173088 Use of instrumental variables with survival analysis in an observational study of cardiovascular and cerebrovascular events

Wednesday, October 29, 2008: 10:45 AM

Samuel A. Bozzette, MD, PhD , RAND Corporation, Santa Monica, CA
Christopher F. Ake, PhD , HIV Neurobehavioral Research Center, UCSD, San Diego, CA
Thomas A. Louis, PhD , Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
In a retrospective study of over 41,000 people in the Veterans Administration system to assess the risk of cardiovascular and cerebrovascular events among those exposed to highly active antiretroviral therapy, we used instrumental variables (IVs) with our survival analysis to attempt to overcome possible selection bias. We assessed how well each of our candidate IVs predicted treatment and considered how plausible it was that each met the requirement of affecting outcome only through treatment. We combined a logistic regression month-by-month binary treatment selection model with a structural outcome model based on proportional hazards regression. Parameters were estimated using a likelihood-type function which contained one factor per non-censored non-post-first-event person-month in the study, with each factor that represented an event month having the general form [{(probability of event given treatment that month) x (probability of treatment that month)} + {(probability of event given no treatment that month) x (probability of no treatment that month)}], where these terms were variously conditioned on instrument, cumulative treatment (in lieu of complete treatment history), and other covariates.

Showing how the likelihood was derived in our case can serve as an illustration of the flexibility of an IV approach in being able to accommodate some non-linear as well as more familiar linear models. At the same time consideration of the modeling assumptions that would be required to hold in our situation for causal inference to be valid serves to emphasize the cautions that should be heeded in attempting to use IVs to accommodate unobserved confounders.

Learning Objectives:
1. Describe how the assumption of no unmeasured confounders implicit or explicit in usual non-instrumental variable modeling approaches to observational studies is replaced by a different statistically unverifiable assumption in an instrumental variable modeling approach, viz., that the instrument (causally) affects outcome only through treatment, the so-called “exclusion condition.” 2. Describe how a researcher might consider whether an instrumental variable method could be appropriate in a proposed study analysis on the basis of assessing how plausible the exclusion condition in the case of the variable under consideration as an instrument would be. 3. Indicate how instrumental variable methods can sometimes be used in non-linear modeling such as survival analysis in addition to more familiar uses such as in linear regression.

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I participated in study design and conduct, data management, statistical analyses, and reporting of results. I am an author on the JAIDS paper reporting results.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.