meta-edc 0.3.39__py3-none-any.whl → 1.4.0__py3-none-any.whl

meta_ae/model_mixins/__init__.py

@@ -1,2 +1,4 @@
 from .ae_review_model_mixin import AeReviewModelMixin
 from .death_report_model_mixin import DeathReportModelMixin
+
+__all__ = ["AeReviewModelMixin", "DeathReportModelMixin"]

meta_ae/model_mixins/ae_review_model_mixin.py

@@ -44,25 +44,25 @@ class AeReviewModelMixin(models.Model):
         verbose_name="Relationship to study drug:",
         max_length=25,
         choices=STUDY_DRUG_RELATIONSHIP,
-        null=True,
+        default="",
     )
 
     ae_expected = models.CharField(
         verbose_name="Based on the protocol, do you believe this event is expected?",
         max_length=25,
         choices=YES_NO,
-        null=True,
+        default="",
     )
 
     ae_action_required = models.CharField(
         verbose_name="If unexpected, do you believe further action is required?",
         max_length=25,
         choices=YES_NO,
-        null=True,
+        default="",
     )
 
     investigator_comments = models.TextField(
-        blank=True, null=True, verbose_name="This Clinical Reviewer's comments:"
+        blank=True, default="", verbose_name="This Clinical Reviewer's comments:"
     )
 
     original_report_agreed = models.CharField(
@@ -70,11 +70,11 @@ class AeReviewModelMixin(models.Model):
         max_length=15,
         choices=YES_NO,
         blank=False,
-        null=True,
+        default="",
         help_text="If No, explain in the narrative below",
     )
 
-    narrative = models.TextField(verbose_name="Narrative", blank=True, null=True)
+    narrative = models.TextField(verbose_name="Narrative", blank=True, default="")
 
     officials_notified = models.DateTimeField(
         blank=True,

meta_ae/model_mixins/death_report_model_mixin.py

@@ -20,7 +20,7 @@ class DeathReportModelMixin(models.Model):
             "If death occurred at hospital / clinic, please give name of the facility"
         ),
         max_length=150,
-        null=True,
+        default="",
         blank=True,
     )
 
@@ -46,14 +46,14 @@ class DeathReportModelMixin(models.Model):
         related_name="secondary_cause_of_death",
         verbose_name="Secondary cause of death",
         help_text=(
-            "Secondary cause of death in the opinion of the " "local study doctor and local PI"
+            "Secondary cause of death in the opinion of the local study doctor and local PI"
         ),
     )
 
     secondary_cause_of_death_other = models.CharField(
         max_length=100,
         blank=True,
-        null=True,
+        default="",
         verbose_name='If "Other" above, please specify',
     )
 

meta_ae/models/__init__.py

@@ -8,3 +8,16 @@ from .death_report import DeathReport
 from .death_report_tmg import DeathReportTmg
 from .death_report_tmg_second import DeathReportTmgSecond
 from .hospitalization import Hospitalization
+
+__all__ = [
+    "AeFollowup",
+    "AeInitial",
+    "AeLocalReview",
+    "AeSponsorReview",
+    "AeSusar",
+    "AeTmg",
+    "DeathReport",
+    "DeathReportTmg",
+    "DeathReportTmgSecond",
+    "Hospitalization",
+]

meta_ae/models/hospitalization.py

@@ -15,7 +15,7 @@ class Hospitalization(
         verbose_name = "Hospitalization"
         verbose_name_plural = "Hospitalization"
         indexes = (
-            HospitalizationModelMixin.Meta.indexes
-            + BaseUuidModel.Meta.indexes
-            + [models.Index(fields=["subject_identifier", "site"])]
+            *HospitalizationModelMixin.Meta.indexes,
+            *BaseUuidModel.Meta.indexes,
+            models.Index(fields=["subject_identifier", "site"]),
         )

meta_ae/pdf_reports/__init__.py

@@ -1,2 +1,4 @@
 from .ae_pdf_report import AePdfReport
 from .death_pdf_report import DeathPdfReport
+
+__all__ = ["AePdfReport", "DeathPdfReport"]

meta_ae/templatetags/meta_ae_extras.py

@@ -1,19 +1,17 @@
 from textwrap import wrap
 
 import arrow
+from clinicedc_constants import OTHER, YES
 from django import template
 from django.conf import settings
 from django.utils.safestring import mark_safe
 from edc_adverse_event.utils import get_adverse_event_app_label
-from edc_constants.constants import OTHER, YES
-from edc_dashboard.utils import get_bootstrap_version
 
 register = template.Library()
 
 
 format_ae_description_template_name = (
-    f"{get_adverse_event_app_label()}/bootstrap{get_bootstrap_version()}/"
-    f"ae_initial_description.html"
+    f"{get_adverse_event_app_label()}/ae_initial_description.html"
 )
 
 
@@ -26,6 +24,6 @@ def format_ae_description(context, ae_initial, wrap_length):
     context["OTHER"] = OTHER
     context["YES"] = YES
     context["ae_initial"] = ae_initial
-    context["sae_reason"] = mark_safe(formatted_sae_reason)  # nosec B308, B703
-    context["ae_description"] = mark_safe(formatted_ae_description)  # nosec B308, B703
+    context["sae_reason"] = mark_safe(formatted_sae_reason)  # nosec B308, B703  # noqa: S308
+    context["ae_description"] = mark_safe(formatted_ae_description)  # nosec B308, B703  # noqa: S308
     return context

meta_analytics/README.rst

@@ -1,5 +1,4 @@
-
-
+pip install -U pdfkit great_tables scipy django_pandas dj_notebook
 
 import pandas as pd
 import numpy as np

meta_analytics/dataframes/__init__.py

@@ -8,6 +8,9 @@ from .constants import (
     endpoint_columns,
 )
 from .get_eos_df import get_eos_df
+from .get_glucose_df import get_glucose_df
+from .get_glucose_fbg_df import get_glucose_fbg_df
+from .get_glucose_fbg_ogtt_df import get_glucose_fbg_ogtt_df
 from .get_last_imp_visits_df import get_last_imp_visits_df
 from .glucose_endpoints import EndpointByDate, GlucoseEndpointsByDate
 from .screening import get_glucose_tested_only_df, get_screening_df
@@ -17,3 +20,27 @@ from .utils import (
     get_unique_subject_identifiers,
     get_unique_visit_codes,
 )
+
+__all__ = [
+    "CASE_EOS",
+    "CASE_FBGS_WITH_FIRST_OGTT",
+    "CASE_FBGS_WITH_SECOND_OGTT",
+    "CASE_FBG_ONLY",
+    "CASE_OGTT",
+    "EndpointByDate",
+    "GlucoseEndpointsByDate",
+    "endpoint_cases",
+    "endpoint_columns",
+    "get_empty_endpoint_df",
+    "get_eos_df",
+    "get_glucose_df",
+    "get_glucose_fbg_df",
+    "get_glucose_fbg_ogtt_df",
+    "get_glucose_tested_only_df",
+    "get_last_imp_visits_df",
+    "get_screening_df",
+    "get_test_string",
+    "get_unique_subject_identifiers",
+    "get_unique_visit_codes",
+    "glucose_endpoints",
+]

meta_analytics/dataframes/constants.py

@@ -2,10 +2,12 @@ CASE_EOS = 7
 CASE_FBGS_WITH_FIRST_OGTT = 2
 CASE_FBGS_WITH_SECOND_OGTT = 3
 CASE_FBG_ONLY = 4
+CASE_FBG_VERY_HIGH = 5
 CASE_OGTT = 1
 EOS_DM_MET = "EOS - Patient developed diabetes"
 OGTT_THRESHOLD_MET = "OGTT >= 11.1"
-
+FBG_BEYOND_THRESHOLD = 20.0
+FBG_THRESHOLD_MET = f"FBG >= {FBG_BEYOND_THRESHOLD}"
 endpoint_columns = [
     "subject_identifier",
     "site_id",
@@ -16,7 +18,7 @@ endpoint_columns = [
     "fbg_value",
     "ogtt_value",
     "fbg_datetime",
-    "fasting",
+    "fasted",
     "endpoint_label",
     "endpoint_type",
     "endpoint",
@@ -29,5 +31,6 @@ endpoint_cases = {
     CASE_FBGS_WITH_FIRST_OGTT: "FBG >= 7 x 2, first OGTT<=11.1",
     CASE_FBGS_WITH_SECOND_OGTT: "FBG >= 7 x 2, second OGTT<=11.1",
     CASE_FBG_ONLY: "FBG >= 7 x 2, OGTT not considered",
+    CASE_FBG_VERY_HIGH: FBG_THRESHOLD_MET,
     CASE_EOS: EOS_DM_MET,
 }

meta_analytics/dataframes/get_eos_df.py

@@ -13,14 +13,26 @@ def get_eos_df() -> pd.DataFrame:
     df_eos = get_eos("meta_prn.endofstudy")
     df_visit = get_subject_visit("meta_subject.subjectvisit")
     df_last_visit = (
-        df_visit.groupby(["subject_identifier", "site"])
-        .agg({"last_visit_code": "max", "last_visit_datetime": "max"})
+        df_visit.groupby(["subject_identifier", "site_id"])
+        .agg({"endline_visit_code": "max", "endline_visit_datetime": "max"})
         .reset_index()
     )
-    df_last_visit = df_last_visit.rename(columns={"site": "site_id"})
+    # df_last_visit = df_last_visit.rename(columns={"site": "site_id"})
 
     df_eos = df_eos.merge(
         df_last_visit, on="subject_identifier", how="left", suffixes=("", "_y")
     )
     df_eos = df_eos.drop(columns=["site_id_y"])
-    return df_eos
+    df_visit_grp = (
+        df_visit.groupby(by=["subject_identifier"])[["baseline_datetime", "visit_datetime"]]
+        .max()
+        .reset_index()
+    )
+    df_visit_grp["followup_days"] = (
+        df_visit_grp["visit_datetime"] - df_visit_grp["baseline_datetime"]
+    ).dt.days
+    return df_eos.merge(
+        df_visit_grp[["subject_identifier", "followup_days"]],
+        on="subject_identifier",
+        how="left",
+    ).reset_index(drop=True)

meta_analytics/dataframes/get_glucose_df.py

@@ -0,0 +1,166 @@
+import numpy as np
+import pandas as pd
+from django_pandas.io import read_frame
+from edc_appointment.constants import MISSED_APPT  # noqa
+from edc_pdutils.dataframes import get_eos, get_subject_consent, get_subject_visit
+
+from meta_subject.models import Glucose, GlucoseFbg
+
+
+def get_glucose_df(subject_identifiers: list[str] | None = None) -> pd.DataFrame:
+    subject_visit_df = (
+        get_subject_visit("meta_subject.subjectvisit", subject_identifiers)
+        .rename(columns={"id": "subject_visit_id"})
+        .query("appt_timing!=@MISSED_APPT")
+    )
+    if subject_identifiers:
+        df_glucose_fbg = read_frame(
+            GlucoseFbg.objects.filter(
+                subject_visit__subject_identifier__in=subject_identifiers
+            ),
+            verbose=False,
+        )
+    else:
+        df_glucose_fbg = read_frame(GlucoseFbg.objects.all(), verbose=False)
+
+    df_glucose_fbg = df_glucose_fbg.rename(
+        columns={"fasting": "fasted", "subject_visit": "subject_visit_id"},
+    )
+    df_glucose_fbg["fasting_hrs"] = np.nan
+    df_glucose_fbg["fasting_hrs"] = df_glucose_fbg["fasting_duration_delta"].apply(
+        lambda x: x.total_seconds() / 3600
+    )
+    df_glucose_fbg.loc[
+        :,
+        ["ogtt_value", "ogtt_units", "ogtt_datetime"],
+    ] = [np.nan, None, pd.NaT]
+    df_glucose_fbg["source"] = "meta_subject.glucosefbg"
+    df_glucose_fbg = df_glucose_fbg[
+        [col for col in df_glucose_fbg.columns if "site_id" not in col]
+    ].merge(
+        subject_visit_df[
+            [
+                "subject_identifier",
+                "site_id",
+                "visit_code",
+                "visit_datetime",
+                "baseline_datetime",
+                "subject_visit_id",
+            ]
+        ],
+        on="subject_visit_id",
+        how="left",
+    )
+
+    if subject_identifiers:
+        df_glucose = read_frame(
+            Glucose.objects.filter(subject_visit__subject_identifier__in=subject_identifiers),
+            verbose=False,
+        ).rename(columns={"subject_visit": "subject_visit_id", "fasting": "fasted"})
+    else:
+        df_glucose = read_frame(Glucose.objects.all(), verbose=False).rename(
+            columns={"subject_visit": "subject_visit_id", "fasting": "fasted"}
+        )
+    df_glucose = df_glucose.rename(
+        columns={"subject_visit": "subject_visit_id", "fasting": "fasted"}
+    )
+    df_glucose["fasting_hrs"] = np.nan
+    df_glucose["fasting_hrs"] = df_glucose["fasting_duration_delta"].apply(
+        lambda x: x.total_seconds() / 3600
+    )
+    df_glucose["source"] = "meta_subject.glucose"
+
+    df_glucose = subject_visit_df[
+        [
+            "subject_identifier",
+            "site_id",
+            "visit_code",
+            "visit_datetime",
+            "baseline_datetime",
+            "subject_visit_id",
+        ]
+    ].merge(
+        df_glucose[[col for col in df_glucose.columns if "site_id" not in col]],
+        on="subject_visit_id",
+        how="left",
+    )
+
+    keep_cols = [
+        "subject_identifier",
+        "site_id",
+        "visit_code",
+        "visit_datetime",
+        "baseline_datetime",
+        "subject_visit_id",
+        "fasted",
+        "fasting_hrs",
+        "fbg_value",
+        "fbg_units",
+        "fbg_datetime",
+        "ogtt_value",
+        "ogtt_units",
+        "ogtt_datetime",
+        "source",
+        "revision",
+        "report_datetime",
+    ]
+    df = df_glucose[keep_cols].merge(
+        df_glucose_fbg[keep_cols],
+        on="subject_visit_id",
+        how="outer",
+        # indicator=True,
+        suffixes=("", "_2"),
+    )
+
+    for suffix in ["", "_2"]:
+        df[[f"fasting_hrs{suffix}", f"fbg_value{suffix}", f"ogtt_value{suffix}"]] = df[
+            [f"fasting_hrs{suffix}", f"fbg_value{suffix}", f"ogtt_value{suffix}"]
+        ].apply(pd.to_numeric)
+        df.loc[
+            (df[f"fbg_units{suffix}"] != "mmol/L (millimoles/L)")
+            & (df[f"fbg_value{suffix}"] >= 0),
+            f"fbg_units{suffix}",
+        ] = "mmol/L (millimoles/L)"
+        df.loc[
+            (df[f"ogtt_units{suffix}"] != "mmol/L (millimoles/L)")
+            & (df[f"ogtt_value{suffix}"] >= 0),
+            f"ogtt_units{suffix}",
+        ] = "mmol/L (millimoles/L)"
+
+    # reconcile all to single column
+    for col in ["fasted", "fbg_value", "ogtt_value", "fbg_datetime", "ogtt_datetime"]:
+        df.loc[(df[col].isna()) & (df[f"{col}_2"].notna()), col] = df[f"{col}_2"]
+
+    df_consent = get_subject_consent("meta_consent.subjectconsent")
+    df_eos = get_eos("meta_prn.endofstudy")
+    df = df.merge(
+        df_consent[["subject_identifier", "gender", "consent_datetime", "dob"]],
+        on="subject_identifier",
+        how="left",
+    ).merge(
+        df_eos[["subject_identifier", "offstudy_datetime", "offstudy_reason"]],
+        on="subject_identifier",
+        how="left",
+    )
+
+    df[[col for col in df.columns if "datetime" in col]] = df[
+        [col for col in df.columns if "datetime" in col]
+    ].apply(lambda x: x.dt.tz_localize(None) if x.dtype == "datetime64[ns, UTC]" else x)
+
+    df["visit_days"] = df["baseline_datetime"].rsub(df["visit_datetime"]).dt.days
+    df["fgb_days"] = df["baseline_datetime"].rsub(df["fbg_datetime"]).dt.days
+    df["ogtt_days"] = df["baseline_datetime"].rsub(df["ogtt_datetime"]).dt.days
+    df["visit_days"] = pd.to_numeric(df["visit_days"], downcast="integer")
+    df["fgb_days"] = pd.to_numeric(df["fgb_days"], downcast="integer")
+    df["ogtt_days"] = pd.to_numeric(df["ogtt_days"], downcast="integer")
+
+    return (
+        df.query(
+            "offstudy_reason != 'Patient fulfilled late exclusion criteria "
+            "(due to abnormal blood values or raised blood pressure at enrolment'"
+        )
+        .copy()
+        .drop(columns=[col for col in df.columns if "_2" in col])
+        .sort_values(by=["subject_identifier", "visit_code"])
+        .reset_index(drop=True)
+    )

meta_analytics/dataframes/get_glucose_fbg_df.py

@@ -0,0 +1,26 @@
+import pandas as pd
+from clinicedc_constants import NO, YES
+from edc_pdutils.dataframes import get_crf
+
+from meta_analytics.dataframes.utils import calculate_fasting_hrs
+
+__all__ = ["get_glucose_fbg_df"]
+
+
+def get_glucose_fbg_df(subject_identifiers: list[str] | None = None) -> pd.DataFrame:
+    """Returns a prepared Dataframe of CRF
+    meta_subject.glucosefbg.
+
+    Note: meta_subject.glucosefbg has only FBG measures.
+    """
+    df = get_crf(
+        model="meta_subject.glucosefbg",
+        subject_identifiers=subject_identifiers or [],
+        subject_visit_model="meta_subject.subjectvisit",
+    )
+    df["source"] = "meta_subject.glucosefbg"
+    df = df.rename(columns={"fbg_fasting": "fasting"})
+    df.loc[(df["fasting"] == "fasting"), "fasting"] = YES
+    df.loc[(df["fasting"] == "non_fasting"), "fasting"] = NO
+    df = calculate_fasting_hrs(df)
+    return df.reset_index(drop=True)

meta_analytics/dataframes/get_glucose_fbg_ogtt_df.py

@@ -0,0 +1,21 @@
+import pandas as pd
+from edc_pdutils.dataframes import get_crf
+
+from .utils import calculate_fasting_hrs
+
+__all__ = ["get_glucose_fbg_ogtt_df"]
+
+
+def get_glucose_fbg_ogtt_df(subject_identifiers: list[str] | None = None) -> pd.DataFrame:
+    """Returns a prepared Dataframe of CRF meta_subject.glucose.
+
+    Note: meta_subject.glucose has FBG and OGTT measures.
+    """
+    df = get_crf(
+        model="meta_subject.glucose",
+        subject_identifiers=subject_identifiers or [],
+        subject_visit_model="meta_subject.subjectvisit",
+    )
+    df["source"] = "meta_subject.glucose"
+    df = calculate_fasting_hrs(df)
+    return df.reset_index(drop=True)

meta_analytics/dataframes/get_last_imp_visits_df.py

@@ -5,12 +5,12 @@ from django_pandas.io import read_frame
 from edc_pdutils.dataframes import get_appointments, get_crf
 
 
-class InvalidLotNumber(Exception):
+class InvalidLotNumber(Exception):  # noqa: N818
     pass
 
 
 def site_cond(df, site_id):
-    return (0 == 0) if not site_id else df.site_id == site_id
+    return (0 == 0) if not site_id else df.site_id == site_id  # noqa: PLR0133
 
 
 def get_last_imp_visits_df(
@@ -28,8 +28,8 @@ def get_last_imp_visits_df(
         lot_number_model_cls = django_apps.get_model("meta_pharmacy.lotnumber")
         try:
             lot_obj = lot_number_model_cls.objects.get(lot_no=lot_no)
-        except ObjectDoesNotExist:
-            raise ObjectDoesNotExist("The lot number given is invalid")
+        except ObjectDoesNotExist as e:
+            raise ObjectDoesNotExist("The lot number given is invalid") from e
 
     df_meds = get_crf(
         "meta_subject.studymedication", subject_visit_model="meta_subject.subjectvisit"
@@ -37,11 +37,14 @@ def get_last_imp_visits_df(
     df_meds = (
         df_meds[(df_meds.refill == "Yes") & (site_cond(df_meds, site_id))]
         .groupby(by=["subject_identifier", "site_id"])
-        .agg({"last_visit_code": "max", "last_visit_datetime": "max"})
+        .agg({"endline_visit_code": "max", "endline_visit_datetime": "max"})
         .reset_index()
     )
     df_meds = df_meds.rename(
-        columns={"last_visit_code": "imp_visit_code", "last_visit_datetime": "imp_visit_date"}
+        columns={
+            "endline_visit_code": "imp_visit_code",
+            "endline_visit_datetime": "imp_visit_date",
+        }
     )
     df_meds.reset_index()
 
@@ -59,8 +62,8 @@ def get_last_imp_visits_df(
     df_off = df_off.set_index("subject_identifier")
 
     df_meds = df_meds.set_index("subject_identifier")
-    df_final = pd.merge(
-        df_meds, df_off, left_index=True, right_index=True, how="outer"
+    df_final = df_meds.merge(
+        df_off, left_index=True, right_index=True, how="outer"
     ).reset_index()
 
     # merge with RandomizationList if lot_obj
@@ -97,5 +100,4 @@ def get_last_imp_visits_df(
     # calculate days since the IMP visit
     df_final["days_since"] = pd.to_datetime("today").normalize() - df_final.imp_visit_date
     df_final["days_until"] = df_final.next_appt_datetime - pd.to_datetime("today").normalize()
-    df_final = df_final.reset_index()
-    return df_final
+    return df_final.reset_index()

meta_analytics/dataframes/glucose_endpoints/__init__.py

@@ -1,2 +1,4 @@
 from .endpoint_by_date import EndpointByDate
 from .glucose_endpoints_by_date import GlucoseEndpointsByDate
+
+__all__ = ["EndpointByDate", "GlucoseEndpointsByDate"]