CHAIN: Subset of variables from the CHAIN project

Description Usage Format Details Source References


The CHAIN project was a longitudinal cohort study of people living with HIV in New York City, which was recruited in 1994 from a large number of medical care and social service agencies serving HIV in New York City. This subset of data pertain to the sixth round of interviews.




A data.frame with 532 observations on the following 8 variables.


log of self reported viral load level, where zero represents an undetectable level.


age at time of the interview


annual family income in 10 intervals


a continuous scale of physical health with a theoretical range between 0 and 100 where better health is associated with higher scale values


a binary measure of poor mental health ( 1=Yes, 0=No )


ordered interval for the CD4 count, which is an indicator of how much damage HIV has caused to the immune system


a three-level ordered variable: 0=Not currently taking HAART (Highly Active AntiretRoviral Therapy) 1=taking HAART but nonadherent, 2=taking HAART and adherent


A missing value in the log virus load level was assigned to individuals who either could not recall their viral load level, did not have a viral load test in the six month preceding the interview, or reported their viral loads as "good" or "bad".



Messeri P, Lee G, Abramson DA, Aidala A, Chiasson MA, Jones JD. (2003). “Antiretroviral therapy and declining AIDS mortality in New York City”. Medical Care 41:512–521.

Search within the mi package
Search all R packages, documentation and source code

Questions? Problems? Suggestions? or email at

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.