Introduction
Washington, DC, has been among some of the urban centres with the highest incidence and prevalence rates of HIV in the United States (US) [1]. In the latest annual surveillance report for HIV, the District of Columbia (DC) Department of Health (DC Health) reported a prevalence rate of 1.8% in 2019 [2], which continues to exceed the World Health Organization (WHO)'s threshold of 1% for a generalised epidemic [3]. In response to high rates of HIV in the nation's capital, DC Health and the National Institutes of Health (NIH) launched a multipronged initiative to improve efforts to curb the local epidemic in 2009 [4]. The DC Cohort study, a longitudinal cohort study enrolling people living with HIV (PWH) receiving care in Washington DC, was conceived as one of the pillars of this partnership. The overall objectives of the cohort are to characterise HIV outcomes in DC and to improve the quality of care for PWH [Reference Greenberg5, Reference Castel6].
The National HIV/AIDS Strategy (NHAS) uses the HIV care continuum as a functional tool and strategy to improve HIV treatment outcomes nationwide [7]. The HIV care continuum model describes the sequential and essential steps of HIV clinical care [8], namely that PWH that have been tested, linked to care, prescribed antiretroviral therapy (ART), and are virally suppressed [Reference Gardner9]. Viral suppression (VS) is the ultimate goal of the care continuum as it improves clinical outcomes for PWH [Reference Samji10, Reference Xia11] and reduces transmission rates as individuals who are suppressed are unable to transmit HIV [Reference Gardner9, Reference Dieffenbach and Fauci12–Reference Cohen14]. While DC Health surveillance data indicates that newly diagnosed HIV cases steadily decreased from 405 in 2015 to 217 in 2020, annual infection rates continue to exceed the national average [2]. Applying the HIV care continuum model to persons with newly diagnosed HIV in Washington DC can help improve our understanding of treatment dynamics in this population, identify gaps in the test-and-treat approach, gauge the impact of current evidence-based interventions and policies, and align with the NHAS's Ending the HIV Epidemic (EHE): A Plan for America [15]. The objective of this analysis was to examine sociodemographic and clinical factors associated with retention in care (RIC), ART initiation and VS among newly diagnosed PWH in the DC Cohort.
Methods
Study population and data sources
The DC Cohort study is a city-wide multisite longitudinal cohort study of PWH seeking care in 15 major community-based, hospital/academic and government clinical sites in Washington, DC. At the time of this analysis, there were 14 participating outpatient sites: 9 hospital-based clinics and 5 community-based clinics [Reference Greenberg5]. All participants provided written informed consent for inclusion in the cohort. The study is approved by the Institutional Review Board (IRB) at George Washington University and by participating sites with their own IRB. Sociodemographic and clinical data are abstracted from the clinics' electronic medical records. They are biannually linked to the DC Health HIV/AIDS surveillance system, which provides additional HIV laboratory results and supplementary clinical data from hospitals and clinics where participants seek care. HIV and AIDS diagnosis dates are also confirmed or added when missing [Reference Castel16].
The DC Cohort is a prospective cohort. We performed secondary data analysis of existing cohort data. For this analysis, we included participants enrolled in the DC Cohort between 1 January 2011 and 31 March 2016, unrestricted by age, who were diagnosed with HIV within 12 months of the enrolment date and had at least one year of follow-up data (hereafter referred to as people who were recently diagnosed with HIV). We examined an additional 12 months of HIV RNA laboratory results after one year of diagnosis to assess sustained VS.
Care continuum outcomes
We used a diagnosis-based care continuum model [8] as we included only diagnosed individuals enrolled in the DC Cohort. We used the Institute of Medicine's definition for RIC, which was defined as having at least two HIV-related medical encounters at least 90 days apart in the 12 months post-enrolment in the DC Cohort [17]. Since medical encounter data is only available after enrolment in the DC Cohort, as it is not routinely captured in local HIV surveillance data, RIC could only be estimated after consent to study participation. ART initiation was defined as being prescribed any antiretroviral medication during two time points: 3 and 12 months following the HIV diagnosis date. In the DC Cohort, data collection includes ART initiation date, which can be a date prior to DC Cohort enrolment. VS was defined as at least one HIV RNA laboratory result <200 copies/μl [18] within the time frame of interest. This measure was assessed at two time points: 3 and 12 months following HIV diagnosis date. Frequency of HIV RNA assessment is at the discretion of the individual providers. HIV RNA labs were eligible to be assessed at 3 months if they were taken between DC Cohort enrolment and the end of month 3. The window for the 12-month HIV RNA was DC Cohort enrolment until the end of month 12.
The source of data for this information could be either local HIV surveillance data or data sent to the DC Cohort by the clinic. For all participants who achieved VS within one year of follow-up, an additional 12 months of HIV RNA labs were evaluated from the first date of VS to determine if VS was sustained. Again, these values could come from either local HIV surveillance data or clinic data. Participants with all HIV RNA values suppressed during the additional 12 months of follow-up were considered to have achieved sustained VS. Based on data availability, RIC was measured in the 12 months post-enrolment in the DC Cohort. However, ART initiation and VS were measured in the 12 months post-HIV diagnosis.
Covariates
Additional measures of interest were collected at enrolment. These consisted of sociodemographic factors and HIV-related characteristics, including state of residence, HIV transmission risk, insurance (public vs. private), housing and employment status, site of care types (community vs. hospital-based clinic), smoking (ever vs. never), substance use (ever vs. never), alcohol abuse (ever vs. never), AIDS diagnosis, nadir CD4 at enrolment, HIV RNA laboratory results, HIV duration and year of diagnosis. AIDS diagnosis was defined as having a diagnosis of an AIDS-defining illness, a low CD4 count result of less than 200 cells/μL, or a record of an AIDS diagnosis date prior to enrolment [19].
Statistical analysis
The number of participants not retained and retained in care was calculated. The association between RIC and covariates were assessed among categorical variables using proportions (%) and median (IQR) continuous variables, testing for significance using Chi-square (Fisher's exact test for small sample sizes) for frequencies and Wilcoxon rank-sum tests medians, respectively.
Separate multivariate logistic regression (LR) models were used to explore the association between covariates and six care continuum outcomes: (1) RIC (2) ART initiation at 3 months (3) ART initiation at 12 months (4) VS achieved at 3 months (5) VS achieved at 12 months (6) VS sustained for 12 months among those who achieved VS during the first year of follow-up. Unadjusted odds ratios and adjusted odds ratios (aOR) with 95% confidence intervals (CI) are presented for each model. The sex at birth, age, race/ethnicity and site of care type were selected a priori and included in all models. In addition to a priori variables, predictors were added to the models based on the statistical significance (P < 0.05) of the association during bivariate analyses. Final models were chosen based on how the models fit the data using the Hosmer and Lemeshow goodness-of-fit test and assuming the principle of parsimony. All analyses were performed using SAS software, version 9.4 (copyright SAS Institute, Inc., Cary, NC, USA).
Results
Newly diagnosed participants comprised 5.5% (455/8302) of DC Cohort enrolees. As shown in Table 1, they were mostly Black (69%), male (79%), men who have sex with men (60%), and were a median age of 33 years (IQR 25, 45). Most resided in DC (69%), were on public insurance (53%), and had stable housing (78%), while a lower proportion was employed (35%). Less than half had ever smoked (40%); about a quarter had ever used illicit substances (24%) or ever abused alcohol (24%). At enrolment, the median duration of HIV diagnosis was 4.9 months (IQR 2.3, 7.7), the median nadir CD4 count was 346 cells/μL (IQR 224, 494), median log HIV RNA was 4.08 copies/mL (IQR 2.04, 4.76), and 38% had an AIDS diagnosis.
Demographic and clinical characteristicsa, by retention in care (RIC)b status, among individuals newly diagnosed with HIV, DC Cohort, 2011–2016
NH, non-Hispanic; IDU, male or female injection drug user; IQR, interquartile range; MSM, men who have sex with men. Other race groups include those of multiple race groups and unknown.
a Characteristics measured at the time of enrolment.
b Retention in care is defined as having at least two HIV care visits at least 90 days apart in the 12 months post-enrolment.
c Opportunistic infections and CD4 counts <200 cells/μl or CD4% <14.
At 12 months after enrolment, 92% of newly diagnosed PWH were RIC. By 3 months post-HIV diagnosis, 65% had initiated ART; by 12 months post-diagnosis, this proportion increased to 91%. The proportion of participants with VS was 17% and 82% by 3- and 12- months post HIV diagnosis, respectively, and of those with VSV at 12 months, 55% sustained VS for an additional 12 months (Fig. 1).
Proportion of newly diagnosed PWH achieving HIV care continuum outcomes, DC Cohort, 2011–2016, n = 455.
As shown in Tables 2 and 3, multivariable LR analysis revealed no differences by sex, age or race for RIC, ART initiation, VS and sustained VS. Compared to those without a history of AIDS-defining diagnoses, those with an AIDS diagnosis were three times more likely to be RIC (aOR 2.99; 95% CI 1.13–7.88) and more than two and half times more likely to start ART by 3 months (aOR 2.58; 95% CI 1.61–4.12); they were also almost five times more likely to attain VS by 12 months (aOR 4.87; 95% CI 1.69–14.03).
Summary of logistic regression Analysis for factors associated with retention in care and ART Initiation
aOR, adjusted odds ratio; CI, confidence interval.
a Characteristics measured at the time of enrolment.
b Retention in care is defined as having at least two HIV care visits at least 90 days apart in the 12 months post-enrolment; ART initiation was defined as having been prescribed any antiretroviral medication was assessed at 3- and 12-months following HIV diagnosis date.
c Opportunistic infections and CD4 counts <200 cells/μl3 or CD4% <14.
Summary of logistic regression analysis for factors associated with VS at 3 months, VS at 12 months and sustained viral suppression
aOR, adjusted odds ratio; CI, confidence interval.
a Characteristics measured at the time of enrolment.
b VS is defined as having at least one HIV RNA laboratory results <200 by 3 and 12 months after diagnosis; sustained VS is defined as no unsuppressed HIV RNA measure during one year of follow-up after achieving VS 12 months post-diagnosis .
c The number of participants in each model varies according to the HIV RNA laboratory results available at different times. At 3 months post-diagnosis, 283 participants had HIV RNA laboratory results; at 12 months all participants (N = 455) had HIV RNA results; 347 participants had HIV RNA results available to calculate sustained viral suppression after 12 months of diagnosis.
d Viral suppression sustained for 12 months among those who achieved VS during the first year of follow-up.
e Opportunistic infections and CD4 counts <200 cells/μl or CD4% <14.
Discussion
DC Cohort participants recently diagnosed with HIV met the 2020 WHO care continuum outcome indicator goals of 90% on ART and 90% virally suppressed [3] by 12 months after their HIV diagnosis. More than 90% of newly diagnosed PWH were retained in care 12 months after DC Cohort enrolment, and at least two-thirds of DC Cohort participants recently diagnosed with HIV had sustained VS. These findings demonstrate that although most PWH were retained in care, not all individuals were prescribed ART soon after diagnosis, and it took several months for most individuals to start treatment. While we did not have data on the reasons for delayed time to ART in our analysis, some possible explanations include delay in securing insurance coverage, needing time to process an HIV diagnosis, concerns about treatment side effects [Reference Lee20], and provider hesitancy to prescribe due to concerns about adherence [Reference Beer21]. Our data include periods when practices for prescribing ART regardless of CD4 count, had not yet been widely accepted.
Over a third of newly-diagnosed DC Cohort participants from 2011–2016 did not have an HIV RNA viral load result within 3 months of diagnosis, and among the 283 participants that did, only 28% were virally suppressed. This improved to 82% at 12 months. Data from 2011–2015 reported by the DC Health in the 2017 surveillance report [22] revealed that 81.4% of patients were linked to care within 3 months, and 59.7% were suppressed at 12 months, comparable to our 12-month results. VS status at three months from the 2011–2015 surveillance data was not reported.
An analysis comparing demographic and clinical characteristics of DC Cohort participants, compared with Washington DC residents with HIV identified with surveillance data, was previously undertaken [Reference Opoku23]. The DC Cohort includes individuals receiving HIV care in Washington, DC, who reside in DC and the surrounding metropolitan areas, which include parts of Maryland and Virginia. Approximately 71% of DC Cohort participants reside in DC. The Cohort participants represent a population that is retained in care, and there are some demographic and clinical differences from the general population. In the previous evaluation of surveillance data, there were 12 964 known people living with HIV in DC at the end of 2016. Of those individuals, 40.1% were DC Cohort participants. Compared with nonparticipants, participants were less likely to be male (68.0% vs. 74.9%, P < 0.001) and more likely to be Black (82.3% vs. 69.5%, P < 0.001) and have a heterosexual contact HIV transmission risk (30.3% vs. 25.9%, P < 0.001). DC Cohort participants were also more likely to have ever been diagnosed with stage 3 HIV disease (59.6% vs. 47.0%, P < 0.001), have a CD4 <200 cells/μl in 2017 (6.2% vs. 4.6%, P < 0.001) and to be virally suppressed in 2017 (61.4% vs. 50.5%, P < 0.001) [Reference Opoku23]. We do not have a recent comparison between recently diagnosed individuals within the DC Cohort vs. those living in DC identified through surveillance data. However, we believe the cohort to be representative of individuals receiving care within Washington DC, since we have enrolled a large proportion of individuals receiving care at the largest HIV clinics.
We found that more advanced HIV was associated with improvements in most of the outcomes evaluated. Those with an AIDS diagnosis at enrolment were approximately twice as likely to be retained in care, to start ART by the third month post-diagnosis and have at least a virally suppressed lab by 12 months after diagnosis. These findings are similar to other large cohorts where a large number of newly diagnosed participants presented with low CD4 counts at diagnosis and at the start of ART [24, Reference Mocroft25]. Our results suggest an urgency among providers to promptly treat those with more advanced HIV disease, and also may reflect that patients are more likely to seek care and accept ART when they are ill. However, of equal concern is that patients without AIDS-defining criteria were not started as quickly on ART. This could represent a missed opportunity to prevent the development of HIV reservoirs and reduce underlying inflammation—one of the key rationales for starting ART as early as possible and without consideration of baseline CD4 count or HIV RNA level.
Closer monitoring or more rapid treatment initiation may also explain why those with an AIDS diagnosis were more likely to attain VS compared with those patients with less advanced disease at diagnosis. Additionally, providers may be less likely to immediately treat individuals with high CD4 or low HIV RNA [Reference Lee26]. DC Health has scaled up HIV testing efforts to capture undiagnosed PWH; however, providers and public health officials continue to face challenges due to stigma, complacency about HIV, high migration patterns, poor access to health care, poverty and unemployment [2, Reference Greenberg27, 28].
Strengths/limitations
The linkage with the DC Health HIV/AIDS surveillance system is a major strength of this analysis. Merging data from the DC Cohort with those collected by mandatory reporting to the local surveillance system improves completeness and accuracy of HIV and AIDS diagnoses and provides additional laboratory results that have been collected at clinics and hospitals other than the enrolling DC Cohort site [Reference Castel16]. The DC Cohort study has a consent rate of over 80% of those eligible patients approached and enrols patients seeking HIV care in 14 major clinics in the District. Although there are limitations in the representativeness of the DC Cohort population to all people with HIV in Washington, DC [Reference Opoku23] it is a data source that is extremely useful for describing HIV outcomes within DC.
Some of the limitations of this study are due to the nature of the DC Cohort study and DC Health surveillance data collection protocols. All patients included in this study were participants in the DC Cohort and already established in care, precluding assessment of linkage to care (an element of the HIV Care Continuum). Although we were able to assess RIC and other outcomes of the care continuum, we could not determine the proportion of newly diagnosed individuals within all of DC that were linked to care, as not all clinics providing HIV care in DC are part of the DC Cohort. Lastly, the most recent linkage with DC Health before this analysis occurred in 2018. This analysis required at least one year of follow-up data; therefore, patients had to be enrolled by 2016. We are working to establish more rapid data linkage protocols to decrease the delay in using linked data.
Conclusions
Despite the limitations of this study, this analysis is useful to improve our understanding of persons with recently diagnosed HIV infection, implement the process indicators of the HIV care continuum and assess VS as an outcome indicator in this population. This analysis demonstrates the importance of early HIV testing and timely linkage to care. Prompt identification and RIC of individuals with HIV are crucial steps to ensuring better clinical outcomes and preventing new HIV infections. Additionally, same-day ART initiation is becoming increasingly utilised among newly diagnosed PWH regardless of the patient's HIV stage at diagnosis [Reference Doshi and Greenberg29].
Acknowledgements
Data in this manuscript were collected by the DC Cohort Study Group with investigators and research staff located at: Children's National Hospital Pediatric clinic (Natella Rakhmanina); the Senior Deputy Director of the DC Department of Health HAHSTA (Clover Barnes); Family and Medical Counseling Service (Angela Wood); Georgetown University (Princy Kumar); The George Washington University Biostatistics Center (Marinella Temprosa, Vinay Bhandaru, Tsedenia Bezabeh, Nisha Grover, Lisa Mele, Susan Reamer, Alla Sapozhnikova and Greg Strylewicz); The George Washington University Department of Epidemiology (Shannon Barth, Morgan Byrne, Amanda Castel, Alan Greenberg, Shannon Hammerlund, Paige Kulie, Anne Monroe, James Peterson and Bianca Stewart) and Department of Biostatistics and Bioinformatics (Yan Ma); The George Washington University Medical Faculty Associates (Jose Lucar); Howard University Adult Infectious Disease Clinic (Jhansi L. Gajjala) and Pediatric Clinic (Sohail Rana); Kaiser Permanente Mid-Atlantic States (Michael Horberg); La Clinica Del Pueblo (Ricardo Fernandez); MetroHealth (Duane Taylor); Washington Health Institute, formerly Providence Hospital (Jose Bordon); Unity Health Care (Gebeyehu Teferi); Veterans Affairs Medical Center (Debra Benator); Washington Hospital Center (Glenn Wortmann); and Whitman-Walker Institute (Stephen Abbott). The DC Cohort is funded by the National Institute of Allergy and Infectious Diseases, UM1 AI069503 and 1R24AI152598-01.
Author contributions
MJ, AG, ADC and AKM designed the study, MJ, MB, LPH, ML processed the data, performed the analysis, and designed the tables and figures. MJ, MB, LPH, ML, MH, AG, ADC and AKM aided in interpreting the results. MJ took the lead in writing the manuscript with assistance from AG and AKM. All authors provided critical feedback and helped shape the research, analysis and manuscript.
Financial support
National Institute of Allergy and Infectious Diseases, 1R24AI152598-01.
Conflicts of interest
The authors have no relevant financial or non-financial interests to disclose.
Data availability statement
De-identified data available upon written request.
Ethical standards
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the George Washington University Institutional Review Board (IRB 071029).
Consent to participate
Informed consent was obtained from all individual participants included in the study.
Consent for publication
N/A
Code availability (software application or custom code)
Analytic code available upon written request.
Open access
