis
39	Prostate
40	Testis
42	Kidney
43	Renal pelvis
44	Ureter
45	Bladder
47	Eye
48	Brain, nervous system
49	Thyroid
50	Adrenal gland
51	Other endocrine
52	Hodgkin disease
53	Non-Hodgkin lymphoma
54	Immunoproliferative diseases
55	Multiple myeloma
56	Lymphoid leukaemia
57	Myeloid leukaemia
58	Leukaemia unspecified
59	Myeloproliferative disorders
60	Myelodysplastic syndromes
62	All sites
63	All sites but skin

See: csu_ageSpecific

Source

https://ci5.iarc.fr/ci5-xii/

See Also

csu_ageSpecific

Examples

data(csu_registry_data_1)
test <- subset(csu_registry_data_1 , registry_label == "Colombia, Cali")
csu_ageSpecific(test,
                plot_title = "Colombia, Liver, male",
                CI5_comparison = "Liver")
 
#See more examples here:
help(csu_ageSpecific)

---

Cancer in five continent volume XII.

Description

CI5 XII summary database.

Usage

data("csu_CI5XII_data")

Format

A data frame with 1113210 observations on the following 10 variables.

id_code

population code (integer)

id_label

population label (character)

country_code

UN country code (integer)

ethnic_code

ethnic code (integer)

cancer_code

cancer code (integer)

cancer_label

cancer label (character)

sex

sex (integer)

age

age variable from 1 to 19. 1 is 0-4 years, 2 is 5-9, etc..., 17 is 80-84, 18 is 85+, 19 represents missing age. (integer)

cases

Number of cases (incidence) (integer)

py

Population-year: Reference population of the registry (integer)

period

Period (character)

Details

All information are available here: https://ci5.iarc.fr/ci5-xii/

Source

https://ci5.iarc.fr/ci5-xii/

See Also

csu_asr csu_cumrisk csu_eapc csu_ageSpecific

Examples

data("csu_CI5XII_data")

#get the id code asssociate to id_label
#print(unique(csu_CI5XII_data[,c("id_code", "id_label" )]),nrows = 1000)

#get the cancer code asssociate to cancer_label
#print(unique(csu_CI5XII_data[,c("cancer_label", "cancer_code")]),nrows = 1000)

#remove all cancers:
df_data <- subset(csu_CI5XII_data ,cancer_code < 62)

#change group_by option to factor
df_data$sex <- factor(df_data$sex, levels=c(1,2), labels=c("Male", "Female"))


#select Thailand changmai
df_data_1 <- subset(df_data, id_code==476400199)



dt_result_1 <- csu_ageSpecific_top(df_data_1,
                    var_age="age", 
                    var_cases="cases", 
                    var_py="py",
                    var_top="cancer_label", 
                    group_by="sex", 
                    plot_title= "Thailand, Chiangmai",
                    plot_subtitle = "Top 5 cancer",
                    missing_age = 19)



					
#select USA
df_data_2 <- subset(df_data,id_code== 384008099 & ethnic_code == 99)

dt_result_2 <- csu_ageSpecific_top(df_data_2,
                                   var_age="age", 
                                   var_cases="cases", 
                                   var_py="py",
                                   var_top="cancer_label", 
                                   group_by="sex", 
                                   plot_title= "USA",
                                   plot_subtitle = "Top 5 cancer",
                                   missing_age = 19)

---

csu_cumrisk

Description

csu_cumrisk Calculate cumulative risk across different population (Registry, year, sex...)

Usage

csu_cumrisk(df_data,
		var_age = "age",
		var_cases = "cases",
		var_py ="py",
		group_by=NULL,
		missing_age = NULL,
		last_age = 15,
		var_st_err=NULL,
		correction_info=FALSE,
		var_cumrisk="cumrisk")

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_age	Age variable. Several format are accepted 1 "0-4" 0 2 "5-9" 5 3 "10-14" 10 ... ... ... 17 "80-84" 80 18 "85+" 85 Missing age value must be precise in the option missing_age. Last age group will always be considere without size (ie: 80+,85+, etc..).
var_cases	Number of event (cases, deaths, ...) variable.
var_py	Population year variable.
group_by	A vector of variables to compare different ASR (sex, country, cancer ...).
missing_age	Age value representing the missing age cases. If not precise, missing age value will not affect the calculation.
last_age	Last age group included, must be between 2 and 17. 2 represents 5-9, ... 5 represents 20-24, ..., 17 represents 80-84, 18 represents 85+ etc. default is 15 (calculates cumulative risk 0-74)
var_st_err	Calculate the Standard error and name of the new variable.
correction_info	Logical value. if TRUE compute the variable "correction" : percentage of missing age cases.
var_cumrisk	Name of the new variable for the cumulative risk.

Details

Calculate cumulative risk across different population
the age group include in cumulative risk will always exclude the last age group since we do not know the size of the last age group (ie: 85+, 80+ etc..)

Value

Return a data.frame.

Author(s)

Mathieu Laversanne

References

https://publications.iarc.fr/_publications/media/download/3753/609d0d7711047dd76d7f3dbaa25d7f041fcd013e.pdf

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

data(csu_registry_data_1)
data(csu_registry_data_2)

# you can import your data from csv file using read.csv:
# mydata <-  read.csv("mydata.csv", sep=",")

# Cumulative risk, 0-74 years, with no missing age cases.
result <- csu_cumrisk(csu_registry_data_1, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label" ))
				  
# you can export your result as csv file using write.csv:
# write.csv(result, file="result.csv")
				  
# Cumulative risk, 0-74 years, with the percentage of correction due to missing age cases. 
result <- csu_cumrisk(csu_registry_data_1, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label" ),
				  missing_age = 19,				  
				  correction_info = TRUE)	
				  
# Cumulative risk 0-74 years and standard error with missing age.
result <- csu_cumrisk(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_st_err = "st_err",
				  missing_age = 99)
				  
# Cumulative risk, 0-69 years.
result <- csu_cumrisk(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_st_err = "st_err",
				  last_age = 14,
				  missing_age = 99)

---

csu_eapc

Description

csu_eapc calculate the Estimated Annual Percentage Change (EAPC) of rates during a time period with the Confidence Interval (CI) across different population (Registry, year, sex...)

Usage

csu_eapc(df_data,
	var_rate="asr",
	var_year="year",
	group_by=NULL,
	var_eapc = "eapc",
	CI_level = 0.95)

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_rate	Rate variable. (Standardized or not, incidence, mortality, etc..)
var_year	Period variable. (Year, month, etc...)
group_by	A vector of variables to compare different EAPC (sex, country, cancer ...).
var_eapc	Name of the new variable for the EAPC. The variable for the CI will be name based on var_eapc + "_" + up/Low.
CI_level	Confidence interval level. Default is 0.95.

Details

This function use Generalized Linear Model (GLM):
glm(log(rate) ~ year, family=gaussian(link = "identity")).
0 value are ignored. More details in reference below.

Value

Return a dataframe.

Author(s)

Mathieu Laversanne

References

https://regstattools.iconcologia.net/stats/sart/eapc/eapc_method.pdf

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

data(csu_registry_data_2)

# you import your data from csv file using read.csv:
# mydata <-  read.csv("mydata.csv", sep=",")

# Estimated Annual Percentage Change (EAPC) base on ASR.


df_asr <- 
	csu_asr(csu_registry_data_2, 
	  "age", "cases", "py",
	  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
	  var_age_group = c("registry_label"), 
	  missing_age = 99
	  )
				  
result <- 
	csu_eapc(df_asr,
		"asr", "year",
		group_by=c("registry", "registry_label", "sex", "ethnic" )
		)

					
# you can export your result as csv file using write.csv:
# write.csv(result, file="result.csv")

---

csu_group_cases

Description

csu_group_cases groups individual data into 5 years age-group data and other user defined variable (sex, registry, etc...).
Optionally: Group cancer based on a standard ICD10 coding; Extract year from custom year format.

Usage

csu_group_cases(df_data, 
	var_age ,
	group_by=NULL,
	var_cases = NULL,
	df_ICD = NULL,
	var_ICD=NULL,
	var_year = NULL,
	all_cancer=FALSE)

Arguments

df_data	Individual data (need to be R data.frame format, see examples to import csv file).
var_age	Age variable. (Numeric). Value > 150 will be considered as missing age.
group_by	(Optional) A vector of variables to create the different population (sex, country, etc...).
var_cases	(Optional) cases variable: If there is already a variable for the number of cases.
df_ICD	(Optional) ICD file for ICD grouping information. Must have 2 fields: "ICD", "LABEL" . 2 formats are possible: Each ICD code separated by ICD group ICD LABEL C82 NHL C83 NHL C84 NHL C85 NHL C96 NHL ICD code already grouped. ICD_group LABEL C82-85,C96 NHL 2 ICD codes separated by "-" includes all the ICD code between. 2 ICD codes separated by "," includes only these 2 ICD code. For instance, C82-85, C96 (or C82-C85, C96) includes: C82, C83, C84, C85 and C96 Must be filled if var_ICD argument is defined example: ICD_group_GLOBOCAN
var_ICD	(Optional) ICD variable: ICD variable in the individual data. Must be filled if df_ICD argument is defined
var_year	(Optional) Year variable: Extract year from custom format , as long as the year is expressed with 4 digits (i.e. ("yyyymmdd","ddmmyyyy", "yyyy/mm","dd-mm-yyyy", etc..) and group data by year.
all_cancer	(Optional) If TRUE, will calculate the number of cases for all cancers (C00-97) and all cancers but non-melanoma of skin (C00-97 but C44) Need var_ICD and df_ICD arguments to be defined

Details

For most analysis, individual cases database need to be grouped by category.
This function groups data by 5 years age-group and other user defined variable.
Next step will be to add 5 years population data. (see csu_merge_cases_pop).

Value

Return a dataframe.

Author(s)

Mathieu Laversanne

See Also

csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

# you can import your data from csv file using read.csv:
# mydata <-  read.csv("mydata.csv", sep=",")



	
data(ICD_group_GLOBOCAN)
data(data_individual_file)

#group individual data by 
# 5 year age group 
df_data_age <- csu_group_cases(data_individual_file,
  var_age="age",
  group_by=c("sex", "regcode", "reglabel", "site")) 



	#group individual data by 
	# 5 year age group 
	# ICD grouping from dataframe ICD_group_GLOBOCAN

	df_data_icd <- csu_group_cases(data_individual_file,
	  var_age="age",
	  group_by=c("sex", "regcode", "reglabel"),
	  df_ICD = ICD_group_GLOBOCAN,
	  var_ICD  ="site") 

	#group individual data by 
	# 5 year age group 
	# ICD grouping from dataframe ICD_group_GLOBOCAN
	# year (extract from date of incidence)

	df_data_year <- csu_group_cases(data_individual_file,
	  var_age="age",
	  group_by=c("sex", "regcode", "reglabel"),
	  df_ICD = ICD_group_GLOBOCAN,
	  var_ICD  ="site",
	  var_year = "doi")       
	


# you can export your result as csv file using write.csv:
# write.csv(result, file="result.csv")

---

csu_merge_cases_pop

Description

csu_merge_cases_pop merges registry data and population data, group by year and other user defined variable (sex, registry, etc...).

Usage

csu_merge_cases_pop(df_cases, 
	df_pop,
	var_age,
	var_cases="cases",
	var_py=NULL,
	group_by=NULL)

Arguments

df_cases	Registry data group by 5 years-age group (need to be R data.frame format, see examples to import csv file).
df_pop	Population data group by 5-years age group (need to be R data.frame format, see examples to import csv file).
var_age	Age variable. Several format are accepted 1 0-4 0 2 5-9 5 3 10-14 10 ... ... ... 17 80-84 80 18 85+ 85 This variable must be a variable with the same column name in both dataset (df_cases and df_pop). Age >= 85 in the df_pop dataset will be aggregated as 85+.
var_cases	Cases variable in the df_cases dataset.
var_py	(Optional) If population is "long format", name of the population variable in the df_pop dataset. If population data is wide format (see details), var_py must be NULL.
group_by	(Optional) A vector of variables to create the different population (sex, country, etc...). Each variable must be a variable with the same column name in both dataset (df_cases and df_pop). Do not include the "year" variable since it is automatically detected (see details).

Details

This function merges registry data and population for further analysis.
Both datasets must be group by 5-years age group.
If present, the year information in format "yyyy" will be detected automatically.
2 formats are accepted for population data:.
Long format: (year and population are 2 variables)

sex	age	pop	year
1	1	116128	2005
1	2	130995	2005
1	3	137556	2005
...	...	...	...
2	16	27171	2007
2	17	13585	2007
2	18	13585	2007

Wide format: (One column per year and no population variable, "yyyy" year format must be included in columns name)

sex	age	Y2013	Y2014	Y2015
1	0-4	215607	237346	247166
1	5-9	160498	152190	152113
1	10-14	175676	171794	165406
...	...	...	...	...
2	75-79	20625	20868	23434
2	80-84	7187	7276	7620
2	85+	2551	2597	2617

Value

Return a dataframe.

Author(s)

Mathieu Laversanne

See Also

csu_group_cases csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

# you can import your data from csv file using read.csv:
# mydata <-  read.csv("mydata.csv", sep=",")

data(ICD_group_GLOBOCAN)
data(data_individual_file)
data(data_population_file)

#group individual data by 
# 5 year age group 
# ICD grouping from dataframe ICD_group_GLOBOCAN
# year (extract from date of incidence)

df_data_year <- csu_group_cases(data_individual_file,
  var_age="age",
  group_by=c("sex", "regcode", "reglabel"),
  df_ICD = ICD_group_GLOBOCAN,
  var_ICD  ="site",
  var_year = "doi")     

#Merge 5-years age grouped data with population by year (automatic) and sex

df_data <- csu_merge_cases_pop(
	df_data_year, 
	data_population_file, 
	var_age = "age_group",
	var_cases = "cases",
	var_py = "pop",
	group_by = c("sex"))


# you can export your result as csv file using write.csv:
# write.csv(result, file="result.csv")

---

cancer registry data

Description

Cancer registry data for liver cancer, males, 2007, 4 registries.

Usage

data("csu_registry_data_1")

Format

A data frame with 76 observations on the following 5 variables.

age

age variable from 1 to 19. 1 is 0-4 years, 2 is 5-9, etc..., 17 is 80-84, 18 is 85+, 19 represents missing age.

cases

Number of cases (incidence)

py

Population-year: Reference population of the registry

registry_label

Name of the registry

registry

Registry code

Details

this 4 registries have been selected for this example as they different number of age group (75+, 80+, 85+) and some have missing age cases.

Source

https://ci5.iarc.fr/ci5-xii/

See Also

csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top

Examples

data(csu_registry_data_1)

# Age standardized rate (ASR) with no missing age cases.
result <- csu_asr(csu_registry_data_1, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label" ),
                  var_age_group = c("registry_label"))
				  
#See more examples here:
help(csu_asr)

---

cancer registry data

Description

Cancer registry data for liver cancer

Usage

data("csu_registry_data_2")

Format

A data frame with 125856 observations on the following 8 variables.

sex

sex variable: 1 male, 2 female

year

year variable, from 1953 to 2007

age

age variable from 0 to 85 with missing age.
0 is 0-4 years, 5 is 5-9, ..., 80 is 80-84, 85 is 85+, 99 represents missing age.

cases

Number of cases (incidence)

py

Population-year: Reference population of the registry

registry_label

Name of the registry (118 populations)

registry

Registry code (102 registries)

ethnic

ethnic code: white (10), black (30), ..., all(99).

Source

https://ci5.iarc.fr/ci5-xii/

See Also

csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend csu_trendCohortPeriod

Examples

data(csu_registry_data_2)
				  
# ASR and standard error with missing age.
result <- csu_asr(csu_registry_data_2, 
                  "age", "cases", "py",
                  group_by = c("registry", "registry_label", "sex", "year", "ethnic" ),
                  var_age_group = c("registry_label"), 
				  missing_age = 99,
                  var_st_err = "st_err")
				  
#See more examples here:
help(csu_asr)

---

csu_time_trend

Description

csu_time_trend plot stats over year.

Usage

csu_time_trend(df_data,
		var_trend = "asr",
		var_year = "year",
		group_by = NULL,
		logscale = FALSE,
		smoothing = NULL,
		legend = csu_trend_legend(),
		color_trend = NULL,
		ytitle = "Age standardized rate per 100,000",
		plot_title = "csu_title")

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_trend	Statistics variable to be plot on Y axis. Usually for the ASR, but it could be the cumulative risk, or the number of cases.
var_year	Time variable.
group_by	Variable to compare different age specific rate (sex, country, cancer ...). Only one variable can be chosen.
logscale	Logical value: if TRUE Y-axis use logscale.
smoothing	Apply a smoothing using the R loess function. . The numerical parameter controls the degree of smoothing. See option span in loess.
legend	legend option: see csu_trend_legend. Use only if group_by is not NULL.
color_trend	Vector of color for the trend. The color codes are hexadecimal (e.g. "#FF0000") or predefined R color names (e.g. "red").
ytitle	Y-axis title. Default is "Age standardized rate per 100,000".
plot_title	Title of the plot.

Details

This function is design the plot a statistics over time. It has been design for the ASR by year, but can be used for other statistics over time period. The group_by option allow to compare different population or cancer.

Value

Return a plot.

Author(s)

Mathieu Laversanne

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_trendCohortPeriod

Examples

data(csu_registry_data_2)

	# you can import your data from csv file using read.csv:
	# mydata <-  read.csv("mydata.csv", sep=",")

	# to select only 1 population 

	test <- subset(csu_registry_data_2 , registry_label == "Colombia, Cali")

	# to change sex variable to factor with label
	test$sex <- factor(test$sex, levels=c(1,2), labels=c("Male", "Female"))

	# to calculate the asr
	df_asr <- csu_asr(
		test,missing_age = 99,
		group_by  = c("registry", "registry_label", "year", "sex", "ethnic"),
		var_age_group =  c("registry", "registry_label")
		)


	# plot ASR ove year, by sex.
	csu_time_trend(df_asr, group_by="sex",
			  plot_title = "Colombia, Liver")

	# plot ASR over year, by sex, with small smoothing.
	csu_time_trend(df_asr, group_by="sex",
			  plot_title = "Colombia, Liver",
			  smoothing = 0.3)


	# plot ASR over year, by sex, with high smoothing.
	csu_time_trend(df_asr, group_by="sex",
			  plot_title = "Colombia, Liver",
			  smoothing = 0.5)

	# Plot embedded in a graphic device
	pdf(paste0(tempdir(),"/test.pdf"),width = 11.692 , height =  8.267) 
	csu_time_trend(df_asr, group_by="sex",
			  plot_title = "Colombia, Liver",
			  smoothing = 0.3)

	csu_time_trend(df_asr, group_by="sex",
			  plot_title = "Colombia, Liver",
			  smoothing = 0.5)

	dev.off()

---

csu_trend_legend

Description

csu_trend_legend legend option use in Rcan package for trends.

Usage

csu_trend_legend(title=NULL, position="bottom",nrow=1, right_space_margin=1)

Arguments

title	Title of the legend. (only if position = "bottom").
position	Position of the legend: "bottom" The legend is place at the bottom of the graph. use with option nrow. "right" The legend is place at the right of each trend. use with option right_space_margin. If there is already a plot in the device, use plot.new() to avoid overprint
nrow	Number of row of the legend (only if position = "bottom").
right_space_margin	If the position = "right", the right margin need to be enough big, to contain the text legend. value should be between 1 and 10 depending the size of the text legend.

Details

This function return a list for the trend graphics of the package Rcan

Value

Return a structured list.

Author(s)

Mathieu Laversanne

See Also

csu_ageSpecific

Examples

data(csu_registry_data_1)

csu_ageSpecific(csu_registry_data_1,
                group_by="registry_label",
                legend=csu_trend_legend(title="registry", position="bottom", nrow = 2),
                plot_title = "Legend: bottom")
				
csu_ageSpecific(csu_registry_data_1,
                group_by="registry_label",
                legend=csu_trend_legend(position="right", right_space_margin = 2),
                plot_title = "Legend: right, cut")

csu_ageSpecific(csu_registry_data_1,
                group_by="registry_label",
                legend=csu_trend_legend(position="right", right_space_margin = 6.5),
                plot_title = "Legend: right")
				
#See more examples here:
help(csu_ageSpecific)

---

csu_trendCohortPeriod

Description

csu_trendCohortPeriod plot cohort period age specific graph.

Usage

csu_trendCohortPeriod(
  df_data,
  var_age = "age",
  var_cases="cases",
  var_py="py",
  var_year = "year",
  type = "Cohort",
  missing_age = NULL,
  logscale = TRUE,
  db_rate = 100000,
  first_age = 6,
  last_age = 16,
  year_group = 5,
  age_dropped=FALSE,
  plot_title = "csu_title",
  format_export = NULL,
  graph_dev =FALSE)

Arguments

df_data	Data (need to be R data.frame format, see example to import csv file).
var_age	Age variable. Several format are accepted 1 "0-4" 0 2 "5-9" 5 3 "10-14" 10 ... ... ... 17 "80-84" 80 18 "85+" 85 Missing age value must be precise in the option missing_age. Last age group will always be considere without size (ie: 80+,85+, etc..).
var_cases	Number of event (cases, deaths, ...) variable.
var_py	Population year variable.
var_year	Time variable.
type	Type of the plot: "Cohort" Cohort graph. "Period" Period graph. "Both" Cohort Period graph.
missing_age	Age value representing the missing age cases.
logscale	Logical value: if TRUE Y-axis use logscale.
db_rate	The denominator population. Default is 100000.
first_age	First age group included, must be between 1 and 17. 1 represents 0-4, 2 represents 5-9, ... 5 represents 20-24 etc. Default is 6 (25-29 years).
last_age	Last age group included, must be between 2 and 18. 2 represents 5-9, ... 5 represents 20-24, ... 18 represents 85+ etc. Default is 16 (75-79 years). To plot the age group from 15-19 years until 65-69 years, use: first_age(4), last_age(14)
year_group	Usually, data are regrouped in 5 years period. . The numerical parameter controls the size of the group.
age_dropped	Only if some age grouped are missing in the data. Logical value: if TRUE, assume the age group not used are already dropped from the data. First_age and last_age option must still be filled.
plot_title	Title of the plot.
format_export	export the graph in different format: NULL Plot in R studio windows. "pdf" Export in PDF format. "tiff" Export in TIFF 300dpi format. "png" Export in PNG 200dpi format. "svg" Export in SVG format. Can be edit with https://inkscape.org/fr/. The filename is the plot_title option.
graph_dev	If the plot is embedded in a graphics Device function (such as pdf()), the graph_dev option should be set to TRUE for the first graph to avoid a blank page.

Details

This function is design the plot a the age-specific cohort and period plot. The type option allow to choose between the 3 different graphics: "Cohort", "Period", or "Both". Please note than the cohort plot and the period plot can be superimposed if the first_age is too low.

Value

Return a plot.

Author(s)

Mathieu Laversanne

See Also

csu_group_cases csu_merge_cases_pop csu_asr csu_cumrisk csu_eapc csu_ageSpecific csu_ageSpecific_top csu_bar_top csu_time_trend

Examples

data(csu_registry_data_2)

	# you can import your data from csv file using read.csv:
	# mydata <-  read.csv("mydata.csv", sep=",")
	
	# to select only 1 population 
	test <- subset(csu_registry_data_2,registry == 84020 & sex == 1)


	
	# plot cohort graph from 25-29 years until 75-79 years.
	csu_trendCohortPeriod(df_data=test,
                        missing_age =99,
                        plot_title = "USA, Liver, males")
						
	# plot Period graph from 0-5 until 85+.
	csu_trendCohortPeriod(df_data=test,
                        missing_age =99,
                        plot_title = "USA, Liver, males",
						type="Period",
						first_age=1,
						last_age=18)
						
	# plot Cohort-Period graph from 30-34 years until 70-74 years.
	csu_trendCohortPeriod(df_data=test,
                        missing_age =99,
                        plot_title = "USA, Liver, males",
						type="Both",
						first_age=7,
						last_age=15)

	# plot Cohort-Period graph from 30-34 years until 70-74 years with Y axis normal scale.
	csu_trendCohortPeriod(df_data=test,
                        missing_age =99,
                        plot_title = "USA, Liver, males",
						type="Both",
						first_age=7,
						last_age=15,
						logscale=FALSE)

					
	# plot Cohort graph from 25-29 years until 75-79 years, with data grouped in 2 years period.
	csu_trendCohortPeriod(df_data=test,
						  missing_age =99,
						  plot_title = "USA, Liver, males",
						  type="Cohort",
						  year_group = 2)
						  
	# Plot embedded in a graphic device
	pdf(paste0(tempdir(),"/example_test.pdf"))
	csu_trendCohortPeriod(df_data=test,
                        missing_age =99,
                        plot_title = "USA, Liver, males",
						type="Both",
						first_age=7,
						last_age=15,
						graph_dev=TRUE)
			  
	csu_trendCohortPeriod(df_data=test,
                        missing_age =99,
                        plot_title = "USA, Liver, males",
						type="Both",
						first_age=7,
						last_age=15,
						logscale=FALSE)
				
	dev.off()

---

Data individual example

Description

Data individual example for the function csu_group_cases
1 line = 1 cases.

Usage

data("data_individual_file")

Format

A data frame with 19284 observations on the following 10 variables.

regcode

registry code

reglabel

registry label

sex

sex

age

age

doi

date of birth (yyyymmdd)

site

ICD10 code

histo

histology

beh

behavior code

grade

grade

basis

basis

Details

This dataset provide an example how to group individual cases to 5 years data, grouped by ICD code and year using the function csu_group_cases and the database ICD_group_GLOBOCAN

See Also

csu_group_cases ICD_group_GLOBOCAN ICD_group_CI5

Examples

data(ICD_group_GLOBOCAN)
data(data_individual_file)

#group individual data by 
#	5 year age group 

df_data_age <- csu_group_cases(data_individual_file,
	var_age="age",
	group_by=c("sex", "regcode", "reglabel", "site"))


#group individual data by 
#	5 year age group 
#	ICD grouping from dataframe ICD_group_GLOBOCAN

df_data_icd <- csu_group_cases(data_individual_file,
	var_age="age",
	group_by=c("sex", "regcode", "reglabel"),
	df_ICD = ICD_group_GLOBOCAN,
	var_ICD  ="site")

---

Population data example

Description

Population data example for the function csu_merge_cases_pop.

Usage

data("data_population_file")

Format

A data frame with 160 observations on the following 4 variables.

year

year

sex

sex

age_group

5 year age group

pop

Count of population year

Details

This dataset provide an example to merge 5 years age grouped data with population data using the function csu_merge_cases_pop

See Also

csu_merge_cases_pop

Examples

data(ICD_group_GLOBOCAN)
data(data_individual_file)
data(data_population_file)

#group individual data by 
# 5 year age group 
# ICD grouping from dataframe ICD_group_GLOBOCAN
# year (extract from date of incidence)

df_data_year <- csu_group_cases(data_individual_file,
  var_age="age",
  group_by=c("sex", "regcode", "reglabel"),
  df_ICD = ICD_group_GLOBOCAN,
  var_ICD  ="site",
  var_year = "doi")     

#Merge 5-years age grouped data with population by year (automatic) and sex

df_pop <- csu_merge_cases_pop(df_data_year, data_population_file, var_age = "age_group", 
                           var_cases = "cases", var_py = "pop", group_by = c("sex"))

---

ICD10 group detailed example

Description

ICD10 group example for the function csu_group_cases based on CI5XII ICD grouping

Usage

data("ICD_group_CI5")

Format

A data frame with 97 observations on the following 2 variables.

ICD

ICD10 code

LABEL

label for cancer group

Details

This dataset provide an example how to regroup ICD code using the function csu_group_cases For instance this group

ICD	LABEL
C18	COLORECTUM
C19	COLORECTUM
C20	COLORECTUM
C21	COLORECTUM

Will become:

ICD_group	LABEL
C18-C21	COLORECTUM

See: csu_group_cases

See Also

csu_group_cases ICD_group_GLOBOCAN

Examples

data(ICD_group_CI5)
data(data_individual_file)

#group individual data by 
#	5 year age group 
#	ICd grouping from dataframe ICD_group_CI5

df_data_icd <- csu_group_cases(data_individual_file,
	var_age="age",
	group_by=c("sex", "regcode", "reglabel"),
	df_ICD = ICD_group_CI5,
	var_ICD  ="site")

---

ICD10 group example

Description

ICD10 group example for the function csu_group_cases based on GLOBOCAN

Usage

data("ICD_group_GLOBOCAN")

Format

A data frame with 97 observations on the following 2 variables.

ICD

ICD10 code

LABEL

label for cancer group

Details

This dataset provide an example how to regroup ICD code using the function csu_group_cases For instance this group

ICD	LABEL
C18	COLORECTUM
C19	COLORECTUM
C20	COLORECTUM
C21	COLORECTUM

Will become:

ICD_group	LABEL
C18-C21	COLORECTUM

See: csu_group_cases

See Also

csu_group_cases ICD_group_CI5

Examples

data(ICD_group_GLOBOCAN)
data(data_individual_file)

#group individual data by 
#	5 year age group 
#	ICd grouping from dataframe ICD_group_GLOBOCAN

df_data_icd <- csu_group_cases(data_individual_file,
	var_age="age",
	group_by=c("sex", "regcode", "reglabel"),
	df_ICD = ICD_group_GLOBOCAN,
	var_ICD  ="site")

---