generated from opensafely/research-template
-
Notifications
You must be signed in to change notification settings - Fork 0
/
calculate_dose_scaled_back.py
151 lines (114 loc) · 7.08 KB
/
calculate_dose_scaled_back.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
import pandas as pd
import numpy as np
import os
import re
import json
from collections import Counter
def match_input_files(file: str) -> bool:
"""Checks if file name has format outputted by cohort extractor"""
pattern = r"^input_20\d\d-(0[1-9]|1[012])-(0[1-9]|[12][0-9]|3[01])\.feather"
return True if re.match(pattern, file) else False
def get_date_input_file(file: str) -> str:
"""Gets the date in format YYYY-MM-DD from input file name string"""
# check format
if not match_input_files(file):
raise Exception("Not valid input file format")
else:
date = result = re.search(r"input_(.*)\.feather", file)
return date.group(1)
def round_5(x):
return int(5 * round(float(x)/5))
OUTPUT_DIR = "output"
for file in os.listdir(OUTPUT_DIR):
if match_input_files(file):
df = pd.read_feather(os.path.join(OUTPUT_DIR, file))
date = get_date_input_file(file)
# e.g date='2020-01-01'
# calculate recommended dose for each doac based on recorded crcl
# deal with null values in numeric value vars
df["crcl"] = df["crcl"].fillna(-1)
# apixaban
apixaban_conditions = [
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 29)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30)),
]
apixaban_values = ["nr", "A2.5", "A5"]
df["apixaban"] = np.select(apixaban_conditions, apixaban_values)
# rivaroxaban
rivaroxaban_conditions = [
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 49)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 50)),
]
rivaroxaban_values = ["nr", "R15", "R20"]
df["rivaroxaban"] = np.select(rivaroxaban_conditions, rivaroxaban_values)
# edoxaban
edoxaban_conditions = [
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 50)),
((df["atrial_fib"] == 1) & (df["crcl"] > 50)),
]
edoxaban_values = ["nr", "E30", "E60"]
df["edoxaban"] = np.select(edoxaban_conditions, edoxaban_values)
# dabigatran
dabigatran_conditions = [
((df["atrial_fib"] == 1) & (df["crcl"] < 30) & (df["crcl"] >= 0)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["crcl"] <= 50)),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30)),
]
dabigatran_values = ["nr", "D110", "D150"]
df["dabigatran"] = np.select(dabigatran_conditions, dabigatran_values)
# now need to check if on correct dose!
conditions = [
(df["doac_dose_calculated"] == df["apixaban"]),
(df["doac_dose_calculated"] == df["rivaroxaban"]),
(df["doac_dose_calculated"] == df["edoxaban"]),
(df["doac_dose_calculated"] == df["dabigatran"]),
]
values = [1, 1, 1, 1]
df["dose_match"] = np.select(conditions, values)
# with af & crcl recorded & exclusions
afcrcl_conditions = [
((df["atrial_fib"] == 1) & (df["crcl_recorded"] == 1)),
((df["atrial_fib"] == 0)),
((df["crcl_recorded"] == 0)),
(df["crcl_exclude"] == 0),
]
afcrcl_values = [1, 0, 0, 0]
df["af_&_crcl"] = np.select(afcrcl_conditions, afcrcl_values)
# dose summary over/under/match
summary_conditions = [
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "A2.5")),
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "A5")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 29) & (df["doac_dose_calculated"] == "A2.5")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 29) & (df["doac_dose_calculated"] == "A5")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["doac_dose_calculated"] == "A2.5")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["doac_dose_calculated"] == "A5")),
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "R15")),
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "R20")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 49) & (df["doac_dose_calculated"] == "R15")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 49) & (df["doac_dose_calculated"] == "R20")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 50) & (df["doac_dose_calculated"] == "R15")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 50) & (df["doac_dose_calculated"] == "R20")),
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "E30")),
((df["atrial_fib"] == 1) & (df["crcl"] < 15) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "E60")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 50) & (df["doac_dose_calculated"] == "E30")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 15) & (df["crcl"] <= 50) & (df["doac_dose_calculated"] == "E60")),
((df["atrial_fib"] == 1) & (df["crcl"] > 50) & (df["doac_dose_calculated"] == "E30")),
((df["atrial_fib"] == 1) & (df["crcl"] > 50) & (df["doac_dose_calculated"] == "E60")),
((df["atrial_fib"] == 1) & (df["crcl"] < 30) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "D110")),
((df["atrial_fib"] == 1) & (df["crcl"] < 30) & (df["crcl"] >= 0) & (df["doac_dose_calculated"] == "D150")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["crcl"] <= 50) & (df["doac_dose_calculated"] == "D110")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["crcl"] <= 50) & (df["doac_dose_calculated"] == "D150")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["doac_dose_calculated"] == "D110")),
((df["atrial_fib"] == 1) & (df["crcl"] >= 30) & (df["doac_dose_calculated"] == "D150")),
((df["atrial_fib"] == 1) & (df["doac_dose_calculated"] == "D75")),
((df["atrial_fib"] == 1) & (df["doac_dose_calculated"] == "E15")),
((df["atrial_fib"] == 1) & (df["doac_dose_calculated"] == "R10")),
((df["atrial_fib"] == 1) & (df["doac_dose_calculated"] == "R2.5")),
]
summary_values = ["over", "over", "match", "over", "under", "match", "over", "over", "match", "over", "under", "match", "over", "over", "match", "over", "under", "match", "over", "over", "match", "over", "under", "match", "under", "under", "under", "under"]
df["dose_summary"] = np.select(summary_conditions, summary_values)
# df.to_csv(f'output/df_with_calculation_{date}.csv') # this will be a new file
df.to_feather(os.path.join(OUTPUT_DIR, file)) # this will overwrite