To fully benefit from antiretroviral therapy (ART), HIV-infected individuals must be aware of their infection, link to and consistently engage in care, and receive and adhere to HIV treatment.1,2 Retention in HIV care is a critical step in this process, associated with improved survival, decreased HIV-related complications, and reduced HIV transmission to others.3–13 These benefits, in part, are because of the strong relationship between retention in care and HIV viral suppression.14–17
Multiple studies indicate that patients retained in care are more likely to achieve viral suppression compared with those not engaged in regular care.14–17 Among 2197 South Carolina HIV-infected residents entering care between 2004 and 2007, 50% were retained in care. Patients retained in care had a greater decrease in HIV viral load from baseline compared with those with suboptimal retention in care.16 Likewise, among 8235 patients followed for 12 months at 6 academic HIV clinics, retention in care, regardless of the measure used, was significantly associated with viral suppression.14
However, previous studies did not account for HIV disease severity, as measured by CD4 count. It is unclear whether retention in care is more strongly associated with viral suppression in patients with lower versus higher CD4 counts. Retention in care is potentially more important for medication adherence in people with lower CD4 counts, who may have higher pill burdens and greater chances for drug toxicity related to treatment of opportunistic infections and other HIV complications.18,19 Alternatively, patients with higher CD4 counts, who may experience minimal symptoms related to their HIV infection, may require more consistent engagement in care to promote high levels of adherence to therapy.20 This study extends previous research by evaluating whether the association between retention in care and viral suppression differs among patients with different disease severity.
Study Sample and Data Collection
We conducted a series of annual cross-sectional analyses using data from the HIV Research Network (HIVRN), a consortium of 22 clinics that provide care to HIV-infected patients.21,22 Data were abstracted from medical records at each site and sent to a data coordinating center after personal identifying information was removed. After quality control and verification, data were combined across sites to produce a uniform database. The study was approved by Institutional Review Board at the Johns Hopkins School of Medicine and at each participating site.
Data from 18 sites, located in the northeastern (8), midwestern (1), southern (5), and western (4) United States, were included in this analysis. The remaining 4 HIVRN sites discontinued participation or did not provide complete data during the study period. Adult patients (age ≥18 years) with at least 1 primary HIV outpatient visit and 1 CD4 test in any calendar year between January 1, 2006 and December 31, 2010 were eligible for inclusion.
We applied 3 previously described measures of retention for each patient in every calendar year included in the analysis.23 First, the US Health Resources and Services Administration HIV/AIDS Bureau (HRSA HAB) medical visits performance measure dichotomously defines retention as having 2 or more outpatient visits separated by ≥90 days during a calendar year.24 Second, the 6-month gap in care measure reflects whether a patient had ≥6 months between sequential outpatient visits, with no gap signifying retention in care. Patients were also coded as having a gap if there were no outpatient visits in the last 6 months of a calendar year. Third, 3-month visit constancy, an ordinal measure, is the number of 3-month intervals in a calendar year in which a patient completes at least 1 outpatient visit (range, 1–4). Outpatient visits refer only to primary HIV care appointments made to HIVRN clinics and do not include nursing, pharmacy, laboratory, or other types of visits.
HIV viral suppression was the binary outcome of interest, categorized as suppressed (HIV-1 RNA ≤400 copies/mL) and not suppressed (HIV-1 RNA >400 copies/mL). We used the last HIV-1 RNA value reported in each calendar year. Earlier studies focused on test results reported in a window 120 days before the end of a calendar year.14 Overall, 69% of values occurred in the 120-day window. Because many patients only had HIV-1 RNA tests before this window, we included an indicator variable in analyses reflecting whether the last test occurred earlier than September 2 in a calendar year.
Sociodemographic and Clinical Variables
For each year of observation, patients' age as of January 1 was divided into 4 groups: 18–29, 30–39, 40–49, and older than 50 years. Race/ethnicity was categorized as non-Hispanic white, non-Hispanic black, Hispanic, and other/missing. HIV transmission risk factor was grouped into men who had sex with men (MSM), heterosexual transmission (HET), injection drug use (IDU), and other/missing. Patients who had IDU in combination with another risk factor (eg, MSM, HET) were classified as IDU. Insurance coverage in each year was categorized as private, Medicaid, Medicare (including dual eligibles), uninsured, or other/missing. Patients whose care was funded by Ryan White, those recorded as self-pay, and those covered by local governmental programs were classified as uninsured. Patients were considered to be on ART if they concomitantly received 3 antiretroviral drugs during the calendar year. First CD4 count recorded in each calendar year was grouped as ≤200, 201–350, 351–500, and >500 cells per cubic millimeter.
The patient-year was the unit of analysis, reflecting the common practice of measuring retention in care and viral suppression on a calendar year basis.24 Analyses were limited to patient-years in which the patient was aged at least 18 years and in care, defined as having at least 1 primary HIV outpatient visit and 1 CD4 test in the year. We excluded 1368 patient-years in which individuals died, as they did not provide adequate time to measure retention and the outcome. Of the 123,991 patient-years between 2006 and 2011 that met inclusion criteria, 1968 (1.6%) had no HIV-1 RNA test reported; these observations were also removed from analyses.
Because the HRSA HAB medical visits and 6-month gap in care measures require at least 6 months of observation, we excluded data for 8263 patient-years for those who enrolled at HIVRN clinics between July and December of their first calendar year in care; data for subsequent years were included for these patients. Moreover, the 3-month visit constancy measure cannot be applied to patients without 4 calendar quarters of data; therefore, for analyses using this variable, we excluded the first calendar year in care for an additional 4562 patient-years (total = 12,825) in which persons entered care at HIVRN clinics after March. Thus, analyses using the HRSA HAB or the 6-month gap measures incorporated 113,760 patient-years from 35,433 persons; analyses using the 3-month visit constancy measure incorporated 109,198 patient-years for 33,439 persons. Eighteen percent of the larger sample contributed 1 year of data to the analyses, 62% between 2 and 6 years, and 20% all 7 years.
Standard descriptive analyses of demographic and clinical characteristics of the sample were conducted using the larger sample of 113,760 person-years. Because viral suppression was a binary outcome, multivariable logistic regression models were used to estimate the association between each measure of retention and viral suppression, adjusting for CD4 group and time of HIV-1 RNA measurement, age, sex, race/ethnicity, HIV transmission risk factor, insurance, and use of ART. Separate models were estimated for each retention measure. To evaluate whether the association between retention in care and viral suppression differed for patients with different disease severity, we included interaction terms between CD4 group and each measure of retention. All models adjusted for calendar year to account for any changes in clinic policies or treatment guidelines during the study period. In addition, we adjusted for differences in the odds of viral suppression across sites by including indicator variables for each site. All covariates were treated as categorical to allow a flexible specification for their association with the odds of viral suppression.25 Generalized estimating equations with a robust variance estimator and an exchangeable correlation structure were used to account for repeated measures of individuals over time.26 Statistical analyses were performed using Stata 12.1 (Stata Corporation, College Station, TX).
A total of 35,433 unique adult patients received care at 18 HIVRN sites between 2006 and 2011 (Table 1). Yearly sample size increased from 16,594 patients in 2006 to 22,237 patients in 2011. The proportion of patients who were aged 50 years or older increased from 26% in 2006 to 36% in 2011. Distributions of sex and race/ethnicity were stable over time, with the majority of patients being men and of minority race/ethnicity. MSM and HET were the predominant HIV risk behaviors. Use of ART increased from 76% to 88% during the study period. Median first CD4 count in the year rose from 399 to 476 cells per cubic millimeter. In each year, between 83% and 85% of patients met the HRSA HAB measure, 75%–78% did not have a 6-month gap, and 34%–39% had visits in all 4 quarters. The percentage achieving viral suppression increased from 60% in 2006 to 79% in 2011.
Figure 1 shows unadjusted proportions of patient-years in which viral suppression was achieved, by CD4 group and each retention measure. For each retention measure, 3 results are clear: (1) the higher the initial CD4 count, the greater the probability of viral suppression; (2) patients who were retained in care had a higher probability of viral suppression than those not retained in care; and (3) the difference in the effect of retention in care on viral suppression was greater at lower than higher CD4 counts.
Adjusting for CD4 group and sociodemographic factors, all 3 retention measures were significantly associated with viral suppression (Tables 2 and 3). The interaction between retention in care and CD4 group was significant (P < 0.01) for each retention measure: χ2 (3df) = 103.69 for HRSA HAB, χ2 (3df) = 185.39 for 6-month gap, and χ2 (9df) = 248.88 for 3-month visit constancy.
Table 4 shows adjusted odds ratios (AORs) for the effect of retention in care on viral suppression at different CD4 counts. For each retention measure, associations with viral suppression were stronger at more advanced disease stages. Using the HRSA HAB measure as a representative example, the association between retention in care and viral suppression was strongest for patients with lower CD4 counts: ≤200 [AOR = 2.33, 95% confidence interval (CI): 2.16 to 2.51], 201–350 (AOR = 1.96, CI: 1.81 to 2.12), 351–500 (AOR = 1.65, CI: 1.53 to 1.78), and >500 cells per cubic millimeter (AOR = 1.22, CI: 1.14 to 1.30). Pairwise comparisons of AOR in adjacent CD4 groups were significant (P < 0.05), with 2 exceptions (Table 4).
These results, from a large multisite sample, provide new information on the relationship between HIV disease severity, retention in care, and viral suppression. All 3 retention measures (HRSA HAB, 6-month gap in care, and 3-month visit constancy) were significantly associated with viral suppression. However, the association between retention in care and viral suppression differed by disease severity; it was strongest among patients with low CD4 counts. Although it is well established that retention in care is important for all HIV-infected patients, our data suggest that retention in care may be even more central to achieving optimal virological outcomes for persons with advanced HIV disease.
Consistent with earlier studies, patients retained in care were more likely to achieve viral suppression.14–17 Variations in adherence to ART and patients' ability to manage their HIV infection may explain these differences. Among 1972 patients receiving ART at 60 clinics in Brazil, missed appointments were independently associated with poor adherence to therapy.27 Similarly, Kleeberge et al28 document a significant association between poor retention in care and low self-reported ART adherence. An evaluation of factors influencing engagement in care notes that sporadic users had difficulty integrating new information, managing stigma, and maintaining normality in their lives compared with regular users of care.29 New patient-centered, retention in care interventions are needed to address the complex socioeconomic and clinical needs of HIV-infected individuals not engaged in care.
In our cohort, patients with higher CD4 counts were more likely to achieve viral suppression; however, we note that the difference in the effect of retention in care on viral suppression was greater in patients with lower compared with higher CD4 counts. This pattern persisted independent of the measure of retention used. People with low CD4 counts are at increased risk of polypharmacy, opportunistic infections, and other HIV-related complications.30,31 In addition, HIV-related stigma, fear of disclosing one's HIV status, and lack of psychological coping resources contribute to late entry into care with lower CD4 counts.32 Maintaining a continuous, high-quality, relationship with a provider may help patients with advanced HIV disease better manage these issues, and may explain why viral suppression is more strongly associated with retention in care in this population compared with individuals with higher CD4 counts.33
Several limitations of this study should be acknowledged. We did not have access to appointment schedules and thus could not examine other measures of retention such as appointment adherence and missed visits. Second, our data do not reflect visits to multiple providers by the same patient. It is possible that patients may switch to a different provider in the same locality, emigrate from the area, or become incarcerated or institutionalized but still be receiving care. Third, we did not measure adherence to or duration of HIV treatment. Future studies should investigate how ART adherence and duration of ART treatment influences the relationship between retention in care and viral suppression. Fourth, although multisite studies have greater generalizability than single-site studies, the HIVRN data are not nationally representative; rates of retention may differ among providers with smaller HIV patient caseloads or with a different mix of patients. Fifth, we removed data for the first year in care for those patients with insufficient time to observe the retention measures. Some patients with only 1 year in care were thus excluded altogether. Of 9483 persons with only 1 year of data, 3255 were removed from analyses using the HRSA HAB or 6-month gap measures and 5213 from analyses using the continuity measure. Those with only 1 year in care linked to but did not establish consistent care. Previous studies have described and examined retention in this population,9,16,26,34 including a study from the HIVRN, which demonstrated that 22% of 22,984 HIV-infected adults initiating care between 2001 and 2009 never established care (defined as having no outpatient visits 6 months after enrollment).2 Additional research is warranted to better understand patterns of care and their effect on clinical outcomes during the first year of care and the last year of life.
The timing of the HIV-1 RNA measure is another limitation. For 31% of the person-years, the measure occurred relatively early in the calendar year (ie, before September). Failure to be retained could have occurred after the HIV-1 RNA measurement. It is possible that virological failure could lead some patients to conclude that treatment was ineffective and motivate dropout. Thus, the analyses show an association between retention and virological suppression, but the parameter estimates cannot be interpreted as representing a causal effect. Ancillary analyses (results not shown) included a 3-way interaction between retention, CD4 count, and the indicator of HIV-1 RNA measure before September. This interaction was not significant for the HRSA HAB and the gap measures. It was significant for the quarters measure, but inspection of results showed that the major overall pattern of weaker association at higher CD4 counts remained both for pre-September and post-September measures of suppression.
This study is one of the first to report the impact of HIV disease severity on retention in care and viral suppression, demonstrating that retention in care is more strongly associated with viral suppression in patients with low CD4 counts. Our findings have important implications for improving the health of patients with advanced HIV disease and emphasize the role of retention in care in test and treat approaches to HIV prevention, demonstrating the added value of retaining people with lower CD4 counts in care.
The authors are grateful to all patients, physicians, investigators, and staff involved in the HIV Research Network.
Participating Sites: Alameda County Medical Center, Oakland, CA (Howard Edelstein, MD), Children's Hospital of Philadelphia, Philadelphia, PA (Richard Rutstein, MD), Community Health Network, Rochester, NY (Roberto Corales, DO), Drexel University, Philadelphia, PA (Jeffrey Jacobson, MD and Sara Allen, CRNP), Fenway Health, Boston, MA (Stephen Boswell, MD), Johns Hopkins University, Baltimore, MD (Kelly Gebo, MD, Richard Moore, MD, and Allison Agwu, MD), Montefiore Medical Group, Bronx, NY (Robert Beil, MD and Carolyn Chu, MD), Montefiore Medical Center, Bronx, NY (Lawrence Hanau, MD), Oregon Health and Science University, Portland, OR (P. Todd Korthuis, MD, MPH), Parkland Health and Hospital System, Dallas, TX (Muhammad Akbar, MD and Laura Armas-Kolostroubis, MD), St. Jude's Children's Hospital and University of Tennessee, Memphis, TN (Aditya Gaur, MD), St. Luke's Roosevelt Hospital Center, NY, New York (Victoria Sharp, MD and Stephen Arpadi, MD), Tampa General Health Care, Tampa, FL (Charurut Somboonwit, MD), University of California, San Diego, CA (W. Christopher Mathews, MD), and Wayne State University, Detroit, MI (Jonathan Cohn, MD).
Sponsoring Agencies: Agency for Healthcare Research and Quality, Rockville, MD (Fred Hellinger, PhD, John Fleishman, PhD, and Irene Fraser, PhD) and Health Resources and Services Administration, Rockville, MD (Robert Mills, PhD and Faye Malitz, MS).
Data Coordinating Center: Johns Hopkins University (Richard Moore, MD, Jeanne Keruly, CRNP, Kelly Gebo, MD, Cindy Voss, MA, and Nikki Balding, MS).
1. Vital signs: HIV prevention through care and treatment—United States. MMWR Morb Mortal Wkly Rep. 2011;60:1618–1623.
2. Fleishman JA, Yehia BR, Moore RD, et al.. Establishment, retention, and loss to follow-up in outpatient HIV care. J Acquir Immune Defic Syndr. 2012;60:249–259.
3. Mugavero MJ. Improving engagement in HIV care: what can we do? Top HIV Med. 2008;16:156–161.
4. Horstmann E, Brown J, Islam F, et al.. Retaining HIV-infected patients in care: Where are we? Where do we go from here? Clin Infect Dis. 2010;50:752–761.
5. Ulett KB, Willig JH, Lin HY, et al.. The therapeutic implications of timely linkage and early retention in HIV care. AIDS Patient Care STDS. 2009;23:41–49.
6. Berg MB, Safren SA, Mimiaga MJ, et al.. Nonadherence to medical appointments is associated with increased plasma HIV RNA and decreased CD4 cell counts in a community-based HIV primary care clinic. AIDS Care. 2005;17:902–907.
7. Giordano TP, White AC Jr, Sajja P, et al.. Factors associated with the use of highly active antiretroviral therapy in patients newly entering care in an urban clinic. J Acquir Immune Defic Syndr. 2003;32:399–405.
8. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions. Ann Intern Med. 1999;131:81–87.
9. Mugavero MJ, Lin HY, Willig JH, et al.. Missed visits and mortality among patients establishing initial outpatient HIV treatment. Clin Infect Dis. 2009;48:248–256.
10. Metsch LR, Pereyra M, Messinger S, et al.. HIV transmission risk behaviors among HIV-infected persons who are successfully linked to care. Clin Infect Dis. 2008;47:577–584.
11. Thompson MA, Mugavero MJ, Amico KR, et al.. Guidelines for improving entry into and retention in care and antiretroviral adherence for persons with HIV: evidence-based recommendations from an International Association of Physicians in AIDS Care panel. Ann Intern Med. 2012;156:817–833.
12. Yehia BR, Kangovi S, Frank I. Patients in transition: avoiding detours on the road to HIV treatment success. AIDS. Epub ahead of print Feb 21, 2013.
13. Yehia BR, Fleishman JA, Moore RD, et al.. Retention in care and health outcomes of transgender persons living with HIV. Clin Infect Dis. Epub ahead of print June 23, 2013.
14. Mugavero MJ, Westfall AO, Zinski A, et al.. Measuring retention in HIV care: the elusive gold standard. J Acquir Immune Defic Syndr. 2012;61:574–580.
15. Mugavero MJ, Amico KR, Westfall AO, et al.. Early retention in HIV care and viral load suppression: implications for a test and treat approach to HIV prevention. J Acquir Immune Defic Syndr. 2012;59:86–93.
16. Tripathi A, Youmans E, Gibson JJ, et al.. The impact of retention in early HIV medical care on viro-immunological parameters and survival: a statewide study. AIDS Res Hum Retroviruses. 2011;27:751–758.
17. Giordano TP, Gifford AL, White AC Jr, et al.. Retention in care: a challenge to survival with HIV infection. Clin Infect Dis. 2007;44:1493–1499.
18. Mohammadpour A, Yekta ZP, Nikbakht Nasrabadi AR. HIV-infected patients' adherence to highly active antiretroviral therapy: a phenomenological study. Nurs Health Sci. 2010;12:464–469.
19. Prosperi MC, Fabbiani M, Fanti I, et al.. Predictors of first-line antiretroviral therapy discontinuation due to drug-related adverse events in HIV-infected patients: a retrospective cohort study. BMC Infect Dis. 2012;12:296.
20. Remien RH, Hirky AE, Johnson MO, et al.. Adherence to medication treatment: a qualitative study of facilitators and barriers among a diverse sample of HIV+ men and women in four US cities. AIDS Behav. 2003;7:61–72.
21. Yehia BR, Gebo KA, Hicks PB, et al.. Structures of care in the clinics of the HIV Research Network. AIDS Patient Care STDS. 2008;22:1007–1013.
22. Yehia BR, Agwu AL, Schranz A, et al.. Conformity of pediatric/adolescent HIV clinics to the patient-centered medical home care model. AIDS Patient Care STDS. 2013;27:272–279.
23. Yehia BR, Fleishman JA, Metlay JP, et al.. Comparing different measures of retention in outpatient HIV care. AIDS. 2012;26:1131–1139.
25. Mancl LA, Leroux BG. Efficiency of regression estimates for clustered data. Biometrics. 1996;52:500–511.
26. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121–130.
27. Nemes MI, Carvalho HB, Souza MF. Antiretroviral therapy adherence in Brazil. AIDS. 2004;18(suppl 3):S15–S20.
28. Kleeberger CA, Phair JP, Strathdee SA, et al.. Determinants of heterogeneous adherence to HIV-antiretroviral therapies in the multicenter AIDS Cohort Study. J Acquir Immune Defic Syndr. 2001;26:82–92.
29. Mallinson RK, Relf MV, Dekker D, et al.. Maintaining normalcy: a grounded theory of engaging in HIV-oriented primary medical care. ANS Adv Nurs Sci. 2005;28:265–277.
30. Marzolini C, Elzi L, Gibbons S, et al.. Prevalence of comedications and effect of potential drug-drug interactions in the Swiss HIV Cohort Study. Antivir Ther. 2010;15:413–423.
32. Mukolo A, Villegas R, Aliyu M, et al.. Predictors of late presentation for HIV diagnosis: a literature review and suggested way forward. AIDS Behav. 2013;17:5–30.
33. Beach MC, Keruly J, Moore RD. Is the quality of the patient-provider relationship associated with better adherence and health outcomes for patients with HIV? J Gen Intern Med. 2006;21:661–665.
34. Giordano TP, Visnegarwala F, White AC Jr, et al.. Patients referred to an urban HIV clinic frequently fail to establish care: factors predicting failure. AIDS Care. 2005;17:773–783.
Keywords:© 2014 by Lippincott Williams & Wilkins
retention in care; engagement in care; retention measures; viral suppression; HIV disease severity; CD4 count