Meaning of "baseline age" in a medical study From a news report:

Through the Kaplan-Meier plot, researchers indicated that time from schizophrenia diagnoses to death was significantly short for those with HIV than those without (P < 0.001). Controlling for sex, baseline age, and injection drug age, HIV status was significantly associated with mortality in patients with schizophrenia (aOR= 2.31; 95% CI; 1.84-2.89).

What is baseline age? 
P.S. The PDF poster of the study.
 A: The baseline measurements would usually be taken when people entered the study. In a trial they would be taken at recruitment, in a cohort the y would be taken at the time of entry to the cohort however that might be defined (usually when some event occurred), and so on.
From the news report it is not clear when the cohort was established so since it is described as retrospective the baseline values might have been taken when the people first presented to services or they might have been at the time of the start of the study.
A: There are three distinctions of a baseline variable that I can think of:
1) A variable that is constant over the course of the study (e.g., sex, race) or that changes for everyone at the same rate (e.g., age). When the variables change at a uniform rate, these measures are taken at some point just prior to treatment, like the pre-treatment visit.
2) A pre-treatment observation/measurement to be used for measuring outcome (e.g., pre- and post-treatment pain ratings)
3) A pre-treatment observation of a variable that correlates with the outcome.
