generated from opensafely/research-template
/
01-createSDtables.R
180 lines (145 loc) · 7.65 KB
/
01-createSDtables.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
# ##==============================================================================
# Analysis filename: 01-createSDtables
# Project: OC evaluation
# Author: MF
# Date: 16/12/2020
# Version: R
# Description: Produce descriptives of GP consultation and OC consultation
# Output to csv files
# Datasets used: input_ori.csv
# Datasets created: None
# Other output: tables: 'tb*.csv'
# Log file: log-01-createSDtables
#
## ==============================================================================
## open log connection to file
sink(here::here("logs", "log-01-createSDtables.txt"))
flag_gtsummaryoperational = TRUE
## library
library(tidyverse)
if (flag_gtsummaryoperational){
library(gtsummary)
}
library(gt)
library(here)
library(webshot)
#webshot::install_phantomjs()
# create directory for saving tables, if not existent
if (!dir.exists(here::here("output", "tables"))){
dir.create(here::here("output", "tables"))
}
## import and pre-process cohort data
df_input <- read_csv(
here::here("output", "input_ori.csv"))
df_cleaned <- df_input %>%
mutate(age_group = factor(cut(age,breaks = c(0,18,40,50,60,70,80, Inf),dig.lab = 2)),
sex = factor(case_when(sex=="F" ~ "Female",sex=="M" ~ "Male",TRUE ~ "Other/Unknown")),
ethnicity = factor(case_when(ethnicity==1 ~ "White",ethnicity==2 ~ "Mixed",ethnicity==3 ~ "Asian",ethnicity==4 ~ "Black",ethnicity==5 ~ "Other",TRUE~"Other")),
care_home_type=factor(case_when(care_home_type=="PC" ~ "Care home",care_home_type=="PN" ~ "Care home",care_home_type=="PS" ~ "Care home",TRUE ~ "Non")),
gp_consult_had = ifelse(is.na(gp_consult_count)|gp_consult_count==0,0,1),
OC_instance=OC_instance_snomed, # setting snomed as the relevant one
oc_instance_had = ifelse(is.na(OC_instance)|OC_instance==0,0,1),
livingalone = ifelse(hh_size<=1,1,0),
has_disability = ifelse(is.na(has_disability),0,has_disability),
imd_quin=ifelse(is.na(imd)|imd=="U",NA,imd),
rural_urban=factor(case_when(rural_urban %in% c(1,2,3,4)~"Urban",rural_urban %in% c(5,6,7,8)~"Rural",TRUE~"Other"))
)
## Redactor code (W.Hulme)
redactor <- function(n, threshold){
# given a vector of frequencies, this returns a boolean vector that is TRUE if
# a) the frequency is <= the redaction threshold and
# b) if the sum of redacted frequencies in a) is still <= the threshold, then the
# next largest frequency is also redacted
n <- as.integer(n)
leq_threshold <- dplyr::between(n, 1, threshold)
n_sum <- sum(n)
# redact if n is less than or equal to redaction threshold
redact <- leq_threshold
# also redact next smallest n if sum of redacted n is still less than or equal to threshold
if((sum(n*leq_threshold) <= threshold) & any(leq_threshold)){
redact[which.min(dplyr::if_else(leq_threshold, n_sum+1L, n))] = TRUE
}
n_redacted <- if_else(redact, NA_integer_, n)
}
## Rates per characteristic
df_to_tbrates <- function(mydf,myvars,flag_save=0,tb_name="latest",n_redact=6) {
mytb=
mydf %>%
group_by_at(myvars) %>%
summarise(
population=n(),
gp_consult=sum(gp_consult_count,na.rm=T),
gp_consult_covg = sum(gp_consult_had,na.rm=T),
oc_instance=sum(OC_instance,na.rm=T),
oc_instance_covg=sum(oc_instance_had,na.rm=T),
) %>%
mutate(
gp_consult=redactor(gp_consult,n_redact),
gp_consult_covg = redactor(gp_consult_covg,n_redact),
oc_instance=redactor(oc_instance,n_redact),
oc_instance_covg=redactor(oc_instance_covg,n_redact),
gp_consult_rate=gp_consult/population,
gp_consult_covg_rate=gp_consult_covg/population,
oc_instance_rate=oc_instance/population,
oc_instance_covg_rate=oc_instance_covg/population
)
if (flag_save){
write.csv(mytb,paste0(here::here("output","tables"),"/",tb_name,".csv"))
}
return(mytb)
}
## OC and GP rates per STP
#tb02_gpcr_stp <- df_to_tbrates(df_cleaned,c("stp"),1,"tb02_gpcr_stp")
### OC and GP rates per practice
#tb03_gpcr_practice <- df_to_tbrates(df_cleaned,c("practice"),1,"tb03_gpcr_practice")
## OC and GP rates per age and gender
tb04_gpcr_agesex <- df_to_tbrates(df_cleaned %>% filter(sex %in% c("Male","Female")),c("age_group","sex"),1,"tb04_gpcr_agesex")
## OC and GP rates by ethnicity
#tb05_gpcr_ethnicity <- df_to_tbrates(df_cleaned,c("ethnicity"),1,"tb05_gpcr_ethnicity")
## OC and GP rates by region and rurality
#tb06_gpcr_ruc <- df_to_tbrates(df_cleaned,c("region","rural_urban"),1,"tb06_gpcr_ruc")
## OC and GP rates by care home status
#tb07_gpcr_care <- df_to_tbrates(df_cleaned,c("care_home_type"),1,"tb07_gpcr_care")
## OC and GP rates by presence of disability
#tb08_gpcr_dis <- df_to_tbrates(df_cleaned,c("has_disability"),1,"tb08_gpcr_dis")
## OC and GP rates by presence of disability
#tb09_gpcr_imd <- df_to_tbrates(df_cleaned,c("imd_quin"),1,"tb09_gpcr_imd")
if (flag_gtsummaryoperational){
## Characteristics of those with any OC consultation, any GP consultation and overall population
desc_vars=c("sex","age","age_group","ethnicity","livingalone","region","imd_quin","rural_urban","care_home_type","oc_instance_had")
(gt_ocpop <- df_cleaned %>% select(desc_vars) %>% tbl_summary(by=oc_instance_had) %>% add_p() %>% add_overall() %>% modify_header(label="**Characteristic | had OC**") %>% modify_spanning_header(c("stat_1", "stat_2") ~ "**Had any OC instance**"))
desc_vars2=c("sex","age","age_group","ethnicity","livingalone","region","imd_quin","rural_urban","care_home_type","gp_consult_had")
(gt_gpcpop <- df_cleaned %>% select(desc_vars2) %>% tbl_summary(by=gp_consult_had) %>% add_p() %>% add_overall() %>% modify_header(label="**Characteristic | had GP consultation**") %>% modify_spanning_header(c("stat_1", "stat_2") ~ "**Had any GP consultation**"))
# Use function from gt package to save table as neat png
gt::gtsave(as_gt(gt_ocpop), file = file.path(here::here("output","tables"), "gt_ocpop.html"))
gt::gtsave(as_gt(gt_gpcpop), file = file.path(here::here("output","tables"), "gt_gpcpop.html"))
# steps to remove input data and strip further where possible
gt_gpcpop$inputs <- NULL
gt_gpcpop$call_list <- NULL
gt_gpcpop$meta_data <- NULL
gt_ocpop$inputs <- NULL
gt_ocpop$call_list <- NULL
gt_ocpop$meta_data <- NULL
## Save dta with actual table data, but underlying data removed
#save(gt_ocpop,file = file.path(here::here("output","tables"), "gt_ocpop.RData"))
#save(gt_gpcpop,file = file.path(here::here("output","tables"), "gt_gpcpop.RData"))
## Save unformatted for disclosiveness assessment
##unlisted_ocpop <- as.data.frame(matrix(unlist(gt_ocpop$table_body), nrow=length(unlist(gt_ocpop$table_body[1]))))
#unlisted_ocpop <- as.data.frame(matrix(unlist(gt_ocpop$table_body)))
#unlisted_ocpop <- as.data.frame(apply(unlisted_ocpop,2,as.character))
#print(class(unlisted_ocpop))
#write_csv(unlisted_ocpop,paste0(here::here("output","tables"),"/gt_ocpop_unlisted.csv"))
##unlisted_gpcpop <- as.data.frame(matrix(unlist(gt_gpcpop$table_body), nrow=length(unlist(gt_gpcpop$table_body[1]))))
#unlisted_gpcpop <- as.data.frame(matrix(unlist(gt_gpcpop$table_body)))
#unlisted_gpcpop <- as.data.frame(apply(unlisted_gpcpop,2,as.character))
#print(class(unlisted_gpcpop))
#write_csv(unlisted_gpcpop,paste0(here::here("output","tables"),"/gt_gpcpop_unlisted.csv"))
#aux<-as.data.frame(gt_ocpop$table_body); aux <- apply(aux,2,as.character)
#print(class(aux))
#write.csv(aux,paste0(here::here("output","tables"),"/gt_ocpop_unformatted.csv"))
#aux<-as.data.frame(gt_gpcpop$table_body);aux <- apply(aux,2,as.character)
#print(class(aux))
#write.csv(aux,paste0(here::here("output","tables"),"/gt_gpcpop_unformatted.csv"))
}
## close log connection
sink()