Missing followups

[id2359]

Problem:
If patients miss a followup and then submit a later one, the sequence number of the followup ( being sequential ) will be lower than what it should be. This means when comparing to other patients with the same sequence number, the comparisons ( of scores) will be with a group of people who have been on treatments and so forth for a shorter period of time in an absolute sense, so if a patient does better over time, the average score will be artificially raised ( or artificially lowered) .

Solution:
My suggestion : we handle by assigning sequence number based on nearest to projected collection dates using a stored schedule. If we don't do this, the scale group average scores will be polluted by values which come from patients who have been on a given treatment , say, longer than the rest of the average set which will artificially raise the average score ( or artifically lower it)

For this we need the schedule stored in the registry metadata.

[id2359]

comment from LM 280223

It is important so depends on workload. Can you let us know how much time you think will be required and we can make a decision about making the change now or leaving it until later.

The scheduled timeframes for CRC are:

Baseline

6months

12 months

2 year (24 months)

3 years (36 months)

4 years (48 months)

5 years (60 months)

6 years (72 months)

7 years (84 years)

8 years (96 months)

9 years (108 months)

10 years (120 months)

How will we handle it though if there is an unscheduled capture of date?

The volume of cases in say breast (300+) would mean that pollution of the averages wouldn’t be so affected but this won’t be so for cancers/sites with a smaller volume.