TCGA - Is vital_status subject to follow_up duration?

Question

[LayneSadler]

Let's say that we have 2 years worth of follow ups for a patient, and their vital status is "Alive"

(a) Is it correct to say that beyond 2 years we do not know if they are alive/dead?

(b) Or does vital_status=Alive mean that we just don't have formal follow up data for them after 2 years?

---

From the GDC Data Dictionary:

https://docs.gdc.cancer.gov/Data_Dictionary/viewer/#?_top=1

term	definition	values
vital_status	The survival state of the person registered on the protocol.	Enumeration: (Alive, Dead, Unknown, Not Reported)
days_to_last_follow_up	Time interval from the date of last follow up to the date of initial pathologic diagnosis, represented as a calculated number of days.	One of: (number, null)

Answer 1

[Chris Miller]

Imagine for a moment, that you're looking at this data 200 years from now. The table will not have changed - the vital_status colum will still say "Alive" - even though everyone in the table is obviously dead. The vital status is only valid at the latest time of follow up. You'll want to look into the concept of "censoring" when doing survival analysis

[LayneSadler]

That is the stance I am taking. However, the fact that "Not Reported" is a possible value in the vital_status column made me question my assumptions