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,
include_measurement_date=True,
include_month=True,
return_expectations={
"rate": "exponential_increase",
"incidence": 0.6,
"float": {
"distribution": "normal",
"mean": 35,
"stddev": 10
},
"date": {
"earliest": "1900-01-01",
"latest": "today"
},
},
),
)
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_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",
include_measurement_date=True,
include_month=True,
include_day=True,
),
)
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_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,
include_measurement_date=True,
include_month=True,
),
)
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_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",
include_measurement_date=True,
include_month=True,
include_day=True,
),
)
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_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",
include_measurement_date=True,
include_month=True,
include_day=True,
),
)
results = study.to_dicts()
assert [x["BMI"] for x in results] == ["0.0"]
assert [x["BMI_date_measured"] for x in results] == [""]
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",
include_measurement_date=True,
include_month=True,
include_day=True,
),
)
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_bmi_with_zero_values():
session = make_session()
weight_code = "X76C7"
height_code = "XM01E"
patient = Patient(DateOfBirth="1950-01-01")
patient.CodedEvents.append(
CodedEvent(CTV3Code=weight_code, NumericValue=0, ConsultationDate="2001-06-01")
)
patient.CodedEvents.append(
CodedEvent(CTV3Code=height_code, NumericValue=0, ConsultationDate="2001-06-01")
)
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] == ["0.0"]
assert [x["BMI_date_measured"] for x in results] == ["2001-06-01"]
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
bmi_kwargs = {}
elif include_dates == "year":
bmi_date = "2002"
bmi_kwargs = dict(include_measurement_date=True)
elif include_dates == "month":
bmi_date = "2002-06"
bmi_kwargs = dict(include_measurement_date=True, include_month=True)
elif include_dates == "day":
bmi_date = "2002-06-01"
bmi_kwargs = dict(
include_measurement_date=True, include_month=True, include_day=True
)
study = StudyDefinition(
population=patients.all(),
BMI=patients.most_recent_bmi(
on_or_after="1995-01-01", on_or_before="2005-01-01", **bmi_kwargs
),
)
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_no_bmi_when_measurements_of_child():
session = make_session()
bmi_code = "22K.."
patient = Patient(DateOfBirth="2000-01-01")
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="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] == ["0.0"]
assert [x["BMI_date_measured"] for x in results] == [""]
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]
"5": 0.1,
"3": 0.1
}
},
"incidence": 0.75,
},
),
## COVARIATES
bmi=patients.most_recent_bmi(
between=["2010-02-28", "2020-02-29"],
minimum_age_at_measurement=16,
include_measurement_date=True,
include_month=True,
return_expectations={
"date": {},
"float": {
"distribution": "normal",
"mean": 35,
"stddev": 10
},
"incidence": 0.95,
},
),
smoking_status=patients.categorised_as(
{
"S": "most_recent_smoking_code = 'S'",
"E": """
most_recent_smoking_code = 'E' OR (
most_recent_smoking_code = 'N' AND ever_smoked
)
""",
return_expectations={
"incidence": 0.2,
"date": {
"earliest": "1950-01-01",
"latest": "today"
},
},
),
# https://github.com/ebmdatalab/tpp-sql-notebook/issues/10
bmi=patients.most_recent_bmi(
on_or_after="2010-02-01",
minimum_age_at_measurement=16,
include_measurement_date=True,
include_month=True,
return_expectations={
"incidence": 0.6,
"float": {
"distribution": "normal",
"mean": 35,
"stddev": 10
},
},
),
# https://github.com/ebmdatalab/tpp-sql-notebook/issues/35
bp_sys=patients.mean_recorded_value(
systolic_blood_pressure_codes,
on_most_recent_day_of_measurement=True,
on_or_before="2020-02-01",
include_measurement_date=True,
include_month=True,
return_expectations={
"incidence": 0.6,
# https://github.com/ebmdatalab/tpp-sql-notebook/issues/37
rural_urban=patients.address_as_of(
"2020-02-01", returning="rural_urban_classification"
),
# https://github.com/ebmdatalab/tpp-sql-notebook/issues/54
stp=patients.registered_practice_as_of("2020-02-01", returning="stp_code"),
# region - one of NHS England 9 regions
region=patients.registered_practice_as_of("2020-02-01", returning="nhse_region_name"),
# https://github.com/ebmdatalab/tpp-sql-notebook/issues/10
bmi=patients.most_recent_bmi(
on_or_after="2010-02-01",
minimum_age_at_measurement=16,
include_measurement_date=True,
include_month=True,
),
# https://github.com/ebmdatalab/tpp-sql-notebook/issues/6
smoking_status=patients.categorised_as(
{
"S": "most_recent_smoking_code = 'S'",
"E": """
most_recent_smoking_code = 'E' OR (
most_recent_smoking_code = 'N' AND ever_smoked
)
""",
"N": "most_recent_smoking_code = 'N' AND NOT ever_smoked",
"M": "DEFAULT"
},
