knitr::opts_chunk$set(warning=FALSE, message=FALSE, out.width = '100%')
#Cancer LCDI Project syntax example 2----
#Libraries ----
library(dplyr)
library(survival)
library(survminer)
library(forcats)
library(readxl)
library(haven)
library(tidyverse)
library(lubridate)
library(data.table)
library(tidyr)
library(stringr)
library(readr)
library(glue)
library(collapsibleTree)
library(ggplot2)
library(ggalluvial)
library(writexl)
library(networkD3)
library(rmarkdown)
library(DT)
library(knitr)
library(kableExtra)
library(plotly)
#rounding function
round2 <- function(x, n) {
posneg = sign(x)
z = abs(x)*10^n
z = z + 0.5
z = trunc(z)
z = z/10^n
z * posneg
}
#SMR06 ----
#For all columns use this: "Z:/Oncology Data/Cancer LCDI/SMR06/SMR06 col names pos.csv"
#This is a table derived from the NHS Boards file layouts file that comes from the SMR linked file that NHS Lothian downloads from ISD
#Please note that not all variables are read into R
SMR06headers<-read_csv("Z:/Oncology Data/Cancer LCDI/SMR06/SMR06 Kept cols.csv")
PosFrom <- SMR06headers$Position_From
PosTo <- SMR06headers$Position_To
SMR06_header<-SMR06headers$Field_name
#REMEMBER TO SPECIFY GUESS_MAX to avoid warnings!.
read_fwf("Z:/Oncology Data/Cancer LCDI/SMR06/S_mrl_smr06.dat", fwf_positions(PosFrom,PosTo,
col_names = SMR06_header),guess_max = 200000) %>%
select(-c(SURNAME,SURNAME_SNDEX_CODE,PREVIOUS_SNDEX_CODE,PREVIOUS_SURNAME,FIRST_FORENAME)) -> SMR06
#Changing dates to date format
SMR06$INCIDENCE_DATE <- SMR06$INCIDENCE_DATE %>% ymd()
SMR06$DATE_OF_BIRTH <- SMR06$DATE_OF_BIRTH %>% ymd()
SMR06$DATE_OF_DEATH <- SMR06$DATE_OF_DEATH %>% ymd()
SMR06$DATE_1ST_CHEMO <-SMR06$DATE_1ST_CHEMO %>% ymd()
SMR06$DATE_1ST_HORM <- SMR06$DATE_1ST_HORM %>% ymd()
SMR06$DATE_1ST_OTHER_THERAPY <-SMR06$DATE_1ST_OTHER_THERAPY %>% ymd()
SMR06$DATE_1ST_RAD <-SMR06$DATE_1ST_RAD %>% ymd()
SMR06$DATE_1ST_SURGERY <-SMR06$DATE_1ST_SURGERY %>% ymd()
SMR06$INCIDENCE_YEAR <-year(SMR06$INCIDENCE_DATE)
SMR06$INCIDENCE_YEAR <-as.numeric(SMR06$INCIDENCE_YEAR)
#HB code lookups
HB_LOOKUP<- read_sav("N:/Library/Lookups/Geographic/HB_Lookup.sav")
#Selecting patient CHI numbers who have a breast cancer diagnosis, this variable can be used to identify patients
# who have additional tumours as well as early breast cancer
# Filtering to show only incidence dates from 01/01/2015 to 31/12/2016
#Cancer codes for early stage breast cancer
cancerCodes <- c("C50","C500","C501","C502","C503","C504","C505","C506",
"C508","C509","C5091")
#Check to see if first surgery occured within 9 months of being diagnosed
difftime(SMR06$DATE_1ST_SURGERY,SMR06$INCIDENCE_DATE,units=c("days"))->SMR06$time_from_diag_to_surg
#Flagging any surgeries which occur within 9 months of incidence date
ifelse(SMR06$time_from_diag_to_surg<=273.75,1,0)->SMR06$Surg_within_9months
SMR06 %>% filter(ICD10S_CANCER_SITE %in% cancerCodes & INCIDENCE_YEAR>=2015 & INCIDENCE_YEAR<=2016 & HERCEPTIN_RECEPTOR_STATUS==1 & SURGERY==1 & Surg_within_9months==1) %>% select(UPI_NUMBER,DATE_OF_BIRTH,SEX,ETHNIC_GROUP,INCIDENCE_DATE,ICD10S_CANCER_SITE,OESTROGEN_RECEPTOR_STATUS,PROGESTERONE_RECEPTOR_STATUS,HERCEPTIN_RECEPTOR_STATUS,STAGE_CLINICAL_T,STAGE_CLINICAL_N,STAGE_CLINICAL_M,STAGE_PATHOLOGIC_T,STAGE_PATHOLOGIC_N,STAGE_PATHOLOGIC_M,SURGERY,DATE_1ST_SURGERY,HBRES_CURRENT,DATE_1ST_SURGERY,DATE_1ST_CHEMO) -> SMR06_BC_201516
#Creating a flag for metastasis
ifelse(SMR06_BC_201516$STAGE_CLINICAL_M==1 | SMR06_BC_201516$STAGE_PATHOLOGIC_M==1,1,0)->SMR06_BC_201516$M_flag
## Reading in advanced breast cancer dataset
read_csv("Z:/Oncology Data/Cancer LCDI/Secondary Breast Cancer CNS database/Second attempt/Adv_BC_pats_CHI v2.csv") %>% select(-c("NAME")) ->Sec_BC_pats
Sec_BC_pats$UPI_NUMBER %>% as.character() ->Sec_BC_pats$UPI_NUMBER
Sec_BC_pats %>% filter(UPI_NUMBER %in% SMR06_BC_201516$UPI_NUMBER) -> Sec_BC_pats
#Adding Metastatic incidence date to the SMR06 data
Sec_BC_pats %>% select(UPI_NUMBER,Met_inc_date) %>% full_join(SMR06_BC_201516)->SMR06_BC_201516
as.Date(SMR06_BC_201516$Met_inc_date, format="%d/%m/%Y") -> SMR06_BC_201516$Met_inc_date
#Creating a flag to check if the patient has developed metastatic disease before having surgery
ifelse(SMR06_BC_201516$Met_inc_date<SMR06_BC_201516$DATE_1ST_SURGERY,1,0)->SMR06_BC_201516$Mets_before_surgery
#All patients with M stage 1 appear to have presented as metastatic initally and have no other smr06 records where M stage is 0
#Therefore it is safe to remove the records where stage M = 1 as these patients don't have any other records of early breast cancer
#filter out all M stage 1 patients
SMR06_BC_201516 %>% filter(M_flag==0|is.na(M_flag)==1)->SMR06_BC_201516
#Matching Health Board descriptions
SMR06_BC_201516$HBRES_CURRENT_NAME <- with(HB_LOOKUP,HB_Name[match(SMR06_BC_201516$HBRES_CURRENT,
HB2014)])
##Reading in Chemocare data
read_xlsx("Z:/Oncology Data/Cancer LCDI/Chemocare/MedicineUtilisationReport_20-12-2018_09-36-24.xlsx") -> Chemocare
#Filter Chemocare to the patients in the Cohort and the appointment dates between 2015 and 2017
Chemocare$`Appointment Date` %>% as.Date(format="%d-%m-%Y") -> Chemocare$`Appointment Date`
Chemocare %>% filter(`Appointment Date`>="2015-01-01" & `Appointment Date`<="2017-12-31") -> Chemocare
#Filtering to contain only neo-adjuvant treatments
#Filter the Chemocare data to show only the patients from the cohort
#Filter for cytotoxic or biological modifiers - these are Chemotherapy treatments and targeted therapy
Chemocare %>% filter(grepl('Neo-Adjuvant', Intention)) %>%
filter(CHI %in% SMR06_BC_201516$UPI_NUMBER) %>%
filter(`Drug Type` %in% c("Cytotoxic","Biological Modifer")& `Drug Status` %in% c("Authorised","Given"))->Chemocare_cyto_bio
#The CHI numbers from the chemocare_cyto_bio dataset are the patients who fit the following criteria:
#Have been diagnosed with breast cancer between 2015 and 2016
#Have had surgery within 9 months of being diagnosed with early BC
#Have not been diagnosed as metastatic before having surgery
#Are HER2 +ve
#Have had at least one round of neo-adjuvant SACT
Chemocare_cyto_bio %>% select(CHI) %>% unique() -> Cohort
#Filter the datasets to the cohort of interset
SMR06 %>% filter(UPI_NUMBER %in% Cohort$CHI) -> SMR06
SMR06_BC_201516 %>% filter(UPI_NUMBER %in% Cohort$CHI) -> SMR06_BC_201516
Chemocare %>% filter(CHI %in% Cohort$CHI) -> Chemocare The methods in this report has been developed as part of the LCDI project, for information on further methodologies, data sources and definitions please see the Cancer Wiki page: https://www.wiki.ed.ac.uk/display/CAN
This report shows the methods and subsequent results from deriving patient characteristics and outcomes in 42 HER2 positive early breast cancer patients who have met the following criteria:
The outcomes reported are at an aggregate level where patient numbers lower than 5 are suppressed.
Please note that some of the patients in the cohort are from health boards other than NHS Lothian, this can mean that patients who received chemotherapy or surgery at another health board will not be included in the analysis.
The data used to derive the cohort has been extracted from the following datasets:
The cohort is derived by first using SMR06 to identify patients who had a diagnosis date (incidence date) between 01/01/2015 and 31/12/2016, a surgery flag of 1 with a date of first surgery within 9 months of their diagnosis date and a HER2 flag of 1.
Also using SMR06, records were checked to ensure there were no additional incidences of metastatic breast cancer using clinical stage M and pathological stage M variables.
Since recurrence is not recorded on SMR06 the advanced breast cancer nurse database is used to identify any patients that had developed metastatic breast cancer between diagnosis date and surgery.
Chemocare is used to identify patients who had received neo-adjuvant SACT between 2015 and 2017, this time period is picked to include data for patients diagnosed towards the end of 2016. Neo-adjuvant treatments can be identified by using the Intention variable, Chemotherapy is identified by searching for a drug type of “Cytotoxic” and targeted therapy is identified by searching for a drug type of “Biological Modifer”.
As nurses may not have time to change drug status from “authorised” to “given” at the end of a treatment; both are selected to identify treatments given to the patients in the cohort.
Customers are encouraged to think of the time period of interest for the patients to have received chemotherapy.
Age is derived by calculating the difference between the patients date of birth and the early breast cancer diagnosis date in days and converting to years. Both variables are found in SMR06.
#Age
#There are also no patients who have more than one incidence of breast cancer occuring on different dates
floor(difftime(SMR06_BC_201516$INCIDENCE_DATE,SMR06_BC_201516$DATE_OF_BIRTH, units=c("days"))/365)->SMR06_BC_201516$Age_at_diagnosis
SMR06_BC_201516$Age_at_diagnosis %>% as.numeric() -> SMR06_BC_201516$Age_at_diagnosis
SMR06_BC_201516$Age_inc_group<-cut(SMR06_BC_201516$Age_at_diagnosis,c(0,39,49,59,150))
SMR06_BC_201516$Age_inc_group %>%
fct_recode("Under 40" = "(0,39]",
"40-49" = "(39,49]",
"50-59" = "(49,59]",
"60+" = "(59,150]") -> SMR06_BC_201516$Age_inc_group
SMR06_BC_201516 %>% group_by(Age_inc_group) %>% summarise(`Number of patients`= n_distinct(UPI_NUMBER)) %>%
mutate(`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1))-> Age_at_diagnosis_sum
Age_at_diagnosis_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") | Age_inc_group | Number of patients | % of Cohort |
|---|---|---|
| Under 40 | 9 | 21.4 |
| 40-49 | 13 | 31.0 |
| 50-59 | 11 | 26.2 |
| 60+ | 9 | 21.4 |
Sex is derived by using data from SMR06. 100% of patient’s in the cohort are female.
#Sex
SMR06_BC_201516 %>% group_by(SEX) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER)) %>%
mutate(`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1))-> Sex_sum
Sex_sum$SEX %>% as.factor() %>% fct_recode("Female"="2") %>% as.character() ->Sex_sum$SEX
Sex_sum %>% rename(Sex=SEX) ->Sex_sum
Sex_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") | Sex | Number of patients | % of Cohort |
|---|---|---|
| Female | 42 | 100 |
Ethnicity has been derived using the Ethnic_group variable from SMR06. Low numbers have been suppressed.
SMR06_BC_201516 %>% group_by(ETHNIC_GROUP) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER))-> Ethnicity_sum
Ethnicity_sum$ETHNIC_GROUP %>% as.factor() %>% fct_recode("White-Scottish"="1A",
"White-Other British"="1B",
"Polish"="1L",
"Any other white ethnic background"="1Z",
"Pakistani, Pakistani Scottish or Pakistani British"="3F",
"Indian, Indian Scottish or Indian British"="3G",
"Other Asian, Asian Scottish or Asian British"="3Z",
"Refused/not provided by the patient"="98",
"Not known"="99") %>% as.character() ->Ethnicity_sum$ETHNIC_GROUP
Ethnicity_sum %>% rename(`Ethnic Group`=ETHNIC_GROUP) %>% arrange(`Number of patients`)->Ethnicity_sum
ifelse(Ethnicity_sum$`Number of patients`<5,"\\*",Ethnicity_sum$`Number of patients`)->Ethnicity_sum$`Number of patients`
Ethnicity_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| Ethnic Group | Number of patients |
|---|---|
| Polish | * |
| Indian, Indian Scottish or Indian British | * |
| Other Asian, Asian Scottish or Asian British | * |
| Any other white ethnic background | * |
| Refused/not provided by the patient | * |
| Not known | 6 |
| White-Other British | 7 |
| White-Scottish | 19 |
| Note: | |
| *Number of patients less than 5 have been suppressed |
TNM staging has been derived by combining the Clinical and Pathological staging variables in SMR06.
Customers should specify how specific they want the TNM staging to be, for simplicity and aggregation the letters have been removed from the variable e.g. T1a T1b T1c are combined to T1.
# General TNM Staging ----
# table(SMR06_BC_201516$STAGE_CLINICAL_T, useNA = "ifany")
# table(SMR06_BC_201516$STAGE_CLINICAL_N, useNA = "ifany")
# table(SMR06_BC_201516$STAGE_CLINICAL_M, useNA = "ifany")
#No missing clinical T,N,M
#removing the letters off of the staging variables
gsub("[^0-9\\X]", "", SMR06_BC_201516$STAGE_CLINICAL_T) -> SMR06_BC_201516$STAGE_CLINICAL_T
gsub("[^0-9\\X]", "", SMR06_BC_201516$STAGE_CLINICAL_N) -> SMR06_BC_201516$STAGE_CLINICAL_N
gsub("[^0-9\\X]", "", SMR06_BC_201516$STAGE_PATHOLOGIC_T) ->SMR06_BC_201516$STAGE_PATHOLOGIC_T
gsub("[^0-9\\X]", "", SMR06_BC_201516$STAGE_PATHOLOGIC_N) ->SMR06_BC_201516$STAGE_PATHOLOGIC_N
#combining clinical and pathological staging----
SMR06_BC_201516$STAGE_CLINICAL_T ->SMR06_BC_201516$stage_T
ifelse(is.na(SMR06_BC_201516$stage_T)==1 | SMR06_BC_201516$stage_T=="X",SMR06_BC_201516$STAGE_PATHOLOGIC_T,SMR06_BC_201516$STAGE_CLINICAL_T) %>%
as.factor()->SMR06_BC_201516$stage_T
SMR06_BC_201516$STAGE_CLINICAL_N ->SMR06_BC_201516$stage_N
ifelse(is.na(SMR06_BC_201516$stage_N)==1 | SMR06_BC_201516$stage_N=="X",SMR06_BC_201516$STAGE_PATHOLOGIC_N,SMR06_BC_201516$STAGE_CLINICAL_N) %>%
as.factor()->SMR06_BC_201516$stage_N
SMR06_BC_201516$STAGE_CLINICAL_M ->SMR06_BC_201516$stage_M
ifelse(is.na(SMR06_BC_201516$stage_M)==1 | SMR06_BC_201516$stage_M=="X",SMR06_BC_201516$STAGE_PATHOLOGIC_M,SMR06_BC_201516$STAGE_CLINICAL_M) %>%
as.factor()->SMR06_BC_201516$stage_M
SMR06_BC_201516$stage_T %>%
as.factor() %>%
fct_recode("X Cannot be assessed" = "X")->SMR06_BC_201516$stage_T
SMR06_BC_201516$stage_N %>%
as.factor() %>%
fct_recode("X Cannot be assessed" = "X")->SMR06_BC_201516$stage_N
SMR06_BC_201516$stage_M %>%
as.factor() %>%
fct_recode("X Cannot be assessed" = "X")->SMR06_BC_201516$stage_M
SMR06_BC_201516 %>% group_by(stage_T) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER)) -> Stage_T_sum
SMR06_BC_201516 %>% group_by(stage_N) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER)) -> Stage_N_sum
SMR06_BC_201516 %>% group_by(stage_M) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER)) -> Stage_M_sum
ifelse(Stage_N_sum$`Number of patients`<20,"\\*",Stage_N_sum$`Number of patients`) -> Stage_N_sum$`Number of patients`
ifelse(Stage_M_sum$`Number of patients`<5,"\\*",Stage_M_sum$`Number of patients`) -> Stage_M_sum$`Number of patients`
Stage_T_sum %>% rename(`Stage T`=stage_T) ->Stage_T_sum
Stage_N_sum %>% rename(`Stage N`=stage_N) ->Stage_N_sum
Stage_M_sum %>% rename(`Stage M`=stage_M) ->Stage_M_sum
Stage_T_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") | Stage T | Number of patients |
|---|---|
| 1 | 6 |
| 2 | 22 |
| 3 | 5 |
| 4 | 9 |
Stage_N_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 20 have been suppressed due to the small number of categories in the variable")| Stage N | Number of patients |
|---|---|
| 0 | * |
| 1 | 25 |
| 2 | * |
| Note: | |
| *Number of patients less than 20 have been suppressed due to the small number of categories in the variable |
Stage_M_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| Stage M | Number of patients |
|---|---|
| 0 | 38 |
| X Cannot be assessed | * |
| Note: | |
| *Number of patients less than 5 have been suppressed |
Hormone receptor status is derived by using the oestrogen receptor status and progesterone receptor status variables from SMR06.
#Oestrogen receptor status
SMR06_BC_201516 %>% group_by(OESTROGEN_RECEPTOR_STATUS) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER)) %>%
mutate(`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1))-> ER_status_sum
ER_status_sum$OESTROGEN_RECEPTOR_STATUS %>% as.factor() %>% fct_recode("Negative"="0",
"Positive"="1") %>% as.character() ->ER_status_sum$OESTROGEN_RECEPTOR_STATUS
ER_status_sum %>% rename(`Oestrogen receptor status`=OESTROGEN_RECEPTOR_STATUS) ->ER_status_sum
#Progesterone receptor status
SMR06_BC_201516 %>% group_by(PROGESTERONE_RECEPTOR_STATUS) %>%
summarise(`Number of patients`= n_distinct(UPI_NUMBER)) %>%
mutate(`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1))-> Pro_status_sum
Pro_status_sum$PROGESTERONE_RECEPTOR_STATUS %>% as.factor() %>% fct_recode("Negative"="0",
"Positive"="1","Not known"="9") %>% as.character() ->Pro_status_sum$PROGESTERONE_RECEPTOR_STATUS
Pro_status_sum %>% rename(`Progesterone receptor status`=PROGESTERONE_RECEPTOR_STATUS) ->Pro_status_sum
ER_status_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") | Oestrogen receptor status | Number of patients | % of Cohort |
|---|---|---|
| Negative | 18 | 42.9 |
| Positive | 24 | 57.1 |
Pro_status_sum %>% kable() %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") | Progesterone receptor status | Number of patients | % of Cohort |
|---|---|---|
| Negative | 25 | 59.5 |
| Positive | 9 | 21.4 |
| Not known | 8 | 19.0 |
Time from diagnosis to first chemotherapy has been derived using the diagnosis date from SMR06 and date of first chemotherapy. A future potential analysis would be to investigate the difference between date of first chemotherapy from SMR06 and the date of the first chemotherapy in Chemocare. Customers are encouraged to think of how to define a patients first chemotherapy e.g. neo-adjuvant, dosage, Taxane, Anthracycline, time period
difftime(SMR06_BC_201516$DATE_1ST_CHEMO,SMR06_BC_201516$INCIDENCE_DATE,units=c("days"))->SMR06_BC_201516$Time_to_first_chemo
SMR06_BC_201516$Time_to_first_chemo %>% as.numeric()-> SMR06_BC_201516$Time_to_first_chemo
SMR06_BC_201516 %>% summarise(`Minimum`=min(Time_to_first_chemo,na.rm=TRUE),
`Median`=median(Time_to_first_chemo,na.rm=TRUE),
`Mean`=round2(mean(Time_to_first_chemo,na.rm=TRUE),1),
`Maximum`=max(Time_to_first_chemo,na.rm=TRUE))-> Time_to_first_chemo_sum
kable(Time_to_first_chemo_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
add_header_above(c("Time to first chemotherapy (days)" = 4))| Minimum | Median | Mean | Maximum |
|---|---|---|---|
| 20 | 30 | 33 | 57 |
Time from diagnosis to first chemotherapy has been derived using the diagnosis date from SMR06 and date of first chemotherapy. A future potential analysis would be to investigate the difference between date of first surgery from SMR06 and the date of the first surgery in SMR01. Customers are encouraged to think of how to define a patient’s first surgery e.g. Time period, type of surgery etc.
difftime(SMR06_BC_201516$DATE_1ST_SURGERY,SMR06_BC_201516$INCIDENCE_DATE,units=c("days"))->SMR06_BC_201516$Time_to_first_surgery
SMR06_BC_201516$Time_to_first_surgery %>% as.numeric()-> SMR06_BC_201516$Time_to_first_surgery
SMR06_BC_201516 %>% summarise(`Minimum`=min(Time_to_first_surgery),
`Median`=median(Time_to_first_surgery),
`Mean`=round2(mean(Time_to_first_surgery),1),
`Maximum`=max(Time_to_first_surgery))-> Time_to_first_surgery_sum
kable(Time_to_first_surgery_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
add_header_above(c("Time to first surgery (days)" = 4))| Minimum | Median | Mean | Maximum |
|---|---|---|---|
| 63 | 175 | 176.6 | 258 |
Neo-adjuvant SACT recevied shows the Neo-adjuvent Regimes with a drug status of “Given” or “Authorised” administered between 01/01/2015 and 31/12/2017.
Customers are encouraged to think of how they would like the Regimes aggregated if the patient numbers are small.
Chemocare %>% filter(grepl('Neo-Adjuvant', Intention)) %>%
filter(CHI %in% SMR06_BC_201516$UPI_NUMBER) %>%
filter(`Drug Status` %in% c("Authorised", "Given")) %>%
group_by(Regime) %>% summarise(`Number of patients`= n_distinct(CHI)) %>%
mutate(`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1))-> Regime_received_sum
Regime_received_sum %>% arrange(`Number of patients`) -> Regime_received_sum
ifelse(Regime_received_sum$`Number of patients`<5,"\\*",Regime_received_sum$`Number of patients`)->Regime_received_sum$`Number of patients`
ifelse(Regime_received_sum$`Number of patients`=="\\*","\\*",Regime_received_sum$`% of Cohort`)->Regime_received_sum$`% of Cohort`
kable(Regime_received_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| Regime | Number of patients | % of Cohort |
|---|---|---|
| FEC-D (D) | * | * |
| FEC-D (FEC) | * | * |
| FEC 100 | * | * |
| FEC 80 | * | * |
| KAITLIN ARM 1 LD | * | * |
| KAITLIN ARM 1 MD | * | * |
| T ROSCO TRAST LD | * | * |
| T ROSCO TRAST SC | * | * |
| ZOLEND ADJUVANT | * | * |
| DOCETAXEL&CYCLO | * | * |
| EPI/CYCLO ADJ | * | * |
| DOCETAXEL&CARBO | * | * |
| PACLITAX WKLY | 8 | 19 |
| FEC-D NEO (D) | 26 | 61.9 |
| FEC-D NEO (FEC) | 29 | 69 |
| TRASTUZ MD 3WKLY | 35 | 83.3 |
| TRASTUZ LD 3WKLY | 36 | 85.7 |
| TRASTUZ SC | 36 | 85.7 |
| Note: | ||
| *Number of patients less than 5 have been suppressed |
Type of surgery has been defined by extracting inpatient records from SMR01 covering the time period of 01/01/2015 and 31/12/2017. The OPCS4 codes used to identify breast procedures can be found in the syntax for this report.
Customers should specify OPCS4 codes to avoid any ambiguity around the analysis.
# SMR01colpos <- read_excel("Z:/Oncology Data/Cancer LCDI/SMR/SMR01B col names pos example2.xlsx")
#
#
# SMR01B_from <- SMR01colpos$Position_From
# SMR01B_to<-SMR01colpos$Position_To
# SMR01B_names<-SMR01colpos$Field_name
#
# SMR01B <- read_fwf("N:/Upload/Linked Data/Raw Data/S_mrl_smr01b.dat", fwf_positions(SMR01B_from,SMR01B_to,
# col_names = SMR01B_names),guess_max =500000)
#
# SMR01B %>% filter(UPI_NUMBER %in% Cohort$CHI) ->SMR01B
#
# SMR01B$ADMISSION_DATE %>% ymd() ->SMR01B$ADMISSION_DATE
#
#
# SMR01B %>% filter(ADMISSION_DATE>="2015-01-01" & ADMISSION_DATE<="2017-12-31") ->SMR01B
#
# SMR01B %>% write_xlsx("SMR01_example2.xlsx")
read_xlsx("SMR01_example2.xlsx",guess_max = 2000) -> SMR01
left_join(SMR06_BC_201516,SMR01,by=c("UPI_NUMBER"="UPI_NUMBER")) -> SMR06_BC_201516
#Remove some variables that aren't of interest
#Read in OPCS4 codes
read_xlsx("opcs4_codes.xlsx")-> opcs4_codes
#Match code descriptions to operations
SMR06_BC_201516$MAIN_OP_A_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$MAIN_OPERATION_A,
Code)])
SMR06_BC_201516$MAIN_OP_B_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$MAIN_OPERATION_B,
Code)])
SMR06_BC_201516$OTHER_OP_1A_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$OTHER_OPERATION_1_A,
Code)])
SMR06_BC_201516$OTHER_OP_1B_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$OTHER_OPERATION_1_B,
Code)])
SMR06_BC_201516$OTHER_OP_2A_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$OTHER_OPERATION_2_A,
Code)])
SMR06_BC_201516$OTHER_OP_2B_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$OTHER_OPERATION_2_B,
Code)])
SMR06_BC_201516$OTHER_OP_3A_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$OTHER_OPERATION_3_A,
Code)])
SMR06_BC_201516$OTHER_OP_3B_desc <- with(opcs4_codes,Description[match(SMR06_BC_201516$OTHER_OPERATION_3_B,
Code)])
#Filter for only breast codes
c("B00","B27","B278","B279","B28","B281","B282","B283","B284","B285","B286","B287","B288","B289","B29","B291","B292","B293","B294","B295","B298","B299","B30","B301","B302","B303","B304","B308","B309","B31","B318","B319","B32","B321","B322","B323","B328","B329","B33","B332","B333","B338","B339","B34","B348","B349","B37","B371","B372","B373","B374","B375","B378","B379","B38","B381","B382","B388","B389","B39","B391","B392","B393","B394","B395","B398","B399","B40","B401","B408","B409","S482","S493","U18","U182","U188","U189","Z15","Z151","Z152","Z153","Z154","Z155","Z158") -> Breast_opcs4_codes
##
SMR06_BC_201516 %>% filter(MAIN_OPERATION_A %in% Breast_opcs4_codes) %>% group_by(MAIN_OPERATION_A,MAIN_OP_A_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`) -> MAIN_OP_A_sum
ifelse(MAIN_OP_A_sum$`Number of patients`<5,"\\*",MAIN_OP_A_sum$`Number of patients`)-> MAIN_OP_A_sum$`Number of patients`
ifelse(MAIN_OP_A_sum$`Number of patients`=="\\*","\\*",MAIN_OP_A_sum$`% of Cohort`)-> MAIN_OP_A_sum$`% of Cohort`
kable(MAIN_OP_A_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| MAIN_OPERATION_A | MAIN_OP_A_desc | Number of patients | % of Cohort |
|---|---|---|---|
| B278 | OTHER SPECIFIED TOTAL EXCISION OF BREAST | * | * |
| B291 | RECONSTR OF BREAST USING MYOCUTANEOUS FLAP OF LATISSIMUS DORSI MUSCLE | * | * |
| B301 | INSERTION OF PROSTHESIS FOR BREAST | * | * |
| B303 | REMOVAL OF PROSTHESIS FOR BREAST | * | * |
| B393 | RECONSTR OF BREAST USING FREE DEEP INFERIOR EPIGASTRIC PERFORATOR FLAP | * | * |
| B284 | RE-EXCISION OF BREAST MARGINS | * | * |
| B287 | WIRE GUIDED EXCISION OF LESION OF BREAST | * | * |
| B282 | PARTIAL EXCISION OF BREAST NEC | * | * |
| B375 | LIPOFILLING OF BREAST | 5 | 11.9 |
| B279 | UNSPECIFIED TOTAL EXCISION OF BREAST | 8 | 19 |
| B285 | WIRE GUIDED PARTIAL EXCISION OF BREAST | 18 | 42.9 |
| Note: | |||
| *Number of patients less than 5 have been suppressed |
##
SMR06_BC_201516 %>% filter(MAIN_OPERATION_B %in% Breast_opcs4_codes) %>% group_by(MAIN_OPERATION_B,MAIN_OP_B_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> MAIN_OP_B_sum
ifelse(MAIN_OP_B_sum$`Number of patients`<5,"\\*",MAIN_OP_B_sum$`Number of patients`)-> MAIN_OP_B_sum$`Number of patients`
ifelse(MAIN_OP_B_sum$`Number of patients`=="\\*","\\*",MAIN_OP_B_sum$`% of Cohort`)-> MAIN_OP_B_sum$`% of Cohort`
kable(MAIN_OP_B_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| MAIN_OPERATION_B | MAIN_OP_B_desc | Number of patients | % of Cohort |
|---|---|---|---|
| B301 | INSERTION OF PROSTHESIS FOR BREAST | * | * |
| Note: | |||
| *Number of patients less than 5 have been suppressed |
##
SMR06_BC_201516 %>% filter(OTHER_OPERATION_1_A %in% Breast_opcs4_codes) %>% group_by(OTHER_OPERATION_1_A,OTHER_OP_1A_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> OTHER_OP_1A_sum
ifelse(OTHER_OP_1A_sum$`Number of patients`<5,"\\*",OTHER_OP_1A_sum$`Number of patients`)-> OTHER_OP_1A_sum$`Number of patients`
ifelse(OTHER_OP_1A_sum$`Number of patients`=="\\*","\\*",OTHER_OP_1A_sum$`% of Cohort`)-> OTHER_OP_1A_sum$`% of Cohort`
kable(OTHER_OP_1A_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| OTHER_OPERATION_1_A | OTHER_OP_1A_desc | Number of patients | % of Cohort |
|---|---|---|---|
| B284 | RE-EXCISION OF BREAST MARGINS | * | * |
| B292 | RECONSTRUCTION OF BREAST USING LOCAL FLAP OF SKIN NEC | * | * |
| B375 | LIPOFILLING OF BREAST | * | * |
| B298 | OTHER SPECIFIED RECONSTRUCTION OF BREAST | * | * |
| B291 | RECONSTR OF BREAST USING MYOCUTANEOUS FLAP OF LATISSIMUS DORSI MUSCLE | * | * |
| Note: | |||
| *Number of patients less than 5 have been suppressed |
##
SMR06_BC_201516 %>% filter(OTHER_OPERATION_1_B %in% Breast_opcs4_codes) %>% group_by(OTHER_OPERATION_1_B,OTHER_OP_1B_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> OTHER_OP_1B_sum
ifelse(OTHER_OP_1B_sum$`Number of patients`<5,"\\*",OTHER_OP_1B_sum$`Number of patients`)-> OTHER_OP_1B_sum$`Number of patients`
ifelse(OTHER_OP_1B_sum$`Number of patients`=="\\*","\\*",OTHER_OP_1B_sum$`% of Cohort`)-> OTHER_OP_1B_sum$`% of Cohort`
kable(OTHER_OP_1B_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| OTHER_OPERATION_1_B | OTHER_OP_1B_desc | Number of patients | % of Cohort |
|---|---|---|---|
| B301 | INSERTION OF PROSTHESIS FOR BREAST | * | * |
| Note: | |||
| *Number of patients less than 5 have been suppressed |
##
SMR06_BC_201516 %>% filter(OTHER_OPERATION_2_A %in% Breast_opcs4_codes) %>% group_by(OTHER_OPERATION_2_A,OTHER_OP_2A_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> OTHER_OP_2A_sum
ifelse(OTHER_OP_2A_sum$`Number of patients`<5,"\\*",OTHER_OP_2A_sum$`Number of patients`)-> OTHER_OP_2A_sum$`Number of patients`
ifelse(OTHER_OP_2A_sum$`Number of patients`=="\\*","\\*",OTHER_OP_2A_sum$`% of Cohort`)-> OTHER_OP_2A_sum$`% of Cohort`
kable(OTHER_OP_2A_sum, booktabs=TRUE) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive","bordered"),full_width=F, position="left") %>%
footnote(general="*Number of patients less than 5 have been suppressed")| OTHER_OPERATION_2_A | OTHER_OP_2A_desc | Number of patients | % of Cohort |
|---|---|---|---|
| B285 | WIRE GUIDED PARTIAL EXCISION OF BREAST | * | * |
| B375 | LIPOFILLING OF BREAST | * | * |
| Note: | |||
| *Number of patients less than 5 have been suppressed |
##
SMR06_BC_201516 %>% filter(OTHER_OPERATION_2_B %in% Breast_opcs4_codes) %>% group_by(OTHER_OPERATION_2_B,OTHER_OP_2B_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> OTHER_OP_2B_sum
ifelse(OTHER_OP_2B_sum$`Number of patients`<5,"\\*",OTHER_OP_2B_sum$`Number of patients`)-> OTHER_OP_2B_sum$`Number of patients`
ifelse(OTHER_OP_2B_sum$`Number of patients`=="\\*","\\*",OTHER_OP_2B_sum$`% of Cohort`)-> OTHER_OP_2B_sum$`% of Cohort`
#No data for other op 2b
##
SMR06_BC_201516 %>% filter(OTHER_OPERATION_3_A %in% Breast_opcs4_codes) %>% group_by(OTHER_OPERATION_3_A,OTHER_OP_3A_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> OTHER_OP_3A_sum
ifelse(OTHER_OP_3A_sum$`Number of patients`<5,"\\*",OTHER_OP_3A_sum$`Number of patients`)-> OTHER_OP_3A_sum$`Number of patients`
ifelse(OTHER_OP_3A_sum$`Number of patients`=="\\*","\\*",OTHER_OP_3A_sum$`% of Cohort`)-> OTHER_OP_3A_sum$`% of Cohort`
#No data for other op 3a
##
SMR06_BC_201516 %>% filter(OTHER_OPERATION_3_B %in% Breast_opcs4_codes) %>% group_by(OTHER_OPERATION_3_B,OTHER_OP_3B_desc) %>%
summarise(`Number of patients` = n_distinct(UPI_NUMBER),
`% of Cohort` =round2(`Number of patients`/n_distinct(Cohort$CHI)*100,1)) %>% arrange(`Number of patients`)-> OTHER_OP_3B_sum
ifelse(OTHER_OP_3B_sum$`Number of patients`<5,"\\*",OTHER_OP_3B_sum$`Number of patients`)-> OTHER_OP_3B_sum$`Number of patients`
ifelse(OTHER_OP_3B_sum$`Number of patients`=="\\*","\\*",OTHER_OP_3B_sum$`% of Cohort`)-> OTHER_OP_3B_sum$`% of Cohort`
#No data for other op 3b