What side effect occurs when using UNION DISTINCT - Data type mismatch leads to an error - TOP is used without global sorting - Lack of indexes slows down UNION execution - Removing duplicates can unexpectedly reduce the number of returned rows - UNION creates more load than UNION ALL - Aggregate functions in subqueries can distort data

When using UNION DISTINCT (simply UNION), duplicates are removed from the combined datasets. This can lead to an unexpected reduction in the number of returned rows if the different datasets contain identical records.

For example, if the first dataset returns 10 rows, the second returns 8, but 3 rows are the same, the final result will contain not 18, but 15 rows.

It is also important to consider that removing duplicates requires additional sorting or hashing, which increases the load on the system compared to UNION ALL, which simply concatenates the results without removing duplicates.