Reading comma delimited file

I am processing moderately sized (18-20 Meg) comma delimited (.csv) files. My code reads the values from the file using the "INPUT#" statement into explicitly defined variables. As in the following:

'input data from file
Input #1, strDRCCaseId, strDRCID, strDRCRefDate, strCurrentStat, _
strSubStat, strStatChngDate, strStatChngReason, strPatSource, _
strBirthYear, strClinicalTrialSubject

The code goes on to validate the data in terms of length, stated valid values, do dates contain valid dates, etc.

If I remove all screen updates during this file processing, I begin to get phantom characters in the result set as though a buffer is getting written to before it gets cleared. For example, my input file contains the following line:

84302,BC4422591,19991117,A,,19991119,"",E,1987,N,21,02

When I test the status (strCurrentStat), it is expecting a 1 character field, and reports an error that the field contains the following:
A99911171 \ó

This appears to me to be the second field (BC4422591) which has been overlayed by the third field (19991117...resulting in 199911171), overlayed by the fourth field (A... resulting in A99911171 plus some other characters I have no idea where they're coming from).

The kicker is that if I slow the code down by painting to the screen immediately prior to executing the INPUT # statement, the error goes away. Here is the code I'm using to update the screen:
txtRecord.Text = dblCount
txtRecord.Refresh
(note: dblCount is a counter containing the line number of the record being worked with).

Any ideas how to fix the problem without slowing the code with screen updates?