[Skip to Content]
[Skip to Content Landing]
Figure 1.
Prior Diagnosis, ART, and Viral Suppression Among HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya (Open Cohort)
Prior Diagnosis, ART, and Viral Suppression Among HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya (Open Cohort)

ART indicates antiretroviral therapy; HIV, human immunodeficiency virus; SEARCH, Sustainable East Africa Research in Community Health Study. Stable residence was defined as living in the study community for 6 or more months in the past year. Viral suppression was defined as plasma HIV RNA measurement of less than 500 copies/mL.

aEstimates adjusted for incomplete HIV serostatus and HIV RNA measurement (eTables 1-2 in Supplement 3).

Figure 2.
Prior Diagnosis, ART, Viral Suppression, Migration Out of the Community, and Death Among HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya Who Were Diagnosed At or Before Baseline (Closed Cohort)a
Prior Diagnosis, ART, Viral Suppression, Migration Out of the Community, and Death Among HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya Who Were Diagnosed At or Before Baseline (Closed Cohort)a

ART indicates antiretroviral therapy; HIV, human immunodeficiency virus; SEARCH, Sustainable East Africa Research in Community Health Study. Stable residence was defined as living in the study community for 6 or more months in the past year. Viral suppression indicates residents who had less than 500 copies/mL of HIV RNA. Migration out of the study community indicates prior residents who were reported to no longer live in the community.

aAdjusted to account for incomplete plasma HIV RNA measurement (eTable 4 in Supplement 3).

Table 1.  
Characteristics of Baseline Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenyaa
Characteristics of Baseline Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenyaa
Table 2.  
Postbaseline HIV Viral Suppression in a Closed Cohort of HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya Who Were Diagnosed At or Before Baseline (n = 7108)a
Postbaseline HIV Viral Suppression in a Closed Cohort of HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya Who Were Diagnosed At or Before Baseline (n = 7108)a
Table 3.  
Association Between Baseline Demographic Variables and HIV RNA ≥500 copies/mL at Follow-up Year 2 in a Closed Cohort of HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya Who Were Diagnosed At or Before Baseline (n = 7108)a
Association Between Baseline Demographic Variables and HIV RNA ≥500 copies/mL at Follow-up Year 2 in a Closed Cohort of HIV-Positive Stable Residents of 16 SEARCH Intervention Communities in Rural Uganda and Kenya Who Were Diagnosed At or Before Baseline (n = 7108)a
Supplement 3.

eFigure 1. Study Population for Primary Analysis

eFigure 2. Study Population for Sensitivity Analysis

eTable 1. Prior Diagnosis, ART Use, and Viral Suppression (HIV RNA<500 copies/mL) at Baseline and at 1 and 2 Years of Follow-up in an Open Cohort of HIV-Positive Adult Stable Residents

eTable 2. Prior Diagnosis, ART Use, and Viral Suppression at Baseline and at 1 and 2 Years of Follow-up in an Open Cohort of HIV-Positive Adult Stable Residents

eTable 3. Sensitivity Analysis: Prior Diagnosis, ART Use, and Viral Suppression at Baseline and at 1 and 2 Years of Follow-up in an Open Cohort of HIV-Positive Adult Residents, Including Nonstable Baseline Residents and In-Migrants

eTable 4. Proportion of a Closed Cohort of HIV-Positive Adults Diagnosed at or Before Study Baseline who are Dead, Out-Migrated, ART Naive, ART Experienced but Not Virally Suppressed, and Virally Suppressed at Baseline and After 1 and 2 Years (N = 7108)

eTable 5. Probability of Ever Testing for HIV (In a Closed Cohort of Individuals Without a Prior HIV Diagnosis at Study Baseline, N = 72 746) and Ever Initiating ART (In a Closed Cohort of HIV-Positive Adults Diagnosed at or Before Study Baseline, N = 7108) 1 And 2 Years After Study Baseline

eTable 6. Association Between Baseline Demographic Variables and Failure to Test for HIV at Least Once by Follow-up Year 2 In a Closed Cohort of Adult Stable Residents Without a Record of Prior HIV Diagnosis at Baseline (N = 72 746)

eTable 7. Association Between Baseline Demographic Variables and Never Using ART by Follow-Up Year 2 In a Closed Cohort of HIV-Positive Adults Diagnosed at or Before Study Baseline (N = 7108)

1.
Grinsztejn  B, Hosseinipour  MC, Ribaudo  HJ,  et al; HPTN 052-ACTG Study Team.  Effects of early versus delayed initiation of antiretroviral treatment on clinical outcomes of HIV-1 infection: results from the phase 3 HPTN 052 randomised controlled trial.  Lancet Infect Dis. 2014;14(4):281-290.PubMedGoogle ScholarCrossref
2.
Cohen  MS, Chen  YQ, McCauley  M,  et al; HPTN 052 Study Team.  Prevention of HIV-1 infection with early antiretroviral therapy.  N Engl J Med. 2011;365(6):493-505.PubMedGoogle ScholarCrossref
3.
Lundgren  JD, Babiker  AG, Gordin  F,  et al; INSIGHT START Study Group.  Initiation of antiretroviral therapy in early asymptomatic HIV infection.  N Engl J Med. 2015;373(9):795-807.PubMedGoogle ScholarCrossref
4.
Granich  RM, Gilks  CF, Dye  C, De Cock  KM, Williams  BG.  Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model.  Lancet. 2009;373(9657):48-57.PubMedGoogle ScholarCrossref
5.
Walensky  RP, Borre  ED, Bekker  LG,  et al.  The anticipated clinical and economic effects of 90-90-90 in South Africa.  Ann Intern Med. 2016;165(5):325-333.PubMedGoogle ScholarCrossref
6.
Tanser  F, Bärnighausen  T, Grapsa  E, Zaidi  J, Newell  ML.  High coverage of ART associated with decline in risk of HIV acquisition in rural KwaZulu-Natal, South Africa.  Science. 2013;339(6122):966-971.PubMedGoogle ScholarCrossref
7.
Stover  J, Hallett  TB, Wu  Z,  et al; New Prevention Technology Study Group.  How can we get close to zero? the potential contribution of biomedical prevention and the investment framework towards an effective response to HIV.  PLoS One. 2014;9(11):e111956.PubMedGoogle ScholarCrossref
8.
UNAIDS.  90-90-90: An ambitious treatment target to help end the AIDS epidemic. http://www.unaids.org/sites/default/files/media_asset/90-90-90_en.pdf. Accessed May 8, 2017.
9.
WHO.  Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection: What’s New: November 2015. Geneva, Switzerland: World Health Organization; 2015.
10.
Gaolathe  T, Wirth  KE, Holme  MP,  et al; Botswana Combination Prevention Project study team.  Botswana’s progress toward achieving the 2020 UNAIDS 90-90-90 antiretroviral therapy and virological suppression goals: a population-based survey.  Lancet HIV. 2016;3(5):e221-e230.PubMedGoogle ScholarCrossref
11.
Levi  J, Raymond  A, Pozniak  A, Vernazza  PL, Kohler  P, Hill  A.  Can the UNAIDS 90–90–90 target be reached? a systematic analysis of national HIV treatment cascades.  BMJ Global Health. 2016; 1(2):e000010. doi:10.1136/bmjgh-2015-000010Google ScholarCrossref
12.
Granich  R, Gupta  S, Hall  I, Aberle-Grasse  J, Hader  S, Mermin  JEgger  M.  Status and methodology of publicly available national HIV care continua and 90-90-90 targets: A systematic review.  PLoS Med. 2017;14(4):e1002253.PubMedGoogle ScholarCrossref
13.
Cherutich  P, Kim  AA, Kellogg  TA,  et al.  Detectable HIV viral load in Kenya: data from a population-based survey.  PLoS One. 2016;11(5):e0154318.PubMedGoogle ScholarCrossref
14.
Adakun  SA, Siedner  MJ, Muzoora  C,  et al.  Higher baseline CD4 cell count predicts treatment interruptions and persistent viremia in patients initiating ARVs in rural Uganda.  J Acquir Immune Defic Syndr. 2013;62(3):317-321.PubMedGoogle ScholarCrossref
15.
Tenthani  L, Haas  AD, Tweya  H,  et al; Ministry of Health in Malawi and IeDEA Southern Africa.  Retention in care under universal antiretroviral therapy for HIV-infected pregnant and breastfeeding women (Option B+) in Malawi.  AIDS. 2014;28(4):589-598.PubMedGoogle ScholarCrossref
16.
Chamie  G, Clark  TD, Kabami  J,  et al.  A hybrid mobile approach for population-wide HIV testing in rural east Africa: an observational study.  Lancet HIV. 2016;3(3):e111-e119.PubMedGoogle ScholarCrossref
17.
Ayieko  J, Van Rie  A, Owaraganise  A,  et al. A novel strategy for accelerated linkage to care following community-wide HIV testing. Poster presented at: International AIDS Conference; July 18-22, 2016; Durban, South Africa.
18.
Kwarisiima  D, Jain  V, Owaraganise  A,  et al. Virologic efficacy, safety, and retention in adults starting ART at high CD4 with streamlined care in rural Africa. Oral presentation at: Conference on Retroviruses and Opportunistic Infections; February 22-25, 2016; Boston, MA.
19.
Jain  V, Liegler  T, Kabami  J,  et al; SEARCH Collaboration.  Assessment of population-based HIV RNA levels in a rural east African setting using a fingerprick-based blood collection method.  Clin Infect Dis. 2013;56(4):598-605.PubMedGoogle ScholarCrossref
20.
Balzer  L, Schwab  J, van der Laan  M, Petersen  M. Evaluation of progress towards the UNAIDS 90-90-90 HIV care cascade: a description of statistical methods used in an interim analysis of the intervention communities in the SEARCH Study. http://biostats.bepress.com/cgi/viewcontent.cgi?article=1365&context=ucbbiostat. Accessed May 8, 2017.
21.
van der Laan  MJ, Gruber  S.  Targeted minimum loss based estimation of causal effects of multiple time point interventions.  Int J Biostat. 2012;8(1):8.PubMedGoogle ScholarCrossref
22.
van der Laan  MJ, Polley  EC, Hubbard  AE.  Super learner.  Stat Appl Genet Mol Biol. 2007;6:e25.PubMedGoogle ScholarCrossref
23.
Polley  EC, van der Laan  M.  SuperLearner: super learner prediction: R package, version 2.0-21. http://cran.r-project.org/web/packages/SuperLearner/index.html. Accessed May 8, 2017.
24.
R Core Team.  R: a language and environment for statistical computing. http://www.R-project.org/. Accessed May 8, 2017.
25.
Schwab  J, Lendle  S, Petersen  M, Gruber  S, van der Laan  M.  ltmle: Longitudinal targeted maximum likelihood estimation. http://cran.r-project.org/web/packages/ltmle/. Accessed May 8, 2017.
26.
Jain  V, Petersen  ML, Liegler  T,  et al; SEARCH Collaboration.  Population levels and geographical distribution of HIV RNA in rural Ugandan and Kenyan communities, including serodiscordant couples: a cross-sectional analysis.  Lancet HIV. 2017;4(3):e122-e133.PubMedGoogle ScholarCrossref
27.
Losina  E, Bassett  IV, Giddy  J,  et al.  The “ART” of linkage: pre-treatment loss to care after HIV diagnosis at two PEPFAR sites in Durban, South Africa.  PLoS One. 2010;5(3):e9538.PubMedGoogle ScholarCrossref
28.
Geng  EH, Odeny  TA, Lyamuya  R,  et al; East Africa International Epidemiologic Databases to Evaluate AIDS Consortium.  Retention in care and patient-reported reasons for undocumented transfer or stopping care among HIV-infected patients on antiretroviral therapy in Eastern Africa: application of a sampling-based approach.  Clin Infect Dis. 2016;62(7):935-944.PubMedGoogle ScholarCrossref
29.
Henry J. Kaiser Family Foundation.  Financing the response to AIDS in low- and middle-income countries: international assistance from donor governments in 2014. https://kaiserfamilyfoundation.files.wordpress.com/2015/06/7347-11-financing-the-response-to-aids-in-low-and-middle-income-countries.pdf. Accessed May 8, 2017.
30.
Hodgson  I, Plummer  ML, Konopka  SN,  et al.  A systematic review of individual and contextual factors affecting ART initiation, adherence, and retention for HIV-infected pregnant and postpartum women.  PLoS One. 2014;9(11):e111421.PubMedGoogle ScholarCrossref
31.
Treves-Kagan  S, Steward  WT, Ntswane  L,  et al.  Why increasing availability of ART is not enough: a rapid, community-based study on how HIV-related stigma impacts engagement to care in rural South Africa.  BMC Public Health. 2016;16:87.PubMedGoogle ScholarCrossref
32.
Kranzer  K, Lawn  SD, Johnson  LF, Bekker  LG, Wood  R.  Community viral load and CD4 count distribution among people living with HIV in a South African township: implications for treatment as prevention.  J Acquir Immune Defic Syndr. 2013;63(4):498-505.PubMedGoogle ScholarCrossref
33.
Maman  D, Zeh  C, Mukui  I,  et al.  Cascade of HIV care and population viral suppression in a high-burden region of Kenya.  AIDS. 2015;29(12):1557-1565.PubMedGoogle ScholarCrossref
34.
Lippman  SA, Shade  SB, El Ayadi  AM,  et al.  Attrition and opportunities along the HIV care continuum: findings from a population-based sample, North West Province, South Africa.  J Acquir Immune Defic Syndr. 2016;73(1):91-99.PubMedGoogle ScholarCrossref
35.
Maman  D, Chilima  B, Masiku  C,  et al.  Closer to 90-90-90: the cascade of care after 10 years of ART scale-up in rural Malawi: a population study.  J Int AIDS Soc. 2016;19(1):20673.PubMedGoogle ScholarCrossref
36.
Justman  J, Hoos  D, Kalton  G,  et al. 114LB: Real progress in the HIV epidemic: PHIA findings from Zimbabwe, Malawi, and Zambia. Abstract presented at: Conference on Retroviruses and Opportunistic Infections; February 13-16, 2017; Seattle, WA.
37.
Grabowski  M, Nakigozi  G, Nalugoda  F,  et al. 34LB Combination HIV prevention and HIV incidence in Rakai, Uganda. Abstract presented at: Conference on Retroviruses and Opportunistic Infections; February 13-16, 2017; Seattle, WA.
38.
Larmarange  J, Iwuji  C, Orne-Gliemann  J,  et al. Measuring the impact of test and treat on the HIV cascade: the challenge of mobility. Abstract presented at: Conference on Retroviruses and Opportunistic Infections; February 22-25, 2016; Boston, MA.
39.
International AIDS Society.  Differentiated care for HIV: a decision framework for antiretroviral therapy delivery. http://www.differentiatedcare.org/Portals/0/adam/Content/yS6M-GKB5EWs_uTBHk1C1Q/File/Decision%20Framework.pdf. Accessed May 8, 2017.
40.
UNAIDS.  Fast track: ending the AIDS epidemic by 2030. http://www.unaids.org/sites/default/files/media_asset/JC2686_WAD2014report_en.pdf. Accessed May 8, 2017.
Original Investigation
June 6, 2017

Association of Implementation of a Universal Testing and Treatment Intervention With HIV Diagnosis, Receipt of Antiretroviral Therapy, and Viral Suppression in East Africa

Author Affiliations
  • 1School of Public Health, University of California, Berkeley
  • 2University of California, San Francisco
  • 3Harvard T. H. Chan School of Public Health, Boston, Massachusetts
  • 4Infectious Diseases Research Collaboration, Kampala, Uganda
  • 5Kenya Medical Research Institute, Nairobi, Kenya
  • 6Makerere University, Kampala, Uganda
JAMA. 2017;317(21):2196-2206. doi:10.1001/jama.2017.5705
Key Points

Question  Was implementation of a human immunodeficiency virus (HIV) test-and-treat intervention in rural East Africa associated with increases in diagnosis, treatment, and viral suppression among individuals with HIV infection?

Findings  In this study that included 77 774 residents of the intervention communities of an ongoing cluster randomized trial, the proportion of HIV-positive individuals with HIV viral suppression increased from 44.7% at baseline to 80.2% after 2 years, along with increases in HIV diagnosis and initiation of antiretroviral therapy.

Meaning  Implementation of a community-based testing and treatment intervention in East Africa was associated with increased proportion of HIV-positive individuals who achieved viral suppression, along with increased HIV diagnosis and initiation of antiretroviral therapy.

Abstract

Importance  Antiretroviral treatment (ART) is now recommended for all HIV-positive persons. UNAIDS has set global targets to diagnose 90% of HIV-positive individuals, treat 90% of diagnosed individuals with ART, and suppress viral replication among 90% of treated individuals, for a population-level target of 73% of all HIV-positive persons with HIV viral suppression.

Objective  To describe changes in the proportions of HIV-positive individuals with HIV viral suppression, HIV-positive individuals who had received a diagnosis, diagnosed individuals treated with ART, and treated individuals with HIV viral suppression, following implementation of a community-based testing and treatment program in rural East Africa.

Design, Setting, and Participants  Observational analysis based on interim data from 16 rural Kenyan (n = 6) and Ugandan (n = 10) intervention communities in the SEARCH Study, an ongoing cluster randomized trial. Community residents who were 15 years or older (N = 77 774) were followed up for 2 years (2013-2014 to 2015-2016). HIV serostatus and plasma HIV RNA level were measured annually at multidisease health campaigns followed by home-based testing for nonattendees. All HIV-positive individuals were offered ART using a streamlined delivery model designed to reduce structural barriers, improve patient-clinician relationships, and enhance patient knowledge and attitudes about HIV.

Main Outcomes and Measures  Primary outcome was viral suppression (plasma HIV RNA<500 copies/mL) among all HIV-positive individuals, assessed at baseline and after 1 and 2 years. Secondary outcomes included HIV diagnosis, ART among previously diagnosed individuals, and viral suppression among those who had initiated ART.

Results  Among 77 774 residents (male, 45.3%; age 15-24 years, 35.1%), baseline HIV prevalence was 10.3% (7108 of 69 283 residents). The proportion of HIV-positive individuals with HIV viral suppression at baseline was 44.7% (95% CI, 43.5%-45.9%; 3464 of 7745 residents) and after 2 years of intervention was 80.2% (95% CI, 79.1%-81.2%; 5666 of 7068 residents), an increase of 35.5 percentage points (95% CI, 34.4-36.6). After 2 years, 95.9% of HIV-positive individuals had been previously diagnosed (95% CI, 95.3%-96.5%; 6780 of 7068 residents); 93.4% of those previously diagnosed had received ART (95% CI, 92.8%-94.0%; 6334 of 6780 residents); and 89.5% of those treated had achieved HIV viral suppression (95% CI, 88.6%-90.3%; 5666 of 6334 residents).

Conclusions and Relevance  Among individuals with HIV in rural Kenya and Uganda, implementation of community-based testing and treatment was associated with an increased proportion of HIV-positive adults who achieved viral suppression, along with increased HIV diagnosis and initiation of antiretroviral therapy. In these communities, the UNAIDS population-level viral suppression target was exceeded within 2 years after program implementation.

Trial Registration  clinicaltrials.gov Identifier: NCT01864683

Introduction

Early antiretroviral therapy (ART) improves the health of individuals with human immunodeficiency virus (HIV) and reduces HIV transmission.1-3 Mathematical models and observational analyses suggest that an intensive global investment to expand ART coverage could alter the epidemic trajectory and improve longevity, health, and economic productivity.4-7 However, realizing this potential requires diagnosing HIV, initiating ART, and suppressing viral replication in most HIV-positive persons, a progression referred to as the HIV care cascade.

The global health community has responded with a mandate for action. In 2014, the Joint United Nations Programme on HIV/AIDS (UNAIDS) issued an ambitious worldwide target: by 2020, at least 90% of HIV-positive individuals will be diagnosed, at least 90% of those diagnosed will be receiving ART, and at least 90% of those receiving ART will have suppressed viral replication (known as the 90-90-90 target), for an overall target of 73% of all HIV-positive individuals with HIV viral suppression.8 In 2015, the World Health Organization (WHO) recommended that all HIV-positive individuals initiate ART, irrespective of CD4 lymphocyte count.9

Apart from Botswana,10 most nations in sub-Saharan Africa remain substantially below the UNAIDS target.8,11-13 Furthermore, many nonsuppressed HIV-positive individuals are asymptomatic with high CD4 cell counts.13 Such individuals may prove challenging to test, engage in care, and virally suppress.14,15 Scalable strategies to effectively diagnose, treat, and achieve viral suppression among HIV-positive individuals in sub-Saharan Africa are needed.

The Sustainable East Africa Research in Community Health Study (SEARCH) is an ongoing cluster randomized trial in rural Kenya and Uganda evaluating the effect of a cascadewide test-and-treat strategy vs country-specific standard of care on HIV, health, and economic outcomes. The trial intervention integrates community-based multidisease testing, universal eligibility for ART for all HIV-positive individuals, and streamlined patient-centered ART delivery. In this descriptive study, interim data from the trial intervention communities were used to evaluate population-level HIV viral suppression and the HIV care cascade at study baseline and after 1 and 2 years of delivering the test-and-treat intervention.

Methods
Study Population

The SEARCH Study is a cluster randomized trial that enrolled 32 pair-matched rural communities, each with approximately 5000 residents. Between June 2013 and June 2014, all residents in each community were enrolled during a household census.16 The study included data from residents (aged ≥15 years) of 10 Ugandan and 6 Kenyan intervention communities, each followed up for 2 years (until June 2015 to June 2016). The trial protocol is available in Supplement 1.

All participants provided verbal informed consent in their preferred language. The Makerere University School of Medicine Research and Ethics Committee (Uganda), the Ugandan National Council on Science and Technology (Uganda), the Kenya Medical Research Institute Ethical Review Committee (Kenya), and the University of California, San Francisco, Committee on Human Research (United States) approved the consent procedures and the study.

Intervention

Population-based HIV testing was conducted at baseline and annually thereafter using a hybrid model that combined multidisease community health campaigns (including diabetes and hypertension screening) with home-based testing for residents who did not attend the campaign.16 All HIV-positive individuals were eligible for ART with efavirenz plus tenofovir disoproxil fumarate, co-formulated either with emtricitabine or with lamivudine, and were offered facilitated linkage to care consisting of (1) immediate appointments at government clinics, (2) personal introductions to clinic staff, (3) a clinician phone number, (4) a 1-time transport voucher, and (5) tracking of individuals who did not link to care.17 Streamlined ART delivery included (1) a 3-month follow-up schedule for clinically stable patients, (2) flexible hours and a welcoming environment, (3) a clinician phone number, (4) text or telephone-based appointment reminders, and (5) HIV RNA measures, with structured discussion of results with patients to support visit and medication adherence.18

Measures

Demographic data were collected at the baseline census. Stable residence was defined as living in the community 6 or more months in the past year and nonstable residence was defined as living in the community less than 6 months in the past year. At baseline and after 1 and 2 years, HIV serostatus was measured at community health campaigns; adults who did not attend were subsequently tracked and offered home-based testing.16 After baseline, individuals newly identified as community residents were classified as migrants into the community. Prior residents who were reported to no longer live in the community were classified as migrants out of the community.

Participants with a positive antibody test result, detectable HIV RNA, or a Ministry of Health record of prior HIV care were considered HIV positive. Plasma HIV RNA was measured on all HIV-positive individuals annually during hybrid mobile testing (at the health campaign or subsequent tracking).19 Viral suppression at baseline and after 1 and 2 years was defined as an HIV RNA measurement of less than 500 copies/mL, measured during that year’s hybrid mobile testing. Prior diagnosis was defined as a prior positive HIV test or Ministry of Health record of HIV care; baseline self-reported diagnosis was included in sensitivity analysis. ART use was assessed with Ministry of Health records; HIV-positive individuals with suppressed HIV RNA were considered to have received ART.

Statistical Analyses

Power calculations and sample size were calculated for the primary randomized trial; this descriptive study was nested within the larger trial. The analysis was prespecified (Supplement 2).20 The primary outcome was population-level viral suppression (plasma HIV RNA<500 copies/mL) among all HIV-positive residents. Secondary outcomes were HIV diagnosis, ART use among those diagnosed with HIV, and HIV viral suppression among those treated with ART. Primary and secondary outcomes were evaluated in an open cohort of HIV-positive community residents at 3 time points corresponding to the dates of annual community-wide testing at baseline and after 1 and 2 years of follow-up. Individuals entered the cohort at the first time point they were HIV infected, 15 years or older, and community residents. Individuals left the cohort when they died or migrated out of the community.

In this open cohort, we estimated the following proportions: (1) number of HIV-positive individuals with HIV viral suppression among all HIV-positive individuals (population-level suppression), (2) the number of individuals diagnosed with HIV prior to time of annual testing among all HIV-positive individuals, (3) number of individuals who had initiated ART prior to the time of annual testing among individuals diagnosed with HIV prior to time of annual testing, and (4) the number of individuals with HIV viral suppression among individuals who had initiated ART prior to the time of annual testing. Primary analysis was restricted to baseline stable residents (ie, those living in the community 6 months or more during the past year). Secondary analyses (1) stratified on sex, age, and country; and (2) included nonstable baseline residents (ie, those living in the community less than 6 months during the past year) and migrants into the community identified during postbaseline annual testing.

Because diagnosis, ART coverage, and HIV viral suppression in the open cohort of HIV-positive individuals depend not only on effective testing, treatment, and suppression but also on HIV incidence, population migration, and mortality, we also conducted longitudinal closed-cohort analyses. First, we estimated the proportion of stable residents diagnosed with HIV at or before study baseline who, after 1 and 2 years, had (1) never initiated ART, (2) initiated ART and had a plasma HIV RNA measurement of 500 copies/mL or more, (3) initiated ART and had a plasma HIV RNA measurement of less than 500 copies/mL, (4) died, and/or (5) migrated out of the community. Second, follow-up was censored at death or migration out of the community and the risk of and demographic factors associated with (1) never testing for HIV by 2 years (among residents without an HIV diagnosis prior to baseline), (2) never initiating ART by 2 years (among residents diagnosed with HIV at or before baseline), and (3) having an HIV RNA measurement of 500 copies/mL or more at 2 years (among residents diagnosed with HIV at or before baseline) were evaluated. Third, HIV viral suppression at 1 and 2 years, censoring at death or migration out of the community, was evaluated in the following subgroups of HIV-positive residents diagnosed at or before study baseline: (1) those who were newly diagnosed at baseline, (2) those who were previously diagnosed but ART naive at baseline, and (3) those who were currently or previously treated with ART at baseline (overall, and within baseline HIV viral suppression strata).

In primary analyses, targeted maximum likelihood estimation21,22 was used to estimate (1) the number of HIV-positive individuals, adjusting for differences in baseline demographics and testing history between individuals with measured vs missing HIV serostatus (to account for possible overrepresentation or underrepresentation of HIV-positive individuals among those with HIV serostatus measured at annual testing), and (2) the number of HIV-positive individuals with HIV viral suppression, adjusting for differences in baseline demographics, testing, ART, and suppression history between individuals with measured vs missing HIV RNA (to account for possible overrepresentation or underrepresentation of individuals with HIV viral suppression among those with HIV RNA measured at annual testing). Unadjusted estimates were calculated as proportions among individuals seen at annual testing with known HIV serostatus and HIV RNA levels.

In secondary analyses, targeted maximum likelihood estimation was also used to adjust for censoring by death or migration out of the community and, in exploratory analysis, to evaluate demographic variables associated with failure to test, use ART, and suppress HIV replication. To minimize model misspecification bias, propensity score and outcome models were estimated using Super Learning,22 a machine-learning method that used 5-fold cross-validation to combine general additive models, stepwise regression, and logistic regression. Analyses adjusted for community and used household as the independent unit in cross-validation and when calculating influence curve and nonparametric bootstrap-based standard errors. A 2-sided P value of less than 0.05 was considered significant. Analyses were conducted in R (R Foundation), version 3.3.2, using packages ltmle_0.9-8-4 and SuperLearner_2.0-21.23-25

Results
Study Population and Testing Coverage

There were 77 774 stable residents 15 years or older in the 16 intervention communities at baseline, of whom 45.3% were male and 35.1% were aged 15 to 24 years (Table 1). Baseline HIV prevalence was 10.3% (Southwest Uganda, 6.5% [1462 of 22 410 residents]; East Uganda, 3.5% [785 of 22 681 residents]; Kenya, 20.1% [4861 of 24 192 residents]). During 2 years of follow-up, 907 individuals died, 13 257 migrated out of a study community, and 9020 turned age 15 years (eFigure 1 in Supplement 3). Secondary analyses included an additional 11 851 nonstable residents and 6437 adult migrants into a study community (eFigure 2 in Supplement 3).

Annual HIV serostatus and RNA testing levels were high.16 Of stable residents, 89.1% (69 283 of 77 774 residents) had known HIV status (tested at that year’s annual campaign or had previously tested positive) at baseline; 89.4% (64 999 of 72 744 residents) had known HIV status at year 1; and 86.8% (63 045 of 72 630 residents) had known HIV status at year 2. Of those known to be HIV-positive, 70.1% (4983 of 7108 residents) had HIV RNA measured during hybrid mobile testing at baseline, 85.9% (6016 of 7003 residents) had HIV RNA measured at year 1, and 83.6% (5786 of 6925 residents) had HIV RNA measured at year 2. Missing HIV RNA levels were more common at baseline due to assay failures at early campaigns.26

Diagnosis, ART Use, and Suppression in an Open Cohort of HIV-Positive Individuals

At baseline, 44.7% of HIV-positive residents had achieved HIV viral suppression (95% CI, 43.5%-45.9%; 3464 of 7745 residents). Population-level suppression was 75.2% (95% CI, 74.1%-76.3%; 5399 of 7182 residents) after 1 year of intervention and 80.2% (95% CI, 79.1%-81.2%; 5666 of 7068 residents) after 2 years of intervention (Figure 1A; eTables 1-2 in Supplement 3). Over 2 years, population-level HIV viral suppression increased 35.5 percentage points (95% CI, 34.4-36.6; P < .001). At baseline, 64.9% of HIV-positive residents were previously diagnosed (95% CI, 63.8%-66.0%; 5028 of 7745 residents; 70.4% incorporating self-reported diagnosis [95% CI, 69.4%-71.5%], 5454 of 7745 residents), 80.3% of previously diagnosed HIV-positive residents had initiated ART (95% CI, 79.2%-81.4%; 4038 of 5028 residents), and 85.8% of HIV-positive residents with prior ART initiation had achieved HIV viral suppression (95% CI, 84.6%-87.0%; 3464 of 4038 residents). After 2 years, 95.9% of HIV-positive residents were previously diagnosed (95% CI, 95.3%-96.5%; 6780 of 7068 residents), 93.4% of previously diagnosed HIV-positive residents had initiated ART (95% CI, 92.8%-94.0%; 6334 of 6780 residents), and 89.5% of HIV-positive residents with prior ART initiation had achieved HIV viral suppression (95% CI, 88.6%-90.3%; 5666 of 6334 residents). Population-level HIV viral suppression at 2 years was similar when migrants into the community and nonstable residents were included (79.0% [95% CI, 78.0%-80.0%]; 6157 of 7792 HIV-positive residents) and was higher in unadjusted analyses (85.6% [95% CI, 84.6%-86.5%]; 4951 of 5786 residents) (eTables 2-3 in Supplement 3).

Among subgroups of the open cohort at baseline, a smaller proportion of HIV-positive males (55.9%) compared with females (69.7%) were previously diagnosed (difference, 13.8 percentage points [95% CI, 12.1-15.5]), and males were less likely to have achieved HIV viral suppression (39.1% among males vs 47.6% among females; difference, 8.5 percentage points [95% CI, 6.8-10.2]) (Figure 1B-G; eTable 2 in Supplement 3). At baseline, youth (aged 15 to 24 years) were less likely than older residents (aged >24 years) to have been previously diagnosed, treated once diagnosed, and suppressed once treated, and a lower proportion of HIV-positive youth (26.2%) compared with older adults (47.2%) had achieved HIV viral suppression (difference, 21.0 percentage points [95% CI, 18.5-23.5]). Age and sex disparities, although smaller, remained after 2 years. After 2 years, 76.2% of males vs 82.2% of females (difference, 6.0 percentage points [95% CI, 4.3-7.7), and 64.5% of youth vs 81.5% of older adults (difference, 17.0 percentage points [95% CI, 13.6-20.4]) had achieved HIV viral suppression.

ART Use and HIV Viral Suppression in a Closed Cohort of Baseline HIV-Positive Residents

Among a closed cohort of 7108 stable residents diagnosed with HIV at or before study baseline (Figure 2; eTable 4 in Supplement 3), 29.3% (n = 2080) were newly diagnosed and 13.9% (n = 990) were previously diagnosed but ART naive at baseline. Adjusting for missing HIV RNA measurements, an estimated 8.1% (n = 575) had initiated ART but had an HIV RNA measurement of 500 copies/mL or more and 48.7% (n = 3463) had an HIV RNA measurement of less than 500 copies/mL at baseline. In contrast, after 2 years, the estimated proportion of residents with HIV viral suppression was 73.4% (n = 5218), 1.9% (N = 136) had died, and 11.9% (N = 849) had migrated out of the community. Although initial disparities in suppression between ages and sexes declined over time (Figures 2B-G), by 2 years, 28.5% of youth (246 of 864 residents) had migrated out of the community and only 54.7% of youth (473 of 864 residents) were resident in the community with HIV viral suppression (vs 76.0% for age >24 years; 4748 of 6244 residents).

In longitudinal analysis in which follow-up was censored at death or migration out of the community with adjustment for potentially informative censoring and missing HIV RNA measurements, 79.7% (95% CI, 78.7%-80.8%) of the baseline HIV-positive cohort (n = 7108) had achieved HIV viral suppression after 1 year, increasing to 83.8% (95% CI, 82.8%-84.9%) after 2 years (Table 2). In the subgroup of individuals newly diagnosed at baseline (n = 2080), 62.8% (95% CI, 60.4%-65.2%) had achieved HIV viral suppression after 1 year, increasing to 68.8% (95% CI, 66.4%-71.2%) after 2 years. Individuals previously diagnosed but ART naive at baseline (n = 990) were more likely to have achieved HIV viral suppression (78.1% [95% CI, 75.3%-80.8%] at 1 year and 86.5% [95% CI, 84.2%-88.8%] at 2 years), whereas ART-experienced individuals with a baseline HIV RNA measurement of 500 copies/mL or more (n = 426) were less likely to have achieved HIV viral suppression (49.5% [95% CI, 44.2%-54.7%] at 1 year and 62.2% [95% CI, 57.2%-67.2%] at 2 years). Almost all individuals with a suppressed HIV RNA measurement at baseline (n = 2549) maintained a suppressed HIV RNA level 1 and 2 years later (96.3% [95% CI, 95.6%-97.1%] at 1 year and 96.8% [95% CI, 96.0%-97.6%] at 2 years).

Demographic Variables Associated With Testing, ART, and Suppression

Associations between baseline demographics and failure to test for HIV, receive ART, and virally suppress were explored in closed cohort analyses, adjusting for censoring at death or migration out of the community and, for viral suppression outcomes, for missing HIV RNA measurements. Among 72 746 baseline stable residents not diagnosed with HIV before baseline, the probability of testing for HIV at least once over 2 years of follow-up was 96.8% (95% CI, 96.6%97.0%) (eTable 5 in Supplement 3). In multivariable analyses, residents who were male, never married, mobile (≥1 month in the past year away from community), aged 25-44 years (vs aged >44 years), and had less than primary education (vs primary) were at higher risk of never testing for HIV (eTable 6 in Supplement 3).

Among 7108 HIV-positive residents diagnosed at or before baseline, an estimated 93.8% (95% CI, 93.2%-94.4%) had ever used ART and 83.8% (95% CI, 82.8%-84.9%) had achieved HIV viral suppression at 2 years (Table 2; eTable 5 in Supplement 3). In multivariable analyses, residents who were male, younger, and had intermediate wealth were at higher risk of never receiving ART, whereas those who were male, younger, and not employed in the formal sector were at higher risk of nonsuppression (Table 3; eTable 7 in Supplement 3).

Discussion

At baseline, 44.7% of HIV-positive residents had suppressed HIV viral replication. After 2 years of a community-based HIV test-and-treat intervention, 95.9% of HIV-positive individuals were previously diagnosed (prior to baseline or during the 2-year program), 93.4% of those had received ART, and 89.5% of those had achieved HIV viral suppression; overall, 80.2% of HIV-positive individuals had suppressed HIV viral replication.

The WHO global public health mandate to identify and offer ART to the 37 million persons living with HIV is a challenging goal—skepticism regarding the feasibility of reaching the UNAIDS 90-90-90 target by 2020 is legitimate. Globally, approximately half of persons living with HIV know their diagnosis,8 many diagnosed persons do not link to care or drop out after starting ART,27,28 HIV funding has not increased over the past 5 years,29 and stigma remains a barrier, particularly in vulnerable populations.30,31 The 2 most important findings of these analyses are (1) the 2015 WHO guidelines to treat persons regardless of CD4 cell count can be successfully implemented, and (2) the UNAIDS 90-90-90 target of 73% of HIV-positive persons with HIV viral suppression can be achieved and exceeded within a 2-year period in rural East Africa.

Few studies in sub-Saharan Africa have directly evaluated population-level HIV testing, treatment, and suppression of HIV viral replication to less than 500 copies/mL,10,11,13,32-37 and even fewer have data after starting an HIV test-and-treat intervention.38 Botswana—an upper-middle income country with a population of 2 million—is closest to achieving the UNAIDS 90-90-90 target.10 Despite Botswana’s policy to initiate ART at CD4 cell counts of less than 350 CD4 cells/mm3, an estimated 70% of HIV-positive persons resident in Botswana have achieved HIV viral suppression, an achievement attributed to Botswana’s political leadership, early adoption of HIV opt-out testing and ART, decentralization of HIV care, and access to HIV RNA monitoring.

This study demonstrates progress toward achieving the UNAIDS targets in rural Uganda and Kenya in regions with baseline HIV prevalence ranging from 3.5% to 20.1%. Furthermore, among populations in which less than half of HIV-positive adults had achieved HIV viral suppression at study baseline, an intervention that included treatment of adults with more than 350 CD4 cells/mm3 using a streamlined HIV care delivery model integrated into government health clinics exceeded the UNAIDS overall population-level viral suppression target after 1 year.

HIV testing is one of the most challenging obstacles to achieving universal ART coverage. Before the testing intervention, 64.9% of HIV-positive community residents knew their status, less than the 83% with known status in Botswana,10 but substantially more than the regional UNAIDS estimate of 56%.8 Using an “out of facility” mobile testing strategy that combined multidisease testing at community health campaigns with home testing for nonattendees, the proportion of HIV-positive persons aware of their status increased to 95.9% after 2 years. Community engagement contributed to this success. Participation of men, whose knowledge of HIV status lagged that of women at study baseline, was enhanced by targeted outreach that included competitive sporting events, conveniently located testing services with evening hours, and non-HIV services including diabetes and hypertension screening and sexual health consultation.16

Poor retention in HIV care and adherence to ART, particularly among persons starting therapy with a high CD4 cell count, is a potential barrier to achieving the new WHO treatment guidelines.9 A streamlined delivery model was used to enhance retention among patients initiating ART at all CD4 cell counts. This patient-centered model of care was designed to reduce structural barriers, improve patient-clinician relationships, and enhance patient knowledge and attitudes about HIV. It is an example of what is now called “differentiated” HIV care—a client-centered approach that adapts HIV services to reflect the preferences and expectations of people living with HIV.39 There is increasing global consensus that successfully treating all persons living with HIV, as well as generating more efficient health systems, will require differentiated care approaches.

This study provides insight into 4 groups—youth, males, those newly diagnosed, and those without HIV viral suppression despite ART initiation—who are falling short of the UNAIDS 90-90-90 target. At baseline, only 50.3% of HIV-positive youth (15-24 years) were aware of their status. This increased to 86.5% after 2 years, suggesting that the testing strategy was reaching most youth. However, population-level HIV viral suppression at 2 years remained 64.5% in youth compared with 81.5% in older populations, due in part to the lower proportion of viral suppression (82.4%) among youth who had started therapy. Barriers faced by youth included stigma and logistics of HIV care for those in boarding school. More research is needed to design age-specific solutions.

Among HIV-positive males at baseline, HIV testing and linkage for ART start were major barriers in the HIV care cascade. The testing intervention closed much of the sex gap in knowledge of HIV status over 2 years. However, after 2 years, HIV-positive males remained less likely to be diagnosed, treated, and virally suppressed on treatment, resulting in a persistent disparity in population-level suppression (82.2% among females vs 76.2% among males).

HIV-positive individuals who were newly diagnosed at baseline failed to achieve targets for ART initiation and viral suppression. In a closed-cohort analysis, only 68.8% had achieved HIV viral suppression after 2 years. This group will require new approaches to enhance linkage and adherence. Adults with an HIV RNA measurement of 500 copies/mL or more at baseline despite prior ART initiation also did not meet the target—only 62.2% had achieved HIV viral suppression after 2 years. Nonadherence, HIV drug resistance, and unmet structural barriers to care all likely contribute to poor suppression in this often vulnerable and difficult-to-reach population. Additional approaches, including improved implementation of second and third line ART treatment and possibly HIV resistance testing, should be evaluated.

This analysis did not address if HIV incidence decreased with the improvements observed in the care cascade. UNAIDS has already set a new target of 95-95-95 by 2030, recognizing that the current target alone is unlikely to eliminate the HIV epidemic.40 Reaching the HIV-positive individuals who remain without HIV viral suppression is a priority, both for their own health and because of their potential contribution to ongoing HIV transmission.

Limitations

This analysis has several limitations. First, prior to trial completion, progress toward the UNAIDS target cannot be attributed to the intervention. Second, primary analysis focused on baseline stable community residents due to their full exposure to the intervention; however, these individuals may be easier to test, treat, and suppress than more mobile populations. Although sensitivity analyses including nonstable residents and migrants into the community yielded similar estimates, ascertainment of migrants into the community was not comprehensive. Cascade outcomes among individuals who left the community are also unknown, as is the contribution of migrants to ongoing HIV transmission.38

Conclusions

Among individuals with human immunodeficiency virus (HIV) in rural Kenya and Uganda, implementation of community-based testing and treatment was associated with an increased proportion of HIV-positive adults who achieved viral suppression, along with increased HIV diagnosis and initiation of antiretroviral therapy. In these communities, the UNAIDS population-level viral suppression target was exceeded within 2 years after program implementation.

Back to top
Article Information

Corresponding Author: Maya Petersen, MD, PhD, School of Public Health, University of California-Berkeley, 101 Haviland Hall, Berkeley, CA 94110 (mayaliv@berkeley.edu).

Accepted for Publication: May 3, 2017.

Author Contributions: Drs Petersen and Havlir had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Petersen, Balzer, Kwarsiima, Sang, Chamie, Kabami, Liegler, Kadede, Jain, van der Laan, Cohen, Charlebois, Kamya, Havlir.

Acquisition, analysis, or interpretation of data: Petersen, Balzer, Kwarsiima, Ayieko, Owaraganise, Liegler, Mwangwa, Jain, Plenty, Brown, Lavoy, Schwab, Black, Bukusi, Clark, Charlebois, Havlir.

Drafting of the manuscript: Petersen, Balzer, Kwarsiima, Liegler, Jain, Lavoy, Schwab, Havlir.

Critical revision of the manuscript for important intellectual content: Petersen, Balzer, Kwarsiima, Sang, Chamie, Ayieko, Kabami, Owaraganise, Mwangwa, Kadede, Jain, Plenty, Brown, Black, van der Laan, Bukusi, Cohen, Clark, Charlebois, Kamya, Havlir.

Statistical analysis: Petersen, Balzer, Chamie, Kabami, Jain, Plenty, Schwab, van der Laan.

Obtained funding: Petersen, Clark, Charlebois, Havlir.

Administrative, technical, or material support: Kwarsiima, Sang, Chamie, Ayieko, Owaraganise, Mwangwa, Kadede, Lavoy, Black, Bukusi, Cohen, Clark, Charlebois, Kamya.

Supervision: Petersen, Kwarsiima, Lavoy, Bukusi, Cohen, Clark, Kamya.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: This work was supported by grants UM1AI069502, U01AI099959 (both from Dr Havlir), and R01AI074345 (Dr van der Laan) from the National Institute of Allergy and Infectious Diseases at the National Institutes of Health and by the US President's Emergency Plan for AIDS Relief, the Office of the Global AIDS Coordinator, Gilead Sciences (Truvada donation), and the Office of AIDS Research. The University of California, San Francisco-Gladstone Institute of Virology and Immunology Center for Aids Research provided indirect funding through grant P30 AI027763 for laboratory infrastructure and Dr Liegler's support.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We thank the residents of the SEARCH Study communities for their participation and the Uganda and Kenya Ministries of Health. They did not receive compensation for their contribution. We also thank the SEARCH Study Team.

References
1.
Grinsztejn  B, Hosseinipour  MC, Ribaudo  HJ,  et al; HPTN 052-ACTG Study Team.  Effects of early versus delayed initiation of antiretroviral treatment on clinical outcomes of HIV-1 infection: results from the phase 3 HPTN 052 randomised controlled trial.  Lancet Infect Dis. 2014;14(4):281-290.PubMedGoogle ScholarCrossref
2.
Cohen  MS, Chen  YQ, McCauley  M,  et al; HPTN 052 Study Team.  Prevention of HIV-1 infection with early antiretroviral therapy.  N Engl J Med. 2011;365(6):493-505.PubMedGoogle ScholarCrossref
3.
Lundgren  JD, Babiker  AG, Gordin  F,  et al; INSIGHT START Study Group.  Initiation of antiretroviral therapy in early asymptomatic HIV infection.  N Engl J Med. 2015;373(9):795-807.PubMedGoogle ScholarCrossref
4.
Granich  RM, Gilks  CF, Dye  C, De Cock  KM, Williams  BG.  Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model.  Lancet. 2009;373(9657):48-57.PubMedGoogle ScholarCrossref
5.
Walensky  RP, Borre  ED, Bekker  LG,  et al.  The anticipated clinical and economic effects of 90-90-90 in South Africa.  Ann Intern Med. 2016;165(5):325-333.PubMedGoogle ScholarCrossref
6.
Tanser  F, Bärnighausen  T, Grapsa  E, Zaidi  J, Newell  ML.  High coverage of ART associated with decline in risk of HIV acquisition in rural KwaZulu-Natal, South Africa.  Science. 2013;339(6122):966-971.PubMedGoogle ScholarCrossref
7.
Stover  J, Hallett  TB, Wu  Z,  et al; New Prevention Technology Study Group.  How can we get close to zero? the potential contribution of biomedical prevention and the investment framework towards an effective response to HIV.  PLoS One. 2014;9(11):e111956.PubMedGoogle ScholarCrossref
8.
UNAIDS.  90-90-90: An ambitious treatment target to help end the AIDS epidemic. http://www.unaids.org/sites/default/files/media_asset/90-90-90_en.pdf. Accessed May 8, 2017.
9.
WHO.  Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection: What’s New: November 2015. Geneva, Switzerland: World Health Organization; 2015.
10.
Gaolathe  T, Wirth  KE, Holme  MP,  et al; Botswana Combination Prevention Project study team.  Botswana’s progress toward achieving the 2020 UNAIDS 90-90-90 antiretroviral therapy and virological suppression goals: a population-based survey.  Lancet HIV. 2016;3(5):e221-e230.PubMedGoogle ScholarCrossref
11.
Levi  J, Raymond  A, Pozniak  A, Vernazza  PL, Kohler  P, Hill  A.  Can the UNAIDS 90–90–90 target be reached? a systematic analysis of national HIV treatment cascades.  BMJ Global Health. 2016; 1(2):e000010. doi:10.1136/bmjgh-2015-000010Google ScholarCrossref
12.
Granich  R, Gupta  S, Hall  I, Aberle-Grasse  J, Hader  S, Mermin  JEgger  M.  Status and methodology of publicly available national HIV care continua and 90-90-90 targets: A systematic review.  PLoS Med. 2017;14(4):e1002253.PubMedGoogle ScholarCrossref
13.
Cherutich  P, Kim  AA, Kellogg  TA,  et al.  Detectable HIV viral load in Kenya: data from a population-based survey.  PLoS One. 2016;11(5):e0154318.PubMedGoogle ScholarCrossref
14.
Adakun  SA, Siedner  MJ, Muzoora  C,  et al.  Higher baseline CD4 cell count predicts treatment interruptions and persistent viremia in patients initiating ARVs in rural Uganda.  J Acquir Immune Defic Syndr. 2013;62(3):317-321.PubMedGoogle ScholarCrossref
15.
Tenthani  L, Haas  AD, Tweya  H,  et al; Ministry of Health in Malawi and IeDEA Southern Africa.  Retention in care under universal antiretroviral therapy for HIV-infected pregnant and breastfeeding women (Option B+) in Malawi.  AIDS. 2014;28(4):589-598.PubMedGoogle ScholarCrossref
16.
Chamie  G, Clark  TD, Kabami  J,  et al.  A hybrid mobile approach for population-wide HIV testing in rural east Africa: an observational study.  Lancet HIV. 2016;3(3):e111-e119.PubMedGoogle ScholarCrossref
17.
Ayieko  J, Van Rie  A, Owaraganise  A,  et al. A novel strategy for accelerated linkage to care following community-wide HIV testing. Poster presented at: International AIDS Conference; July 18-22, 2016; Durban, South Africa.
18.
Kwarisiima  D, Jain  V, Owaraganise  A,  et al. Virologic efficacy, safety, and retention in adults starting ART at high CD4 with streamlined care in rural Africa. Oral presentation at: Conference on Retroviruses and Opportunistic Infections; February 22-25, 2016; Boston, MA.
19.
Jain  V, Liegler  T, Kabami  J,  et al; SEARCH Collaboration.  Assessment of population-based HIV RNA levels in a rural east African setting using a fingerprick-based blood collection method.  Clin Infect Dis. 2013;56(4):598-605.PubMedGoogle ScholarCrossref
20.
Balzer  L, Schwab  J, van der Laan  M, Petersen  M. Evaluation of progress towards the UNAIDS 90-90-90 HIV care cascade: a description of statistical methods used in an interim analysis of the intervention communities in the SEARCH Study. http://biostats.bepress.com/cgi/viewcontent.cgi?article=1365&context=ucbbiostat. Accessed May 8, 2017.
21.
van der Laan  MJ, Gruber  S.  Targeted minimum loss based estimation of causal effects of multiple time point interventions.  Int J Biostat. 2012;8(1):8.PubMedGoogle ScholarCrossref
22.
van der Laan  MJ, Polley  EC, Hubbard  AE.  Super learner.  Stat Appl Genet Mol Biol. 2007;6:e25.PubMedGoogle ScholarCrossref
23.
Polley  EC, van der Laan  M.  SuperLearner: super learner prediction: R package, version 2.0-21. http://cran.r-project.org/web/packages/SuperLearner/index.html. Accessed May 8, 2017.
24.
R Core Team.  R: a language and environment for statistical computing. http://www.R-project.org/. Accessed May 8, 2017.
25.
Schwab  J, Lendle  S, Petersen  M, Gruber  S, van der Laan  M.  ltmle: Longitudinal targeted maximum likelihood estimation. http://cran.r-project.org/web/packages/ltmle/. Accessed May 8, 2017.
26.
Jain  V, Petersen  ML, Liegler  T,  et al; SEARCH Collaboration.  Population levels and geographical distribution of HIV RNA in rural Ugandan and Kenyan communities, including serodiscordant couples: a cross-sectional analysis.  Lancet HIV. 2017;4(3):e122-e133.PubMedGoogle ScholarCrossref
27.
Losina  E, Bassett  IV, Giddy  J,  et al.  The “ART” of linkage: pre-treatment loss to care after HIV diagnosis at two PEPFAR sites in Durban, South Africa.  PLoS One. 2010;5(3):e9538.PubMedGoogle ScholarCrossref
28.
Geng  EH, Odeny  TA, Lyamuya  R,  et al; East Africa International Epidemiologic Databases to Evaluate AIDS Consortium.  Retention in care and patient-reported reasons for undocumented transfer or stopping care among HIV-infected patients on antiretroviral therapy in Eastern Africa: application of a sampling-based approach.  Clin Infect Dis. 2016;62(7):935-944.PubMedGoogle ScholarCrossref
29.
Henry J. Kaiser Family Foundation.  Financing the response to AIDS in low- and middle-income countries: international assistance from donor governments in 2014. https://kaiserfamilyfoundation.files.wordpress.com/2015/06/7347-11-financing-the-response-to-aids-in-low-and-middle-income-countries.pdf. Accessed May 8, 2017.
30.
Hodgson  I, Plummer  ML, Konopka  SN,  et al.  A systematic review of individual and contextual factors affecting ART initiation, adherence, and retention for HIV-infected pregnant and postpartum women.  PLoS One. 2014;9(11):e111421.PubMedGoogle ScholarCrossref
31.
Treves-Kagan  S, Steward  WT, Ntswane  L,  et al.  Why increasing availability of ART is not enough: a rapid, community-based study on how HIV-related stigma impacts engagement to care in rural South Africa.  BMC Public Health. 2016;16:87.PubMedGoogle ScholarCrossref
32.
Kranzer  K, Lawn  SD, Johnson  LF, Bekker  LG, Wood  R.  Community viral load and CD4 count distribution among people living with HIV in a South African township: implications for treatment as prevention.  J Acquir Immune Defic Syndr. 2013;63(4):498-505.PubMedGoogle ScholarCrossref
33.
Maman  D, Zeh  C, Mukui  I,  et al.  Cascade of HIV care and population viral suppression in a high-burden region of Kenya.  AIDS. 2015;29(12):1557-1565.PubMedGoogle ScholarCrossref
34.
Lippman  SA, Shade  SB, El Ayadi  AM,  et al.  Attrition and opportunities along the HIV care continuum: findings from a population-based sample, North West Province, South Africa.  J Acquir Immune Defic Syndr. 2016;73(1):91-99.PubMedGoogle ScholarCrossref
35.
Maman  D, Chilima  B, Masiku  C,  et al.  Closer to 90-90-90: the cascade of care after 10 years of ART scale-up in rural Malawi: a population study.  J Int AIDS Soc. 2016;19(1):20673.PubMedGoogle ScholarCrossref
36.
Justman  J, Hoos  D, Kalton  G,  et al. 114LB: Real progress in the HIV epidemic: PHIA findings from Zimbabwe, Malawi, and Zambia. Abstract presented at: Conference on Retroviruses and Opportunistic Infections; February 13-16, 2017; Seattle, WA.
37.
Grabowski  M, Nakigozi  G, Nalugoda  F,  et al. 34LB Combination HIV prevention and HIV incidence in Rakai, Uganda. Abstract presented at: Conference on Retroviruses and Opportunistic Infections; February 13-16, 2017; Seattle, WA.
38.
Larmarange  J, Iwuji  C, Orne-Gliemann  J,  et al. Measuring the impact of test and treat on the HIV cascade: the challenge of mobility. Abstract presented at: Conference on Retroviruses and Opportunistic Infections; February 22-25, 2016; Boston, MA.
39.
International AIDS Society.  Differentiated care for HIV: a decision framework for antiretroviral therapy delivery. http://www.differentiatedcare.org/Portals/0/adam/Content/yS6M-GKB5EWs_uTBHk1C1Q/File/Decision%20Framework.pdf. Accessed May 8, 2017.
40.
UNAIDS.  Fast track: ending the AIDS epidemic by 2030. http://www.unaids.org/sites/default/files/media_asset/JC2686_WAD2014report_en.pdf. Accessed May 8, 2017.
×