generated from opensafely/research-template
-
Notifications
You must be signed in to change notification settings - Fork 0
/
plots.py
196 lines (166 loc) · 6.56 KB
/
plots.py
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
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
import pandas as pd
from pathlib import Path
from ebmdatalab import charts
from utilities import (
OUTPUT_DIR,
ANALYSIS_DIR,
plot_measures,
redact_small_numbers,
convert_binary,
calculate_rate
)
Path("output/monthly/joined/redacted").mkdir(parents=True, exist_ok=True)
Path("output/weekly/joined/redacted").mkdir(parents=True, exist_ok=True)
for frequency in ["monthly", "weekly"]:
for test in ["alt", "ast", "bilirubin", "gi_illness", "hepatitis"]:
# plot rates
df = pd.read_csv(
OUTPUT_DIR / f"{frequency}/joined/measure_{test}_rate.csv", parse_dates=["date"]
)
df["rate"] = calculate_rate(df, "value")
df = redact_small_numbers(
df, 5, test, "population", "rate", "date"
)
df.to_csv(
OUTPUT_DIR / f"{frequency}/joined/redacted/measure_{test}_rate.csv", index=False
)
plot_measures(
df=df,
filename=f"{frequency}/joined/plot_{test}",
column_to_plot="rate",
title="",
y_label="Rate per 1000",
as_bar=False,
)
# plot count
plot_measures(
df=df,
filename=f"{frequency}/joined/plot_{test}_count",
column_to_plot=test,
title="",
y_label="Count",
as_bar=False,
)
# deciles chart
df_practice = pd.read_csv(
OUTPUT_DIR / f"{frequency}/joined/measure_{test}_practice_rate.csv",
parse_dates=["date"],
)
df_practice["rate"] = calculate_rate(df_practice, "value")
decile_chart = charts.deciles_chart(
df_practice,
period_column="date",
column="rate",
show_outer_percentiles=False,
ylabel="Rate per 1000",
)
decile_chart.savefig(
OUTPUT_DIR / f"{frequency}/joined/deciles_chart_{test}.png", bbox_inches="tight"
)
# plot out of range rates
if test in ["alt", "ast", "bilirubin"]:
if test == "bilirubin":
input_file = f"{frequency}/joined/measure_{test}_oor_ref_rate.csv"
numerator = f"{test}_numeric_value_out_of_ref_range"
output_file = f"measure_{test}_oor_ref_rate.csv"
else:
input_file = f"{frequency}/joined/measure_{test}_oor_rate.csv"
numerator = f"{test}_numeric_value_out_of_range"
output_file = f"measure_{test}_oor_rate.csv"
df_oor = pd.read_csv(
OUTPUT_DIR / input_file,
parse_dates=["date"],
)
df_oor = redact_small_numbers(
df_oor, 5, numerator, test, "value", "date"
)
df_oor.to_csv(
OUTPUT_DIR / f"{frequency}/joined/redacted/{output_file}", index=False
)
df_oor["rate"] = calculate_rate(df_oor,"value")
plot_measures(
df=df_oor,
filename=f"{frequency}/joined/plot_{test}_oor",
column_to_plot="rate",
title="",
y_label="Rate per 1000",
as_bar=False,
)
# plot counts
plot_measures(
df=df_oor,
filename=f"{frequency}/joined/plot_{test}_oor_count",
column_to_plot=numerator,
title="",
y_label="Count",
as_bar=False,
)
# chart for those with recent test and out of range
df_oor_cov = pd.read_csv(
OUTPUT_DIR / f"{frequency}/joined/measure_{test}_oor_recent_cov_rate.csv",
parse_dates=["date"],
)
df_oor_cov = redact_small_numbers(
df_oor_cov, 5, numerator, test, "value", "date"
)
df_oor_cov.to_csv(
OUTPUT_DIR / f"{frequency}/joined/redacted/measure_{test}_oor_recent_cov_rate.csv",index=False
)
df_oor_cov["rate"] = calculate_rate(df_oor_cov, "value")
convert_binary(df_oor_cov, "recent_positive_covid_test", "Yes", "No")
plot_measures(
df=df_oor_cov,
filename=f"{frequency}/joined/plot_{test}_oor_recent_cov",
column_to_plot="rate",
title="",
y_label="Rate per 1000",
as_bar=False,
category="recent_positive_covid_test",
)
# plot count
plot_measures(
df=df_oor_cov,
filename=f"{frequency}/joined/plot_{test}_oor_recent_cov_count",
column_to_plot=numerator,
title="",
y_label="Count",
as_bar=False,
category="recent_positive_covid_test",
)
for d in ["age_band_months", "region"]:
demographic_df = pd.read_csv(
OUTPUT_DIR / f"{frequency}/joined/measure_{test}_{d}_rate.csv",
parse_dates=["date"],
)
if d == "age_band_months":
demographic_df = demographic_df[
demographic_df["age_band_months"] != "missing"
]
elif d == "region":
demographic_df = demographic_df[demographic_df["region"].notnull()]
demographic_df["rate"] = calculate_rate(demographic_df, "value")
demographic_df = redact_small_numbers(
demographic_df, 5, test, "population", "rate", "date"
)
demographic_df.to_csv(
OUTPUT_DIR / f"{frequency}/joined/redacted/measure_{test}_{d}_rate.csv",index=False
)
plot_measures(
df=demographic_df,
filename=f"{frequency}/joined/plot_{test}_{d}",
column_to_plot="rate",
title="",
y_label="Rate per 1000",
as_bar=False,
category=d,
)
#count
plot_measures(
df=demographic_df,
filename=f"{frequency}/joined/plot_{test}_{d}_count",
column_to_plot=test,
title="",
y_label="Count",
as_bar=False,
category=d,
)