diff --git a/dagster/src/assets/adhoc/health_master_csv_to_gold.py b/dagster/src/assets/adhoc/health_master_csv_to_gold.py
index a86c1bb8..21e28b41 100644
--- a/dagster/src/assets/adhoc/health_master_csv_to_gold.py
+++ b/dagster/src/assets/adhoc/health_master_csv_to_gold.py
@@ -1,3 +1,4 @@
+from functools import reduce
 from io import BytesIO
 from pathlib import Path
 
@@ -85,20 +86,45 @@ def adhoc__health_master_data_transforms(
     )
     sdf = sdf.withColumn("signature", sha2(concat_ws("|", *sorted(sdf.columns)), 256))
 
-    df = sdf.toPandas()
-    df = df.drop_duplicates("health_id_giga")
+    mandatory_columns = [
+        col.name
+        for col in schema_columns
+        if not col.nullable and col.name in sdf.columns
+    ]
+    context.log.info(f"Mandatory columns: {mandatory_columns}")
 
+    country_code = config.country_code
     path = Path(config.filepath)
     stem = path.stem
-    df_failed = df[df["health_id_giga"].isna()]
-    output_filepath = (
-        f"{constants.gold_folder}/dq-results/health-master/failed/{stem}.csv"
+
+    if mandatory_columns:
+        mandatory_null_condition = reduce(
+            lambda a, b: a | b,
+            [f.col(c).isNull() for c in mandatory_columns],
+        )
+        sdf_failed = sdf.filter(mandatory_null_condition)
+        sdf = sdf.filter(~mandatory_null_condition)
+
+        context.log.info(
+            f"Mandatory check: {sdf_failed.count()} rows dropped, {sdf.count()} rows passed"
+        )
+    else:
+        sdf_failed = sdf.limit(0)
+
+    df = sdf.toPandas()
+    df = df.drop_duplicates("health_id_giga")
+
+    df_failed = sdf_failed.toPandas()
+
+    failed_filepath = (
+        f"{constants.gold_folder}/dq-results/failed/{country_code}/{stem}.csv"
     )
     adls_file_client.upload_pandas_dataframe_as_file(
         context=context,
         data=df_failed,
-        filepath=str(output_filepath),
+        filepath=str(failed_filepath),
     )
+    context.log.info(f"Failed rows written to {failed_filepath}: {len(df_failed)} rows")
 
     context.log.info(f"columns: {df.columns.tolist()}")
     context.log.info(f"row count: {len(df)}")
diff --git a/dagster/src/sensors/adhoc.py b/dagster/src/sensors/adhoc.py
index 012096d2..e7fb8afb 100644
--- a/dagster/src/sensors/adhoc.py
+++ b/dagster/src/sensors/adhoc.py
@@ -261,7 +261,7 @@ def health_master__gold_csv_to_deltatable_sensor(
             ),
             "adhoc__health_master_data_transforms": OpDestinationMapping(
                 source_filepath=str(path),
-                destination_filepath=f"{constants.gold_folder}/dq-results/health-master/transforms/{stem}.csv",
+                destination_filepath=f"{constants.gold_folder}/dq-results/transforms/{country_code}/{stem}.csv",
                 metastore_schema=metastore_schema,
                 tier=DataTier.TRANSFORMS,
             ),