If it's the Overtid: that you want to fill out, then the following does that:
awk 'function overtime(s,e) {
if (s) printf("Overtid: %d\n", (e - s > 800) ? e - s : 0);
}
/^Mødt ind:/ { start = $NF; print }
/^Gået hjem:/ { end = $NF; print }
/^Dato:/ { overtime(start, end); print }
END { overtime(start, end) }' file
This picks out the start and end times from the data and for each Dato: line (which signifies a new record in the data), it prints out the overtime for the previous record. This is also done at the end (the END block).
I've done overtime() as a function as we need to do the same thing in two different places in the script. The if (s) in the function is to ensure that we don't get a bogus Overtid: 0 output when hitting the very first Dato: line in the data.
The print statements in the code passes through the existing data to the output.
Output with your data:
Dato: 22-02-2018
Mødt ind: 700
Gået hjem: 1600
Overtid: 900
Dato: 23-02-2018
Mødt ind: 730
Gået hjem: 1600
Overtid: 870