My paper on non-parametric bounds for treatment effects under outcome left-censoring was just published in the Journal of Applied Statistics: Environmental Statistics and Data Science.
Environmental data often exhibit left-censoring, because measurements are subject to a lower limit of detection. For example, instruments might not be able to measure the concentration of a chemical in a sample below a certain threshold. In those cases, all we can say is that the chemical was not detected.
Left-censoring poses problems for traditional analyses. The insight of this paper is to move away from trying to estimate a point effect of an exposure on a left-censored outcome, and instead estimate a range of possible effects that are consistent with the observations.
The paper includes an overview of causal inference for an environmental science audience, and has some sample R code that can be used to implement the proposed analysis.