Using QBASIC's indexing would be faster/"better"

QBASIC has sort very effective indexing methods when working with files open for RANDOM reading & writing. First, you will need to convert the existing text file into a RANDOM access file, which is stored in a special format.
Here's how to perform the conversion:

OPEN mynewfilename$ for RANDOM AS #2 LEN=1000 'change 1000 to the maximum line width in characters required, or just a big number if you're not sure!
OPEN myfilename$ for INPUT AS #1
DO WHILE EOF(1)=0
LINE INPUT #1,theline$
PUT #2,,theline$
numlines=numlines+1
LOOP
CLOSE #1

Now that the file is converted, access it in the following way:

GET #2,5,gotline$ 'stores the contents of line 5 into gotline$
GET #2,numlines,gotline$ 'stores the contents of the last line into gotline$
savethis$="hello"
PUT #2,2,savethis$ 'overwrites line 5 with the string hello
numlines=numlines+1
PUT #2,numlines,savethis$ 'add a new line with the string hello

The file has been edited, so now it will be saved back into its original format:

OPEN mysavefilename$ for OUTPUT AS #1
FOR i=1 to numlines
GET #2,i,myline$
PRINT #1,myline$
NEXT
CLOSE #1