which could hence result in reporter bias. Finally, due to the timing of data collection as well as the longitudinal consistency of the alcohol measures administered, we were unable to account for concurrent alcohol in these models. It is likely, particularly when considering the females’ findings that the increase in depressive symptoms is running in tandem with a parallel increase in alcohol use and misuse. This concurrent association may be complex, with a bi-directional aspect operating on a much finer timescale than might be elucidated using annually collected panel data. Unfortunately, concurrent consistent measures of alcohol use are not available, precluding our ability to address this possibility. We were also unable to conduct cross-lag analyses due to inconsistent alcohol measures across the age range of interest.
