I just noticed that a record on the 1810 Census Index (CD-313) was a duplicate of a record from the 1820 Census Index (CD-314), namely:

Census Index:  U.S. Selected Counties, 1810 (Broderbund CD-313):

Mann, Thomas

TN

Stewart Co.

p. 18

500110-00210

Please notice the numerical data consist of six digits for males and five for females.  There is only one census year in which the digits for males and females were asymmetrical and that was 1820 when the digits were, indeed, 6/5.  (See What the Numbers in the Federal Censuses Mean.)  Please compare this entry to the next one.

Census Index:  U.S. Selected Counties, 1820 (Broderbund CD-314):

Mann, Thomas

TN

Stewart Co.

p. 18

500110-00210

The page number and data are the same as in 1820, a virtually impossible coincidence.  Clearly, there has been corruption of the database and this record belongs only in 1820.

I have no idea how many records are involved in this corruption.

The census data from Ancestry.com seem to be similarly afflicted (apparently they bought their data from the same flawed source):

Year	Surname	Given Name(s)	County	State	Page	Other Info	Record Type	Database	ID#
1810	MANN	THOMAS	Stewart	TN	118	500110-00210	Tax List	TN Early Census Index	TNS1a2055726
1820	MANN	THOMAS	Stewart	TN	118	500110-00210	Federal Population Schedule	TN 1820 Federal Census Index	TNS2a792121

Clearly (again), the page number and the "Other Info" apply to the 1820 record, not the 1810 record.

I just read that the entire 1810 Census of Tennessee was lost/destroyed, except for two counties:  Grainger and Rutherford.  So, except for Grainger and Rutherford Cos., there shouldn't be any Tennessee census records on CD-313.