To compare the accuracy of linkage to care metrics for patients diagnosed with HIV using retention in care and virological suppression as the gold standards of effective linkage.
A retrospective cohort study of patients aged 18 years and older with newly diagnosed HIV infection in the City of Philadelphia, 2007–2008.
Times from diagnosis to clinic visits or laboratory testing were used as linkage measures. Outcome variables included being retained in care and achieving virological suppression, 366–730 days after diagnosis. Positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) for each linkage measure and retention, and virological suppression outcomes are described.
Of the 1781 patients in the study, 503 (28.2%) were retained in care in the Ryan White system and 418 (23.5%) achieved virological suppression 366–730 days after diagnosis. The linkage measure with the highest PPV for retention was having 2 clinic visits within 365 days of diagnosis, separated by 90 days (74.2%). Having a clinic visit between 21 and 365 days after diagnosis had both the highest NPV for retention (94.5%) and the highest adjusted AUC for retention (0.872). Having 2 tests within 365 days of diagnosis, separated by 90 days, had the highest adjusted AUC for virological suppression (0.780).
Linkage measures associated with clinic visits had higher PPV and NPV for retention, whereas linkage measures associated with laboratory testing had higher PPV and NPV for retention. Linkage measures should be chosen based on the outcome of interest.
Supplemental Digital Content is Available in the Text.
*Center for Healthcare Improvement and Patient Safety and
†Division of Infectious Diseases, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA; and
‡City of Philadelphia Department of Public Health, AIDS Activities Coordinating Office, Philadelphia, PA.
Correspondence to: Sara C. Keller, MD, MPH, Center for Healthcare Improvement and Patient Safety and Division of Infectious Diseases, Perelman School of Medicine, University of Pennsylvania, 230 N 21st Street, Unit 907, Philadelphia, PA 19103 (e-mail: email@example.com).
Supported by an unrestricted grant from the Agency for Healthcare Research and Quality, Grant (GIM) 400-4239-4-555854-XXXX-2446-2192 (S.C.K.).
Portions of these data were presented at ID Week 2012, 2012, San Diego, CA.
S.C.K. was supported by an unrestricted grant from the Agency for Healthcare Research and Quality (AHRQ), Grant (GIM) 400-4239-4-555854-XXXX-2446-2192. B.R.Y. was supported by the National Institutes of Health/Institute of Mental Health (K23-MH-097647-01A1). K.A.B. was supported by a Health Resources and Services Administration Ryan White Grant (H89HA0013) and by a Centers for Disease Control and Prevention grant for FOA PS08-802 (5U62PS001044-04). M.G.E. has no conflicts of interest to disclose.
S.C.K. designed the study, acquired the data, performed statistical analyses, performed the data analysis and interpretation, drafted the manuscript, and provided critical revision of the manuscript for important intellectual content. M.G.E. performed statistical analyses, provided administrative support, acquired data, and provided critical revision of the manuscript for important intellectual content. B.R.Y. performed data interpretation and provided critical revision of the manuscript for important intellectual content. K.A.B. assisted in study design, performed data interpretation and provided critical revision of the manuscript for important intellectual content.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.jaids.com).
Received January 21, 2013
Accepted April 10, 2013