Troubleshooting Corrupt EDI Files


On attempting 'Get Data' in Tools...External Data Interface, eNICQ 5 presents a message saying that the data file is corrupt.


NOTE:  Please do NOT submit files to Vermont Oxford Network's support team that contain Protected Health Information.  While this is a complication in our ability to assist you in troubleshooting, it is still possible for us to help.  Our staff are skilled at supporting your file review indirectly by phone while respecting HIPAA limitations.

The file will have to be inspected visually to find the flaw. Typically, this is bad data in either a text field or a datetime field.

Start by taking a screenshot of the message given by eNICQ. Do NOT submit this image to Vermont Oxford Network as it is possible it may contain PHI data.

In the error message, look for any indication of the location of the corrupted record. This may be defined by row, column and position number.

Make a copy of your source file so that you have an original that remains un-edited and a copy that you will make corrections to.

Open the file first in Excel. A small amount of experience in Excel will be helpful. For example, you will want to freeze the header row at the top of the worksheet. If you are unfamiliar with how to do this, it is a common task that most Excel users can show you, or you can check the help file to see how to do this in your version of Excel.

The "row" number given in the error message provides a general starting place to look for corruption. Note that this will refer to the actual row, not the ID number. Scroll down to that row.

The actual corruption might be found in a row nearby the row number listed, usually within a few rows. Because of the corruption, the parsing of row and columns will likely not be exact.

Scroll from left to right on the target row, looking at that row and a couple rows before and after.

Corruption is most often found in text items and datetime items. Look for malformed dates, or for special characters such as quotes, parenthesis or other unexpected characters in text items.

It is ok to see date items with a value of 7/7/1907 or 9/9/1909. It is ok to see blank items. An example of a malformed datetime item would be one that only included the time, or one that had impossible values such as 0/2/00. Corrupt text items generally have something involving special characters, usually quotes. While the system should be able to handle a single quote in a name, such as O'Mallory, any quote use in the vicinity of the corrupted row should be scrutinized.

Make corrections to any corrupted data items as needed, keeping notes of each change so that the end user can be informed of changes to the records.

If you are unable to find the corruption in Excel, try switching to a plain text editor such as Notepad. In Notepad, the separations between items will be marked with a comma. The advantage of using Notepad is that you will see the literal contents of the file. The disadvantage is that the comma separators do not provide column alignment, makind it difficult to match your values up with your header row.

Once you believe you have found and fixed the corruption, try using External Data Interface again to import the data file. It is possible that you may correct the first corrupted elements, only to find that there are others that surface as you work your way through the file. Repeat your review, correction and testing attempts until you have successfully imported, or determined that the file is unable to be salvaged.

Repairing a corrupted file is not an easy task. Make a note of the types of corruption found and consider changes to data entry or file creation processes that might avoid this difficulty in the future.