Entry and Retention in Medical Care among HIV-diagnosed Persons: A Meta-analysis

Gary Marks; Lytt I. Gardner; Jason Craw; Nicole Crepaz


AIDS. 2011;24(17):2665-2678. 

In This Article

Abstract and Introduction


Objective: A 'test and treat' strategy to reduce HIV transmission hinges on linking and retaining HIV patients in care to achieve the full benefit of antiretroviral therapy. We integrated empirical findings and estimated the percentage of HIV-positive persons in the United States who entered HIV medical care soon after their diagnosis; and were retained in care during specified assessment intervals.
Methods: We comprehensively searched databases and bibliographic lists to identify studies that collected data from May 1995 through 2009. Separate meta-analyses were conducted for entry into care and retention in care (having multiple HIV medical visits during specified assessment intervals) stratified by methodological variables. All analyses used random-effects models.
Results: Overall, 69% [95% confidence interval (CI) 66–71%, N = 53 323, 28 findings] of HIV-diagnosed persons in the United States entered HIV medical care averaged across time intervals in the studies. Seventy-two percent (95% CI 67–77%, N = 6586, 12 findings) entered care within 4 months of diagnosis. Seventy-six percent (95% CI 66–84%, N = 561, 15 findings) entered care after testing HIV-positive in emergency/urgent care departments and 67% (95% CI 64–70%, N = 52 762, 13 findings) entered care when testing was done in community locations. With respect to retention in care, 59% (95% CI 53–65%, N = 75 655, 28 findings) had multiple HIV medical care visits averaged across assessment intervals of 6 months to 3–5 years. Retention was lower during longer assessment intervals.
Conclusion: Entry and retention in HIV medical care in the United States are moderately high. Improvement in both outcomes will increase the success of a test and treat strategy.


HIV investigators[1–5] recently have been discussing a 'test and treat' strategy for reducing transmission of HIV infection at the population level. The idea is to increase the number of people who become aware that they are HIV-positive through expanded testing efforts, assist these people in entering medical care, and initiate HAART at earlier stages of disease, for example, when CD4 cell counts are above 350/μl. Patients on HAART who see their medical providers regularly are more likely to have suppressed viral load than those who see their providers irregularly.[6] Those with suppressed viral load are less infectious and, thus, less likely to transmit HIV infection through risk behaviors.[7,8]

We could better gauge the potential success of a test and treat strategy if we knew the proportion of HIV-diagnosed persons who enter care soon after their diagnosis and the proportion of HIV patients who are seen regularly for HIV medical care. Knowing these proportions can help guide resource allocation and programmatic efforts to improve these outcomes, and establish benchmarks for monitoring progress.

Several studies have examined entry and retention in care among HIV-diagnosed persons, but there is no single source for estimating these outcomes. Disparate conclusions can be reached if the results of one study or a small subset of studies are considered in isolation of the others. Quantitatively combining the findings from multiple studies through meta-analytic techniques can generate an integrated conclusion.

We performed meta-analyses of US studies conducted in the era of HAART to estimate the proportion of HIV-diagnosed persons who enter HIV medical care soon after their diagnosis (e.g., within 4 months of diagnosis); and the proportion of HIV-diagnosed persons who are retained in care, conceptualized in the analysis as having multiple HIV medical visits during specified assessment intervals (e.g., three or more visits in prior 12 months). Additionally, we examined each of these outcomes according to several methodological variables that may affect estimates of these outcomes (e.g., assessment interval for entry or retention in care, recruitment settings, sampling methods, source of outcome measures, and number of participants in the study).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: