2
$\begingroup$

My question is whether my data are a survival analysis problem.

I have 3 groups (2 interventions and 1 control). The 200 patients are all followed up for 6 months only. The follow-up is done by observing the records in a public hospital for any readmissions. I am making the assumption that a patient does not visit any other hospital.

I am interested in the time (in days) to readmission of the patients after their initial discharge. The time of the initial admission was the time they entered the study. The start date is not the same for all individuals – a range of 2 years, more or less.

Basically I have different start times, but the censoring time is the same for all censored patients. It is 180 days for all censored patients.

  • Does this affect the survival curve or the comparison between the groups?
  • Is there another analysis that I should apply that nbetter matches my data and setting?

When running the Kaplan Meier curve in SPSS, I get a straight line (going down) at the 180 days (x axis). Is this OK?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
4
$\begingroup$

Yes, these are survival analysis/event history analysis data.

The beginning of time in survival analysis is rarely calendar time, but is the first day the individual was observed in the study. This affects your interpretation in that intervention/treatment effects are understood to affect person time (e.g. to affect the hazard function in a given abstracted notion of "days since start of observation", or "days since diagnosis" or "days since treatment"... depending on the nature of your study design), rather than affecting the hazard function in terms of calendar time (i.e. you are not trying to estimate change in hazard due to treatment on June, 3rd, 2014).

If you only followed people for 6 months: that's 180 days; unless everyone experienced readmission by 180 days, there should be some right censoring, and the survival curve should not plummet to 0 at 180 days.

Improve this answer
$\endgroup$
3
  • $\begingroup$ thank you for your answer! I just discovered what went wrong with the plummet curve. Coding issue. $\endgroup$
    – Lefkios Paikousis
    Commented Jul 7, 2014 at 9:40
  • $\begingroup$ just came across <basic.northwestern.edu/statguidefiles/kaplan_ass_viol.html#Lack of independence> and I was wondering whether by having all my censored cases at 180 days would violate the censoring independence assumption. Additionally, I have approximately 50% of the patients that survived. Does this violate any assumptions? (i.e. high censoring) $\endgroup$
    – Lefkios Paikousis
    Commented Jul 11, 2014 at 14:51
  • 2
    $\begingroup$ If 50% survived, then 50% did not, and cannot be censored by definition, if "(right) censored" means "observed without ever experiencing the outcome." $\endgroup$
    – Alexis
    Commented Jul 11, 2014 at 15:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.