def test_make_df_from_expectations_with_distribution_and_date():
study = StudyDefinition(
population=patients.all(),
bmi=patients.most_recent_bmi(
on_or_after="2010-02-01",
minimum_age_at_measurement=16,
return_expectations={
"rate": "exponential_increase",
"incidence": 0.6,
"float": {
"distribution": "normal",
"mean": 35,
"stddev": 10
},
"date": {
"earliest": "1900-01-01",
"latest": "today"
},
},
),
bmi_date_measured=patients.date_of(
"bmi",
date_format="YYYY-MM",
),
)
population_size = 10000
result = study.make_df_from_expectations(population_size)
assert list(sorted(result.columns)) == ["bmi", "bmi_date_measured"]
# Check that the null-valued rows are aligned with each other
assert (result["bmi"][pd.isnull(
result["bmi"])].fillna(0) == result["bmi_date_measured"][pd.isnull(
result["bmi_date_measured"])].fillna(0)).all()
def test_bmi_dtype_generation():
categorised_codelist = codelist([("X", "Y")], system="ctv3")
categorised_codelist.has_categories = True
study = StudyDefinition(
population=patients.all(),
bmi=patients.most_recent_bmi(
on_or_after="2010-02-01",
minimum_age_at_measurement=16,
),
bmi_date_measured=patients.date_of("bmi", date_format="YYYY-MM"),
)
result = _converters_to_names(study.pandas_csv_args)
assert result == {
"converters": {
"bmi_date_measured": "add_day_to_date"
},
"dtype": {
"bmi": "float"
},
"date_col_for": {
"bmi": "bmi_date_measured"
},
"parse_dates": ["bmi_date_measured"],
}
def test_clinical_events_numeric_value_dtype_generation():
test_codelist = codelist(["X"], system="ctv3")
study = StudyDefinition(
population=patients.all(),
creatinine=patients.with_these_clinical_events(
test_codelist,
find_last_match_in_period=True,
on_or_before="2020-02-01",
returning="numeric_value",
),
creatinine_date=patients.date_of("creatinine", date_format="YYYY-MM"),
)
result = _converters_to_names(study.pandas_csv_args)
assert result == {
"converters": {
"creatinine_date": "add_day_to_date"
},
"dtype": {
"creatinine": "float"
},
"date_col_for": {
"creatinine": "creatinine_date"
},
"parse_dates": ["creatinine_date"],
}
def test_explicit_bmi_fallback():
session = make_session()
weight_code = "X76C7"
bmi_code = "22K.."
patient = Patient(DateOfBirth="1950-01-01")
patient.CodedEvents.append(
CodedEvent(CTV3Code=weight_code, NumericValue=50, ConsultationDate="2001-06-01")
)
patient.CodedEvents.append(
CodedEvent(CTV3Code=bmi_code, NumericValue=99, ConsultationDate="2001-10-01")
)
session.add(patient)
session.commit()
study = StudyDefinition(
population=patients.all(),
BMI=patients.most_recent_bmi(
on_or_after="1995-01-01", on_or_before="2005-01-01",
),
BMI_date_measured=patients.date_of("BMI", date_format="YYYY-MM-DD"),
)
results = study.to_dicts()
assert [x["BMI"] for x in results] == ["99.0"]
assert [x["BMI_date_measured"] for x in results] == ["2001-10-01"]
def test_categorical_clinical_events_with_date_dtype_generation():
categorised_codelist = codelist([("X", "Y")], system="ctv3")
categorised_codelist.has_categories = True
study = StudyDefinition(
population=patients.all(),
ethnicity=patients.with_these_clinical_events(
categorised_codelist,
returning="category",
find_last_match_in_period=True,
),
ethnicity_date=patients.date_of("ethnicity"),
)
result = _converters_to_names(study.pandas_csv_args)
assert result == {
"converters": {
"ethnicity_date": "add_month_and_day_to_date"
},
"date_col_for": {
"ethnicity": "ethnicity_date"
},
"dtype": {
"ethnicity": "category"
},
"parse_dates": ["ethnicity_date"],
}
def test_bmi_when_only_some_measurements_of_child():
session = make_session()
bmi_code = "22K.."
weight_code = "X76C7"
height_code = "XM01E"
patient = Patient(DateOfBirth="1990-01-01")
patient.CodedEvents.append(
CodedEvent(CTV3Code=bmi_code, NumericValue=99, ConsultationDate="1995-01-01")
)
patient.CodedEvents.append(
CodedEvent(CTV3Code=weight_code, NumericValue=50, ConsultationDate="2010-01-01")
)
patient.CodedEvents.append(
CodedEvent(CTV3Code=height_code, NumericValue=10, ConsultationDate="2010-01-01")
)
session.add(patient)
session.commit()
study = StudyDefinition(
population=patients.all(),
BMI=patients.most_recent_bmi(
on_or_after="2005-01-01", on_or_before="2015-01-01",
),
BMI_date_measured=patients.date_of("BMI", date_format="YYYY-MM-DD"),
)
results = study.to_dicts()
assert [x["BMI"] for x in results] == ["0.5"]
assert [x["BMI_date_measured"] for x in results] == ["2010-01-01"]
def test_bmi_rounded():
session = make_session()
weight_code = "X76C7"
height_code = "XM01E"
patient = Patient(DateOfBirth="1950-01-01")
patient.CodedEvents.append(
CodedEvent(
CTV3Code=weight_code, NumericValue=10.12345, ConsultationDate="2001-06-01"
)
)
patient.CodedEvents.append(
CodedEvent(CTV3Code=height_code, NumericValue=10, ConsultationDate="2000-02-01")
)
session.add(patient)
session.commit()
study = StudyDefinition(
population=patients.all(),
BMI=patients.most_recent_bmi("2005-01-01",),
BMI_date_measured=patients.date_of("BMI", date_format="YYYY-MM-DD"),
)
results = study.to_dicts()
assert [x["BMI"] for x in results] == ["0.1"]
assert [x["BMI_date_measured"] for x in results] == ["2001-06-01"]
def test_clinical_event_with_category():
session = make_session()
session.add_all(
[
Patient(),
Patient(
CodedEvents=[
CodedEvent(CTV3Code="foo1", ConsultationDate="2018-01-01"),
CodedEvent(CTV3Code="foo2", ConsultationDate="2020-01-01"),
]
),
Patient(
CodedEvents=[CodedEvent(CTV3Code="foo3", ConsultationDate="2019-01-01")]
),
]
)
session.commit()
codes = codelist([("foo1", "A"), ("foo2", "B"), ("foo3", "C")], "ctv3")
study = StudyDefinition(
population=patients.all(),
code_category=patients.with_these_clinical_events(
codes, returning="category", find_last_match_in_period=True
),
code_category_date=patients.date_of("code_category"),
)
results = study.to_dicts()
assert [x["code_category"] for x in results] == ["", "B", "C"]
assert [x["code_category_date"] for x in results] == ["", "2020", "2019"]
def test_clinical_event_with_numeric_value():
condition_code = "ASTHMA"
_make_clinical_events_selection(
condition_code,
patient_dates=[
None,
# Include date before period starts, which should be ignored
[
("2001-01-01", 1),
("2002-01-01", 2),
("2002-02-01", 3),
("2002-06-01", 4),
],
[("2001-06-01", 7)],
],
)
study = StudyDefinition(
population=patients.all(),
asthma_value=patients.with_these_clinical_events(
codelist([condition_code], "ctv3"),
between=["2001-12-01", "2002-06-01"],
returning="numeric_value",
find_first_match_in_period=True,
),
asthma_value_date=patients.date_of("asthma_value", date_format="YYYY-MM"),
)
results = study.to_dicts()
assert [x["asthma_value"] for x in results] == ["0.0", "2.0", "0.0"]
assert [x["asthma_value_date"] for x in results] == ["", "2002-01", ""]
def test_simple_bmi(include_dates):
session = make_session()
weight_code = "X76C7"
height_code = "XM01E"
patient = Patient(DateOfBirth="1950-01-01")
patient.CodedEvents.append(
CodedEvent(CTV3Code=weight_code, NumericValue=50, ConsultationDate="2002-06-01")
)
patient.CodedEvents.append(
CodedEvent(CTV3Code=height_code, NumericValue=10, ConsultationDate="2001-06-01")
)
session.add(patient)
session.commit()
if include_dates == "none":
bmi_date = None
date_query = None
elif include_dates == "year":
bmi_date = "2002"
date_query = patients.date_of("BMI")
elif include_dates == "month":
bmi_date = "2002-06"
date_query = patients.date_of("BMI", date_format="YYYY-MM")
elif include_dates == "day":
bmi_date = "2002-06-01"
date_query = patients.date_of("BMI", date_format="YYYY-MM-DD")
study = StudyDefinition(
population=patients.all(),
BMI=patients.most_recent_bmi(
on_or_after="1995-01-01", on_or_before="2005-01-01"
),
**dict(BMI_date_measured=date_query) if date_query else {}
)
results = study.to_dicts()
assert [x["BMI"] for x in results] == ["0.5"]
assert [x.get("BMI_date_measured") for x in results] == [bmi_date]
def test_mean_recorded_value_dtype_generation():
test_codelist = codelist(["X"], system="ctv3")
study = StudyDefinition(
population=patients.all(),
bp_sys=patients.mean_recorded_value(
test_codelist,
on_most_recent_day_of_measurement=True,
on_or_before="2020-02-01",
),
bp_sys_date_measured=patients.date_of("bp_sys", date_format="YYYY-MM"),
)
result = _converters_to_names(study.pandas_csv_args)
assert result == {
"converters": {"bp_sys_date_measured": "add_day_to_date"},
"dtype": {"bp_sys": "float"},
"date_col_for": {"bp_sys": "bp_sys_date_measured"},
"parse_dates": ["bp_sys_date_measured"],
}
def test_no_bmi_when_measurement_after_reference_date():
session = make_session()
bmi_code = "22K.."
patient = Patient(DateOfBirth="1900-01-01")
patient.CodedEvents.append(
CodedEvent(CTV3Code=bmi_code, NumericValue=99, ConsultationDate="2001-01-01")
)
session.add(patient)
session.commit()
study = StudyDefinition(
population=patients.all(),
BMI=patients.most_recent_bmi(
on_or_after="1990-01-01", on_or_before="2000-01-01",
),
BMI_date_measured=patients.date_of("BMI", date_format="YYYY-MM-DD"),
)
results = study.to_dicts()
assert [x["BMI"] for x in results] == ["0.0"]
assert [x["BMI_date_measured"] for x in results] == [""]
def test_mean_recorded_value():
code = "2469."
session = make_session()
patient = Patient()
values = [
("2020-02-10", 90),
("2020-02-10", 100),
("2020-02-10", 98),
# This day is outside period and should be ignored
("2020-04-01", 110),
]
for date, value in values:
patient.CodedEvents.append(
CodedEvent(CTV3Code=code, NumericValue=value, ConsultationDate=date)
)
patient_with_old_reading = Patient()
patient_with_old_reading.CodedEvents.append(
CodedEvent(CTV3Code=code, NumericValue=100, ConsultationDate="2010-01-01")
)
patient_with_no_reading = Patient()
session.add_all([patient, patient_with_old_reading, patient_with_no_reading])
session.commit()
study = StudyDefinition(
population=patients.all(),
bp_systolic=patients.mean_recorded_value(
codelist([code], system="ctv3"),
on_most_recent_day_of_measurement=True,
between=["2018-01-01", "2020-03-01"],
),
bp_systolic_date_measured=patients.date_of(
"bp_systolic", date_format="YYYY-MM-DD"
),
)
results = study.to_dicts()
results = [(i["bp_systolic"], i["bp_systolic_date_measured"]) for i in results]
assert results == [("96.0", "2020-02-10"), ("0.0", ""), ("0.0", "")]
