In a group of HIV-positive patients under observation since their first exposure to ART or monitored off of ART, BMD changes over one year will be determined. For each subject, the investigators will also determine associations between changes in BMD and 1) ART initiation, 2) cumulative viremia (measured by copy-years viremia), and 3) inflammation (evaluated through the measurement of interleukin-6 {IL-6}, tumor necrosis factor alpha {TNF-a}, high-sensitivity c-reactive protein {hsCRP}).

Hypotheses: BMD will decrease less in persons initiated on ART than those monitored off of ART, after excluding those subjects treated with tenofovir.

BMD will decrease most significantly in HIV-positive subjects with the highest levels of cumulative viremia.

HIV-positive persons with highest cumulative viremia will have the highest levels of inflammation, as measured by pro-inflammatory cytokines.

Additionally, the investigators will evaluate fracture incidence in a 5% National Medicare sample and fracture association with the use of varying ART medications among dual-eligible persons in Medicare and Medicaid datasets.

Hypotheses: Fracture incidence will be greater in HIV-positive subjects compared to HIV-negative subjects.Fracture incidence will be greatest in subjects with the shortest duration of ART exposure.

Inclusion Criteria:

• treatment naïve patients seen in the 1917 Clinic between January 1, 2000 and May 1, 2010
• currently under care at the time of the initiation of the study (>1 clinic visit in the past 12 months)

Exclusion Criteria:

• history of chronic renal failure (estimated GFR <30ml/min)
• known diagnosis of a metabolic bone disease (i.e. osteoporosis, primary hyperparathyroidism, Paget Disease, Osteogenesis Imperfecta)
• multiple myeloma, cancer, untreated thyroid disease, or inflammatory bowel disease, or persons currently treated with or plans to begin an osteoporosis-specific medication (including estrogen)
• treatment with oral glucocorticoids and anticonvulsants

For primary outcomes: 1917 Clinic cohort. For secondary outcomes: Medicare sample.