generated from opensafely/research-template
/
calculate_tpp_coverage.R
169 lines (129 loc) · 5.16 KB
/
calculate_tpp_coverage.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
################################################################################
# Description: Script to calculate TPP coverage per MSOA according to household
# size of TPP-registered patients and ONS population estimates per MSOA
#
# input:
#
# Author: Emily S Nightingale
# Date: 13/10/2020
#
################################################################################
time_total <- Sys.time()
################################################################################
library(tidyverse)
library(data.table)
library(dtplyr)
# write("Calculating TPP coverage",file="coverage_log.txt")
sink("./coverage_log.txt", type = "output")
options(datatable.old.fread.datetime.character = TRUE)
# ---------------------------------------------------------------------------- #
#----------------------#
# LOAD/CLEAN DATA #
#----------------------#
# * input.csv
# - pull household ID, size and MSOA for all TPP-registered patients
# * msoa_pop.csv
# - total population estimates per MSOA
# - population estimates by single year age
#
# args <- c("./input.csv","./data/SAPE22DT15_mid_2019_msoa.csv", 2)
args = commandArgs(trailingOnly = TRUE)
## TPP-registered patient records (from study definition)
## Include ALL patients with non-missing MSOA in calculation of TPP populations
input <- fread(args[1], data.table = FALSE, na.strings = "") %>%
# Remove individuals w missing/non-England MSOA
filter(grepl("E",msoa) & !is.na(msoa))
## National MSOA population estimates (ONS mid-2019):
msoa_pop <- fread(args[2], data.table = FALSE, na.strings = "") %>%
mutate(msoa = as.factor(`MSOA Code`),
msoa_pop = `All Ages`) %>%
# Filter to England
filter(grepl("E", msoa)) %>%
# Sum older populations
rowwise() %>%
mutate(`70+` = sum(`70-74`:`90+`)) %>%
dplyr::select(msoa, msoa_pop, `70+`) %>%
ungroup()
## MSOA TPP coverage cut off
msoa_cov_cutoff <- args[3]
# ---------------------------------------------------------------------------- #
print("No. MSOAs in England:")
n_distinct(msoa_pop$msoa)
print("No. TPP-registered patients with non-missing MSOA:")
nrow(input)
print("No. unique MSOAs with patients registered in TPP:")
n_distinct(input$msoa)
print("No. TPP-registered patients with non-missing MSOA and non-missing HHID:")
nrow(input[input$household_id > 0,])
print("No. unique MSOAs with patients registered in TPP and non-missing HHID:")
n_distinct(input$msoa[input$household_id > 0])
print("Count probable cases with/without HHID:")
input %>%
group_by((household_id > 0)) %>%
summarise(n_probable = sum(!is.na(primary_care_case_probable)))
# ---------------------------------------------------------------------------- #
#------------------------------------------#
# Aggregate by MSOA and merge with pops #
#------------------------------------------#
input %>%
# Count records per MSOA
group_by(msoa) %>%
tally(name = "tpp_pop_all") %>%
ungroup() -> tpp_pop_all
input %>%
filter(household_id > 0) %>%
# Count records per MSOA
group_by(msoa) %>%
tally(name = "tpp_pop_wHHID") %>%
ungroup() -> tpp_pop_wHHID
tpp_pop_all %>%
full_join(tpp_pop_wHHID) %>%
# Merge MSOAs in OS with total population estimates
left_join(msoa_pop) %>%
mutate(msoa = as.factor(msoa),
tpp_cov_all = tpp_pop_all*100/msoa_pop,
tpp_cov_wHHID = tpp_pop_wHHID*100/msoa_pop,
cov_gt_100 = as.factor(ifelse(tpp_cov_all > 100, "Yes", "No"))) -> tpp_cov
summary(tpp_cov)
over100 <- filter(tpp_cov, cov_gt_100 == "Yes")
write.csv(over100, "./msoa_gt_100_cov.csv", row.names = FALSE)
tpp_msoas <- unique(input$msoa)
write.csv(tpp_msoas, "./msoas_in_tpp.csv", row.names = FALSE)
# Figure
png("./total_vs_tpp_pop.png", height = 800, width = 800)
tpp_cov %>%
pivot_longer(c("tpp_cov_all","tpp_cov_wHHID")) %>%
ggplot(aes(value)) +
geom_histogram(bins = 30, fill = "steelblue") +
facet_wrap(~name) +
labs(x = "TPP coverage per MSOA") +
theme_minimal()
dev.off()
# ---------------------------------------------------------------------------- #
#------------------------------------------#
# Filter on coverage cutoff and save #
#------------------------------------------#
# Filter MSOAs by TPP coverage
exclude_msoa <- tpp_cov %>%
filter(tpp_cov_wHHID < msoa_cov_cutoff) %>%
pull(msoa) %>%
unique()
print(paste0("MSOAs excluded with ",msoa_cov_cutoff,"% coverage cut off: n = ",length(exclude_msoa)))
tpp_cov_incl <- tpp_cov %>%
filter(tpp_cov_wHHID >= msoa_cov_cutoff)
print(paste0("MSOAs included with ",msoa_cov_cutoff,"% coverage cut off: n = ",nrow(tpp_cov_incl)))
summary(tpp_cov_incl$tpp_cov_wHHID)
# Trying to find where the cov = 9% come from...
tpp_cov_incl %>%
mutate(low_cov = (tpp_cov_wHHID < msoa_cov_cutoff)) %>%
group_by(low_cov) %>%
tally()
tpp_cov_incl %>%
filter(tpp_cov_wHHID < msoa_cov_cutoff)
saveRDS(tpp_cov_incl, file = "./tpp_coverage_included.rds")
saveRDS(tpp_cov, file = "./tpp_coverage_all.rds")
write.csv(tpp_cov, "./tpp_coverage_all.csv", row.names = FALSE)
################################################################################
sink()
################################################################################
################################################################################