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 | def compute_completeness_predictor_per_visit(
self,
data: Data,
start_date: datetime,
end_date: datetime,
care_site_levels: Union[bool, str, List[str]],
stay_types: Union[bool, str, Dict[str, str]],
care_site_ids: List[int],
care_site_short_names: List[str],
care_site_specialties: Union[bool, List[str]],
care_sites_sets: Union[str, Dict[str, str]],
specialties_sets: Union[str, Dict[str, str]],
extra_data: Data,
length_of_stays: List[float],
note_types: Union[str, Dict[str, str]],
age_ranges: List[int],
gender_source_values: Union[bool, str, Dict[str, str]],
condition_types: Union[bool, str, Dict[str, str]],
provenance_sources: Union[bool, str, Dict[str, str]],
stay_sources: Union[bool, str, Dict[str, str]],
drg_sources: Union[bool, str, Dict[str, str]],
**kwargs
):
r"""Script to be used by [``compute()``][edsteva.probes.base.BaseProbe.compute]
The ``per_visit`` algorithm computes $c_(t)$ the availability of clinical documents linked to patients' administrative stays:
$$
c(t) = \frac{n_{with\,doc}(t)}{n_{visit}(t)}
$$
Where $n_{visit}(t)$ is the number of administrative stays, $n_{with\,doc}$ the number of visits having at least one document and $t$ is the month.
"""
self._note_columns = list(set(["note_type"]).intersection(set(self._index)))
self._metrics = ["c", "n_visit", "n_visit_with_note"]
check_tables(
data=data,
required_tables=[
"note",
"visit_occurrence",
"care_site",
"fact_relationship",
],
)
care_site_relationship = prepare_care_site_relationship(
data=data,
)
self.care_site_relationship = care_site_relationship
person = (
prepare_person(data, gender_source_values)
if (age_ranges or gender_source_values)
else None
)
cost = prepare_cost(data, drg_sources) if drg_sources else None
visit_occurrence = prepare_visit_occurrence(
data=data,
start_date=start_date,
end_date=end_date,
stay_types=stay_types,
length_of_stays=length_of_stays,
provenance_sources=provenance_sources,
stay_sources=stay_sources,
cost=cost,
person=person,
age_ranges=age_ranges,
)
if condition_types:
check_tables(
data=data,
required_tables=[
"condition_occurrence",
],
)
conditions = prepare_condition_occurrence(
data,
extra_data=None,
visit_occurrence=None,
source_systems="ORBIS",
diag_types=None,
condition_types=condition_types,
start_date=start_date,
end_date=end_date,
)[["visit_occurrence_id", "condition_type"]].drop_duplicates()
visit_occurrence = visit_occurrence.merge(conditions, on="visit_occurrence_id")
care_site = prepare_care_site(
data=data,
care_site_ids=care_site_ids,
care_site_short_names=care_site_short_names,
care_site_relationship=care_site_relationship,
care_site_specialties=care_site_specialties,
care_sites_sets=care_sites_sets,
specialties_sets=specialties_sets,
)
note = prepare_note(data, note_types)
hospital_visit = get_hospital_visit(self, note, visit_occurrence, care_site)
hospital_name = CARE_SITE_LEVEL_NAMES["Hospital"]
note_predictor_by_level = {hospital_name: hospital_visit}
# UF selection
if not hospital_only(care_site_levels=care_site_levels):
if extra_data: # pragma: no cover
visit_detail = prepare_visit_detail(data, start_date, end_date)
uf_visit, uc_visit, uh_visit = get_visit_detail(
self,
extra_data=extra_data,
note=note,
visit_occurrence=visit_occurrence,
visit_detail=visit_detail,
care_site=care_site,
)
uf_name = CARE_SITE_LEVEL_NAMES["UF"]
note_predictor_by_level[uf_name] = uf_visit
uc_name = CARE_SITE_LEVEL_NAMES["UC"]
note_predictor_by_level[uc_name] = uc_visit
uh_name = CARE_SITE_LEVEL_NAMES["UH"]
note_predictor_by_level[uh_name] = uh_visit
pole_visit = get_pole_visit(uf_visit, care_site, care_site_relationship)
pole_name = CARE_SITE_LEVEL_NAMES["Pole"]
note_predictor_by_level[pole_name] = pole_visit
else:
logger.info("Note data are only available at hospital level")
care_site_levels = ["Hospital"]
# Concatenate all predictors
note_predictor = concatenate_predictor_by_level(
predictor_by_level=note_predictor_by_level,
care_site_levels=care_site_levels,
)
if is_koalas(note_predictor):
note_predictor.spark.cache()
return compute_completeness(self, note_predictor)
|