cehrbert-data 0.0.1__py3-none-any.whl

cehrbert_data/cohorts/atrial_fibrillation.py

@@ -0,0 +1,45 @@
+from cehrbert_data.cohorts.query_builder import AncestorTableSpec, QueryBuilder, QuerySpec
+from cehrbert_data.const.common import CONDITION_OCCURRENCE, PERSON, VISIT_OCCURRENCE
+
+COHORT_QUERY_TEMPLATE = """
+SELECT
+    co.person_id,
+    FIRST(DATE(vo.visit_start_date)) OVER (PARTITION BY co.person_id
+        ORDER BY DATE(vo.visit_start_date), vo.visit_occurrence_id) AS index_date,
+    FIRST(vo.visit_occurrence_id) OVER (PARTITION BY co.person_id
+        ORDER BY DATE(vo.visit_start_date), vo.visit_occurrence_id) AS visit_occurrence_id
+FROM global_temp.condition_occurrence AS co
+JOIN global_temp.visit_occurrence AS vo
+    ON co.visit_occurrence_id = vo.visit_occurrence_id
+JOIN global_temp.{atrial_fibrillation_concepts} AS c
+    ON co.condition_concept_id = c.concept_id
+"""
+
+ATRIAL_FIBRILLATION_CONCEPT_ID = [313217]
+
+DEPENDENCY_LIST = [PERSON, VISIT_OCCURRENCE, CONDITION_OCCURRENCE]
+
+DEFAULT_COHORT_NAME = "atrial_fibrillation"
+ATRIAL_FIBRILLATION_CONCEPTS = "atrial_fibrillation_concepts"
+
+
+def query_builder():
+    query = QuerySpec(
+        table_name=DEFAULT_COHORT_NAME,
+        query_template=COHORT_QUERY_TEMPLATE,
+        parameters={"atrial_fibrillation_concepts": ATRIAL_FIBRILLATION_CONCEPTS},
+    )
+
+    ancestor_table_specs = [
+        AncestorTableSpec(
+            table_name=ATRIAL_FIBRILLATION_CONCEPTS,
+            ancestor_concept_ids=ATRIAL_FIBRILLATION_CONCEPT_ID,
+            is_standard=True,
+        )
+    ]
+    return QueryBuilder(
+        cohort_name=DEFAULT_COHORT_NAME,
+        dependency_list=DEPENDENCY_LIST,
+        query=query,
+        ancestor_table_specs=ancestor_table_specs,
+    )

cehrbert_data/cohorts/cabg.py

@@ -0,0 +1,72 @@
+from cehrbert_data.cohorts.query_builder import AncestorTableSpec, QueryBuilder, QuerySpec
+from cehrbert_data.const.common import PERSON, PROCEDURE_OCCURRENCE, VISIT_OCCURRENCE
+
+COHORT_QUERY_TEMPLATE = """
+SELECT DISTINCT
+    c.person_id,
+    c.index_date,
+    c.visit_occurrence_id
+FROM
+(
+    SELECT DISTINCT
+        vo.person_id,
+        FIRST(DATE(vo.visit_start_date)) OVER (PARTITION BY po.person_id
+            ORDER BY DATE(vo.visit_start_date), vo.visit_occurrence_id) AS index_date,
+        FIRST(vo.visit_occurrence_id) OVER (PARTITION BY po.person_id
+            ORDER BY DATE(vo.visit_start_date), vo.visit_occurrence_id) AS visit_occurrence_id
+    FROM global_temp.procedure_occurrence AS po
+    JOIN global_temp.visit_occurrence AS vo
+        ON po.visit_occurrence_id = vo.visit_occurrence_id
+    WHERE EXISTS (
+        SELECT 1
+        FROM global_temp.{cabg_concept_table} AS ie
+        WHERE po.procedure_concept_id = ie.concept_id
+    )
+) c
+WHERE c.index_date >= '{date_lower_bound}'
+"""
+
+DEFAULT_COHORT_NAME = "cabg"
+DEPENDENCY_LIST = [PERSON, VISIT_OCCURRENCE, PROCEDURE_OCCURRENCE]
+CABG_INCLUSION_TABLE = "CABG"
+CABG_CONCEPTS = [
+    43528001,
+    43528003,
+    43528004,
+    43528002,
+    4305852,
+    4168831,
+    2107250,
+    2107216,
+    2107222,
+    2107231,
+    4336464,
+    4231998,
+    4284104,
+    2100873,
+]
+
+
+def query_builder(spark_args):
+    query = QuerySpec(
+        table_name=DEFAULT_COHORT_NAME,
+        query_template=COHORT_QUERY_TEMPLATE,
+        parameters={
+            "cabg_concept_table": CABG_INCLUSION_TABLE,
+            "date_lower_bound": spark_args.date_lower_bound,
+        },
+    )
+
+    ancestor_table_specs = [
+        AncestorTableSpec(
+            table_name=CABG_INCLUSION_TABLE,
+            ancestor_concept_ids=CABG_CONCEPTS,
+            is_standard=True,
+        )
+    ]
+    return QueryBuilder(
+        cohort_name=DEFAULT_COHORT_NAME,
+        dependency_list=DEPENDENCY_LIST,
+        query=query,
+        ancestor_table_specs=ancestor_table_specs,
+    )

cehrbert_data/cohorts/coronary_artery_disease.py

@@ -0,0 +1,84 @@
+from cehrbert_data.cohorts.query_builder import AncestorTableSpec, QueryBuilder, QuerySpec
+from cehrbert_data.const.common import CONDITION_OCCURRENCE, PERSON, PROCEDURE_OCCURRENCE, VISIT_OCCURRENCE
+
+COHORT_QUERY_TEMPLATE = """
+WITH prior_graft_stent AS (
+    SELECT
+        po.person_id,
+        po.procedure_date
+    FROM global_temp.procedure_occurrence AS po
+    WHERE EXISTS (
+        SELECT 1
+        FROM global_temp.{graft_stent_table} AS gs
+        WHERE po.procedure_concept_id = gs.concept_id
+    )
+)
+SELECT DISTINCT
+    c.person_id,
+    c.index_date,
+    c.visit_occurrence_id
+FROM
+(
+    SELECT DISTINCT
+        vo.person_id,
+        FIRST(DATE(vo.visit_start_date)) OVER (PARTITION BY co.person_id
+            ORDER BY DATE(vo.visit_start_date), vo.visit_occurrence_id) AS index_date,
+        FIRST(vo.visit_occurrence_id) OVER (PARTITION BY co.person_id
+            ORDER BY DATE(vo.visit_start_date), vo.visit_occurrence_id) AS visit_occurrence_id
+    FROM global_temp.condition_occurrence AS co
+    JOIN global_temp.visit_occurrence AS vo
+        ON co.visit_occurrence_id = vo.visit_occurrence_id
+    WHERE EXISTS (
+        SELECT 1
+        FROM global_temp.{cad_concept_table} AS ie
+        WHERE co.condition_concept_id = ie.concept_id
+    )
+) c
+WHERE NOT EXISTS (
+    -- The patients who had a graft or stent procedures before the index date
+    -- need to be removed from the cohort
+    SELECT 1
+    FROM prior_graft_stent AS exclusion
+    WHERE exclusion.person_id = c.person_id
+        AND c.index_date > exclusion.procedure_date
+) AND c.index_date >= '{date_lower_bound}'
+"""
+
+DEFAULT_COHORT_NAME = "coronary_artery_disease"
+DEPENDENCY_LIST = [PERSON, VISIT_OCCURRENCE, CONDITION_OCCURRENCE, PROCEDURE_OCCURRENCE]
+CAD_INCLUSION_TABLE = "CAD"
+CAD_CONCEPTS = [317576]
+
+PRIOR_PROCEDURE_TABLE = "graft_stent"
+PRIOR_PROCEDURES = [4296227, 42537730, 762043, 44782770, 42537729]
+
+
+def query_builder(spark_args):
+    query = QuerySpec(
+        table_name=DEFAULT_COHORT_NAME,
+        query_template=COHORT_QUERY_TEMPLATE,
+        parameters={
+            "cad_concept_table": CAD_INCLUSION_TABLE,
+            "graft_stent_table": PRIOR_PROCEDURE_TABLE,
+            "date_lower_bound": spark_args.date_lower_bound,
+        },
+    )
+
+    ancestor_table_specs = [
+        AncestorTableSpec(
+            table_name=CAD_INCLUSION_TABLE,
+            ancestor_concept_ids=CAD_CONCEPTS,
+            is_standard=True,
+        ),
+        AncestorTableSpec(
+            table_name=PRIOR_PROCEDURE_TABLE,
+            ancestor_concept_ids=PRIOR_PROCEDURES,
+            is_standard=True,
+        ),
+    ]
+    return QueryBuilder(
+        cohort_name=DEFAULT_COHORT_NAME,
+        dependency_list=DEPENDENCY_LIST,
+        query=query,
+        ancestor_table_specs=ancestor_table_specs,
+    )

cehrbert_data/cohorts/covid.py

@@ -0,0 +1,43 @@
+from cehrbert_data.cohorts.query_builder import QueryBuilder, QuerySpec
+from cehrbert_data.const.common import CONDITION_OCCURRENCE, MEASUREMENT, PERSON, VISIT_OCCURRENCE
+
+COVID_COHORT_QUERY = """
+SELECT DISTINCT
+    c.person_id,
+    DATE_ADD(FIRST(index_date) OVER (PARTITION BY person_id ORDER BY index_date, visit_occurrence_id), 1) AS index_date,
+    FIRST(visit_occurrence_id) OVER (PARTITION BY person_id ORDER BY index_date, visit_occurrence_id) AS visit_occurrence_id
+FROM
+(
+    SELECT DISTINCT
+        m.person_id,
+        FIRST(visit_start_date) OVER (PARTITION BY v.person_id ORDER BY visit_start_date, v.visit_occurrence_id) AS index_date,
+        FIRST(v.visit_occurrence_id) OVER (PARTITION BY v.person_id ORDER BY visit_start_date, v.visit_occurrence_id) AS visit_occurrence_id
+    FROM global_temp.measurement AS m
+    JOIN global_temp.visit_occurrence AS v
+        ON m.visit_occurrence_id = v.visit_occurrence_id
+    JOIN global_temp.concept AS c
+        ON m.value_as_concept_id = c.concept_id
+    WHERE m.measurement_concept_id IN (723475,723479,706178,723473,723474,586515,706177,706163,706180,706181)
+        AND c.concept_name IN ('Detected', 'Positve')
+
+    UNION
+
+    SELECT
+        co.person_id,
+        FIRST(visit_start_date) OVER (PARTITION BY v.person_id ORDER BY visit_start_date, v.visit_occurrence_id) AS index_date,
+        FIRST(v.visit_occurrence_id) OVER (PARTITION BY v.person_id ORDER BY visit_start_date, v.visit_occurrence_id) AS visit_occurrence_id
+    FROM global_temp.condition_occurrence AS co
+    JOIN global_temp.visit_occurrence AS v
+        ON co.visit_occurrence_id = v.visit_occurrence_id
+    WHERE co.condition_concept_id = 37311061
+) c
+"""
+
+DEFAULT_COHORT_NAME = "covid19"
+DEPENDENCY_LIST = [PERSON, VISIT_OCCURRENCE, MEASUREMENT, CONDITION_OCCURRENCE]
+
+
+def query_builder():
+    query = QuerySpec(table_name=DEFAULT_COHORT_NAME, query_template=COVID_COHORT_QUERY, parameters={})
+
+    return QueryBuilder(cohort_name=DEFAULT_COHORT_NAME, dependency_list=DEPENDENCY_LIST, query=query)

cehrbert_data/cohorts/covid_inpatient.py

@@ -0,0 +1,84 @@
+from cehrbert_data.cohorts.query_builder import QueryBuilder, QuerySpec
+from cehrbert_data.const.common import CONDITION_OCCURRENCE, MEASUREMENT, PERSON, VISIT_OCCURRENCE
+
+COVID_COHORT_QUERY = """
+WITH covid_positive AS
+(
+
+    SELECT DISTINCT
+        ROW_NUMBER() OVER(ORDER BY c.person_id, c.index_date) AS test_row_number,
+        c.*
+    FROM
+    (
+        SELECT DISTINCT
+            m.person_id,
+            COALESCE(vo.visit_start_datetime, m.measurement_datetime) AS index_date,
+            vo.visit_occurrence_id,
+            vo.visit_concept_id
+        FROM global_temp.measurement AS m
+        LEFT JOIN global_temp.visit_occurrence AS vo
+            ON m.visit_occurrence_id = vo.visit_occurrence_id
+        WHERE measurement_concept_id IN (723475,723479,706178,723473,723474,586515,706177,706163,706180,706181)
+            AND value_source_value = 'Detected'
+
+        UNION
+
+        SELECT DISTINCT
+            co.person_id,
+            COALESCE(vo.visit_start_datetime, co.condition_start_datetime) AS index_date,
+            vo.visit_occurrence_id,
+            vo.visit_concept_id
+        FROM global_temp.condition_occurrence AS co
+        LEFT JOIN global_temp.visit_occurrence AS vo
+            ON co.visit_occurrence_id = vo.visit_occurrence_id
+        WHERE condition_concept_id = 37311061
+    ) c
+),
+
+covid_test_with_no_visit AS
+(
+    SELECT DISTINCT
+        c.test_row_number,
+        c.person_id,
+        FIRST_VALUE(vo.visit_start_datetime) OVER(PARTITION BY c.person_id ORDER BY vo.visit_start_datetime DESC) AS index_date,
+        FIRST_VALUE(vo.visit_occurrence_id) OVER(PARTITION BY c.person_id ORDER BY vo.visit_start_datetime DESC) AS visit_occurrence_id,
+        FIRST_VALUE(vo.visit_concept_id) OVER(PARTITION BY c.person_id ORDER BY vo.visit_start_datetime DESC) AS visit_concept_id
+    FROM covid_positive AS c
+    JOIN global_temp.visit_occurrence AS vo
+        ON c.person_id = vo.person_id AND c.index_date BETWEEN DATE_ADD(vo.visit_start_date, -7) AND vo.visit_start_date
+    WHERE c.visit_occurrence_id IS NULL
+),
+
+all_covid_tests AS
+(
+    SELECT DISTINCT
+        c.person_id,
+        COALESCE(c.index_date, cn.index_date) AS index_date,
+        COALESCE(c.visit_occurrence_id, cn.visit_occurrence_id) AS visit_occurrence_id,
+        COALESCE(c.visit_concept_id, cn.visit_concept_id) AS visit_concept_id
+    FROM covid_positive AS c
+    LEFT JOIN covid_test_with_no_visit AS cn
+        ON c.test_row_number = cn.test_row_number
+)
+
+SELECT DISTINCT
+    person_id,
+    FIRST_VALUE(vo.index_date) OVER(PARTITION BY vo.person_id ORDER BY vo.index_date) AS index_date,
+    FIRST_VALUE(vo.visit_occurrence_id) OVER(PARTITION BY vo.person_id ORDER BY vo.index_date) AS visit_occurrence_id
+FROM
+(
+    SELECT
+        co.*
+    FROM all_covid_tests AS co
+    WHERE visit_concept_id IN (262, 9203, 9201)
+) vo
+"""
+
+DEFAULT_COHORT_NAME = "covid19"
+DEPENDENCY_LIST = [PERSON, VISIT_OCCURRENCE, MEASUREMENT, CONDITION_OCCURRENCE]
+
+
+def query_builder():
+    query = QuerySpec(table_name=DEFAULT_COHORT_NAME, query_template=COVID_COHORT_QUERY, parameters={})
+
+    return QueryBuilder(cohort_name=DEFAULT_COHORT_NAME, dependency_list=DEPENDENCY_LIST, query=query)

cehrbert_data/cohorts/death.py

@@ -0,0 +1,46 @@
+from cehrbert_data.cohorts.query_builder import QueryBuilder, QuerySpec, create_cohort_entry_query_spec
+from cehrbert_data.const.common import DEATH, PERSON, VISIT_OCCURRENCE
+
+DEATH_COHORT_QUERY = """
+WITH max_death_date_cte AS
+(
+    SELECT
+        person_id,
+        MAX(death_date) AS death_date
+    FROM global_temp.death
+    GROUP BY person_id
+),
+last_visit_start_date AS
+(
+    SELECT
+        person_id,
+        MAX(visit_start_date) AS last_visit_start_date
+    FROM global_temp.visit_occurrence
+    GROUP BY person_id
+)
+
+SELECT
+    d.person_id,
+    d.death_date AS index_date,
+    CAST(null AS INT) AS visit_occurrence_id
+FROM max_death_date_cte AS d
+JOIN last_visit_start_date AS v
+    ON d.person_id = v.person_id
+        AND v.last_visit_start_date <= d.death_date
+"""
+
+DEFAULT_COHORT_NAME = "mortality"
+DEPENDENCY_LIST = [PERSON, DEATH, VISIT_OCCURRENCE]
+
+
+def query_builder():
+    query = QuerySpec(table_name=DEFAULT_COHORT_NAME, query_template=DEATH_COHORT_QUERY, parameters={})
+
+    entry_cohort_query = create_cohort_entry_query_spec(entry_query_template=DEATH_COHORT_QUERY, parameters={})
+
+    return QueryBuilder(
+        cohort_name=DEFAULT_COHORT_NAME,
+        dependency_list=DEPENDENCY_LIST,
+        query=query,
+        entry_cohort_query=entry_cohort_query,
+    )