ABSTRACT
There is a wide variety of data in HIV/AIDS research. Biomarker data are common in both clinical and basic studies of HIV; these may include markers of inflammation, pharmacokinetics, drug use, or metabolism, and may be biomarkers commonly used in other disease settings. Genomic data, both human and viral, are also important. Many variables are left censored at detection limits or right censored due to finite follow-up. Many are ordered categorical. There are benefits to having statistical methods that are robust and efficient across a wide variety of data types. Such methods can be quickly applied with confidence in many different situations and may be useful as a first pass in big data settings. Such methods may also be useful in smaller analyses because they provide nice, simple summaries. HIV-positive individuals who have been on long-term antiretroviral therapy appear to be at an increased risk of cardiometabolic diseases, including diabetes, compared to HIV-negative individuals.
