Latent Variable Approach to Complier Average Causal Effect (CACE) Analysis in SAS

[David-DuBois]

I am interested in implementing the latent variable approach to CACE analysis in SAS, specifically to estimate effects of an intervention under conditions of non-compliance. I have not been able to locate any source to help with this. I'm looking for something akin to this recent article on how to implement the procedure in Stata (https://journals.sagepub.com/doi/full/10.1177/1536867X221106416).

Thank you!

[sbxkoenk]

Hello,

I don't know about CACE myself, but this is what I find very quickly :

NESUG 2010 conference
Macro for Computing Complier-Average Causal Effect of Treatment (CACE) Using Instrumental Variable Method
Patricia R Houck, MSH, Sati Mazumdar, PhD, Benoit H Mulsant, MD
University of Pittsburgh
https://www.lexjansen.com/nesug/nesug10/po/po12.pdf

AND

Estimating the Complier Average Causal Effect in a Meta-Analysis of Randomized Clinical Trials With Binary Outcomes Accounting for Noncompliance: A Generalized Linear Latent and Mixed Model Approach
Ting Zhou, Jincheng Zhou, James S Hodges, Lifeng Lin, Yong Chen, Stephen R Cole, Haitao Chu
https://pubmed.ncbi.nlm.nih.gov/34564720/

... The above article (not yet published in Am J Epidemiol, I believe) uses SAS PROC MIXED !

BR,
Koen