diff --git a/python/cudf/cudf/__init__.py b/python/cudf/cudf/__init__.py
index 273ab147241..27aaa07f940 100644
--- a/python/cudf/cudf/__init__.py
+++ b/python/cudf/cudf/__init__.py
@@ -17,10 +17,8 @@
     register_index_accessor,
     register_series_accessor,
 )
-from cudf.core.scalar import (
-    NA,
-    Scalar,
-)
+from cudf.core.scalar import Scalar
+
 from cudf.core.index import (
     BaseIndex,
     CategoricalIndex,
@@ -45,6 +43,7 @@
 )
 from cudf.core.dataframe import DataFrame, from_pandas, merge, from_dataframe
 from cudf.core.series import Series
+from cudf.core.missing import NA
 from cudf.core.multiindex import MultiIndex
 from cudf.core.cut import cut
 from cudf.core.algorithms import factorize
diff --git a/python/cudf/cudf/core/missing.py b/python/cudf/cudf/core/missing.py
new file mode 100644
index 00000000000..02bcb7636f4
--- /dev/null
+++ b/python/cudf/cudf/core/missing.py
@@ -0,0 +1,9 @@
+# Copyright (c) 2018-2022, NVIDIA CORPORATION.
+
+
+# Pandas NAType enforces a single instance exists at a time
+# instantiating this class will yield the existing instance
+# of pandas._libs.missing.NAType, id(cudf.NA) == id(pd.NA).
+from pandas import NA
+
+__all__ = ["NA"]
diff --git a/python/cudf/cudf/core/scalar.py b/python/cudf/cudf/core/scalar.py
index 1c81803ed98..189418f651f 100644
--- a/python/cudf/cudf/core/scalar.py
+++ b/python/cudf/cudf/core/scalar.py
@@ -4,14 +4,12 @@
 
 import numpy as np
 import pyarrow as pa
-from pandas._libs.missing import NAType as pd_NAType
 
 import cudf
-from cudf.core.column.column import ColumnBase
+from cudf.api.types import is_scalar
 from cudf.core.dtypes import ListDtype, StructDtype
-from cudf.core.index import BaseIndex
+from cudf.core.missing import NA
 from cudf.core.mixins import BinaryOperand
-from cudf.core.series import Series
 from cudf.utils.dtypes import (
     get_allowed_combinations_for_operator,
     to_cudf_compatible_scalar,
@@ -273,19 +271,19 @@ def _binop_result_dtype_or_error(self, other, op):
         return cudf.dtype(out_dtype)
 
     def _binaryop(self, other, op: str):
-        if isinstance(other, (ColumnBase, Series, BaseIndex, np.ndarray)):
-            # dispatch to column implementation
-            return NotImplemented
-        other = to_cudf_compatible_scalar(other)
-        out_dtype = self._binop_result_dtype_or_error(other, op)
-        valid = self.is_valid and (
-            isinstance(other, np.generic) or other.is_valid
-        )
-        if not valid:
-            return Scalar(None, dtype=out_dtype)
+        if is_scalar(other):
+            other = to_cudf_compatible_scalar(other)
+            out_dtype = self._binop_result_dtype_or_error(other, op)
+            valid = self.is_valid and (
+                isinstance(other, np.generic) or other.is_valid
+            )
+            if not valid:
+                return Scalar(None, dtype=out_dtype)
+            else:
+                result = self._dispatch_scalar_binop(other, op)
+                return Scalar(result, dtype=out_dtype)
         else:
-            result = self._dispatch_scalar_binop(other, op)
-            return Scalar(result, dtype=out_dtype)
+            return NotImplemented
 
     def _dispatch_scalar_binop(self, other, op):
         if isinstance(other, Scalar):
@@ -323,13 +321,3 @@ def _dispatch_scalar_unaop(self, op):
 
     def astype(self, dtype):
         return Scalar(self.value, dtype)
-
-
-class _NAType(pd_NAType):
-    # Pandas NAType enforces a single instance exists at a time
-    # instantiating this class will yield the existing instance
-    # of pandas._libs.missing.NAType, id(cudf.NA) == id(pd.NA).
-    pass
-
-
-NA = _NAType()
diff --git a/python/cudf/cudf/core/udf/typing.py b/python/cudf/cudf/core/udf/typing.py
index ed5fc1d6d23..37e2712f71c 100644
--- a/python/cudf/cudf/core/udf/typing.py
+++ b/python/cudf/cudf/core/udf/typing.py
@@ -17,8 +17,8 @@
 )
 from numba.core.typing.typeof import typeof
 from numba.cuda.cudadecl import registry as cuda_decl_registry
-from pandas._libs.missing import NAType as _NAType
 
+from cudf.core.missing import NA
 from cudf.core.udf import api
 from cudf.core.udf._ops import (
     arith_ops,
@@ -214,7 +214,7 @@ def unify(self, context, other):
 na_type = NAType()
 
 
-@typeof_impl.register(_NAType)
+@typeof_impl.register(type(NA))
 def typeof_na(val, c):
     """
     Tie instances of _NAType (cudf.NA) to our NAType.
diff --git a/python/cudf/cudf/tests/test_udf_masked_ops.py b/python/cudf/cudf/tests/test_udf_masked_ops.py
index 438f46d4266..73bf5160347 100644
--- a/python/cudf/cudf/tests/test_udf_masked_ops.py
+++ b/python/cudf/cudf/tests/test_udf_masked_ops.py
@@ -7,7 +7,7 @@
 from numba import cuda
 
 import cudf
-from cudf.core.scalar import NA
+from cudf.core.missing import NA
 from cudf.core.udf._ops import (
     arith_ops,
     bitwise_ops,