diff --git a/sdmetrics/single_column/statistical/boundary_adherence.py b/sdmetrics/single_column/statistical/boundary_adherence.py
index 2646b24e..3dc74f0b 100644
--- a/sdmetrics/single_column/statistical/boundary_adherence.py
+++ b/sdmetrics/single_column/statistical/boundary_adherence.py
@@ -43,8 +43,11 @@ def compute(cls, real_data, synthetic_data):
             float:
                 The boundary adherence of the two columns.
         """
-        real_data = pd.Series(real_data).dropna()
-        synthetic_data = pd.Series(synthetic_data).dropna()
+        real_data = pd.Series(real_data)
+        synthetic_data = pd.Series(synthetic_data)
+        if any(pd.isna(real_data)):
+            real_data = real_data.dropna()
+            synthetic_data = synthetic_data.dropna()
 
         if is_datetime(real_data):
             real_data = pd.to_numeric(real_data)
diff --git a/tests/unit/single_column/statistical/test_boundary_adherence.py b/tests/unit/single_column/statistical/test_boundary_adherence.py
index 4ef240bf..828a7730 100644
--- a/tests/unit/single_column/statistical/test_boundary_adherence.py
+++ b/tests/unit/single_column/statistical/test_boundary_adherence.py
@@ -1,5 +1,7 @@
+from datetime import datetime
 from unittest.mock import patch
 
+import numpy as np
 import pandas as pd
 
 from sdmetrics.single_column.statistical import BoundaryAdherence
@@ -31,6 +33,65 @@ def test_compute(self):
         # Assert
         assert result == 0.75
 
+    def test_compute_nans(self):
+        """Test the ``compute`` method with nan values.
+
+        Expect that the nan values in synthetic data are considered as
+        out of bounds if the real data does not also containt nan values.
+        """
+        # Setup
+        real_data = pd.Series([1.0, 2.4, 2.6, 0.8])  # 0.8 -> 2.6
+        real_data_nans = pd.Series([1.0, 2.4, 2.6, 0.8, np.nan])
+        synthetic_data = pd.Series([0.9, 1.8, 2.1, 5.0, np.nan])
+
+        metric = BoundaryAdherence()
+
+        # Run
+        result = metric.compute(real_data, synthetic_data)
+        result_ignore_nans = metric.compute(real_data_nans, synthetic_data)
+
+        # Assert
+        assert result == 0.6
+        assert result_ignore_nans == 0.75
+
+    def test_compute_datetime_nans(self):
+        """Test the ``compute`` method with nan values.
+
+        Expect that the nan values in synthetic data are considered as
+        out of bounds if the real data does not also containt nan values.
+        """
+        # Setup
+        real_data = pd.Series([
+            datetime(2020, 10, 1),
+            datetime(2021, 1, 2),
+            datetime(2021, 9, 12),
+            datetime(2022, 10, 1),
+
+        ], dtype='datetime64[ns]')  # 0.8 -> 2.6
+        real_data_nans = pd.Series([
+            datetime(2020, 10, 1),
+            datetime(2021, 1, 2),
+            datetime(2021, 9, 12),
+            datetime(2022, 10, 1),
+            pd.NaT
+        ], dtype='datetime64[ns]')
+        synthetic_data = pd.Series([
+            datetime(2020, 11, 1),
+            datetime(2021, 1, 2),
+            datetime(2021, 2, 9),
+            pd.NaT,
+        ], dtype='datetime64[ns]')
+
+        metric = BoundaryAdherence()
+
+        # Run
+        result = metric.compute(real_data, synthetic_data)
+        result_ignore_nans = metric.compute(real_data_nans, synthetic_data)
+
+        # Assert
+        assert result == 0.75
+        assert result_ignore_nans == 1
+
     @patch('sdmetrics.single_column.statistical.boundary_adherence.SingleColumnMetric.normalize')
     def test_normalize(self, normalize_mock):
         """Test the ``normalize`` method.