Assigning time zero in an observational comparative effectiveness study



This is more of a study design question that I have been grappling with since long. I am doing comparative effectiveness study of two drugs (say, two anticoagulants) using claims data. I require the drugs be prescribed within six months of diagnosis of the condition (say, atrial fibrillation). What should my time zero be in this case: should I use 180th day as time zero regardless of the actual date of first prescription, or should I use first date of prescription for each patient? What if one drug is consistently prescribed within first couple of months and the other, in four/fifth months? How do I tease out the effect of the drug from effect of delayed treatment of one of the drugs? Thank you!


i’m not an expert in this area but i think it’s important to be aware of the ‘immortal time’ issue [see eg Problem of immortal time bias in cohort studies: example using statins for preventing progression of diabetes]. They explain it as follows: “immortal time typically arises when the determination of an individual’s treatment status involves a delay or wait period during which follow-up time is accrued—for example, waiting for a prescription to be dispensed after discharge from hospital when the discharge date represents the start of follow-up. This wait period is considered immortal because individuals who end up in the treated or exposed group have to survive (be alive and event free) until the treatment definition is fulfilled. If they have an event before taking up treatment they are in the untreated or unexposed group. Bias is introduced when this period of “immortality” is either misclassified with regards to treatment status or excluded from the analysis…”


Also not an expert in this area. Suissa, however, has a very good paper on immortal time bias, where he describes a number of common pitfalls in assigning time zero in pharmacoepidemiologic studies: