UBC Faculty Research and Publications

Care cascade structural intervention versus standard of care in the diagnosis and treatment of HIV in… Mao, Yurong; Wu, Zunyou; McGoogan, Jennifer M; Liu, David; Gu, Diane; Erinoff, Lynda; Ling, Walter; VanVeldhuisen, Paul; Detels, Roger; Hasson, Albert L; Lindblad, Robert; Montaner, Julio S G; Tang, Zhenzhu; Zhao, Yan Jun 12, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12913_2017_Article_2323.pdf [ 1.2MB ]
JSON: 52383-1.0348248.json
JSON-LD: 52383-1.0348248-ld.json
RDF/XML (Pretty): 52383-1.0348248-rdf.xml
RDF/JSON: 52383-1.0348248-rdf.json
Turtle: 52383-1.0348248-turtle.txt
N-Triples: 52383-1.0348248-rdf-ntriples.txt
Original Record: 52383-1.0348248-source.json
Full Text

Full Text

STUDY PROTOCOL Open AccessCare cascade structural intervention versusstandard of care in the diagnosis andtreatment of HIV in China: a cluster-randomized controlled trial protocolYurong Mao1, Zunyou Wu1*, Jennifer M. McGoogan1, David Liu2, Diane Gu1, Lynda Erinoff2, Walter Ling3,Paul VanVeldhuisen4, Roger Detels5, Albert L. Hasson3, Robert Lindblad4, Julio S. G. Montaner6, Zhenzhu Tang7and Yan Zhao1AbstractBackground: The high rate of attrition along the care cascade of infection with human immunodeficiency virus(HIV) results in lost opportunities to provide timely antiretroviral therapy (ART) and to prevent unnecessarily highmortality. This study aims to assess the effectiveness of a structural intervention, the one-stop (“One4All”) strategythat streamlines China’s HIV care cascade with the intent to improve testing completeness, ART initiation, viralsuppression, and mortality.Method: A two-arm, cluster-randomized controlled trial was implemented in twelve county hospitals in Guangxi Chinato test the effectiveness of the One4All strategy (intervention arm) compared to the current standard of care (SOC;control arm). The twelve study hospitals were selected for homogeneity and allocated one-to-one to the interventionand control arms. All patients screening HIV positive in study hospitals were enrolled. Target study enrollment was 180participants per arm, 30 participants per hospital. Basic demographic information was collected as well as HIV riskbehavior and route of infection. In intervention hospitals, patients then went on to receive point-of-care CD4 testingand in-parallel viral load (VL) testing whereas patients in control hospitals progressed through the usual SOC cascade.The primary outcome measure was testing completeness within 30 days of positive initial HIV screening result. Testingcompleteness was defined as receipt of all tests, test results, and post-test counseling. The secondary outcomemeasure was ART initiation (receipt of first ART prescriptions) within 90 days of positive initial HIV screening result.Tertiary outcome measures were viral suppression (≤200 copies/mL) and all-cause mortality at 12 months.Discussion: We expect that this first-ever, cluster-randomized controlled trial of a bundle of interventions intended tostreamline the HIV care cascade in China (the One4All strategy) will provide strong evidence for the benefit ofaccelerating diagnosis, thorough clinical assessment, and ART initiation via an optimized HIV care cascade. Wefurthermore anticipate that this evidence will be valuable to policymakers looking to elevate China’s overall HIV/AIDSresponse to meet the UNAIDS 90-90-90 targets and the broader, global goal of eradication of the HIV/AIDS epidemic.Trial registration: ClinicalTrials.gov #NCT02084316. (Registered on March 7, 2014)Keywords: HIV test, Point-of-care, CD4 count, Viral load, Viral suppression, Antiretroviral therapy, Mortality, Linkage tocare, HIV care cascade, HIV continuum of care* Correspondence: wuzy@263.net; wuzunyou@chinaaids.cn1The National Centre for AIDS/STD Prevention and Control, Chinese Centerfor Disease Control and Prevention, 155 Changbai Road, Changping District,Beijing 102206, ChinaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Mao et al. BMC Health Services Research  (2017) 17:397 DOI 10.1186/s12913-017-2323-zBackgroundIn both resource-rich and resource-limited settings alike,patients are commonly lost at each step along the carecascade of infection with human immunodeficiency virus(HIV) that causes acquired immunodeficiency syndromes(AIDS) [1]. In China, the current standard-of-care (SOC)patient pathway from screening HIV-positive to initiatingantiretroviral therapy (ART) involves multiple hospitalvisits, during which patients submit to several separateblood draws, and then experience protracted waitingperiods before they are notified of results. This causessubstantial loss to follow-up, delays in diagnosis, clinicalassessments that are incomplete, and ART initiation atlater stages of disease progression [2, 3]. Recent studieshave indicated that, in some parts of China, only approxi-mately 43% of those who screen HIV-positive in hospitalsettings received confirmatory testing [3], and only 57% ofthose confirmed HIV-positive received CD4 testing within6 months [4]. Since CD4 count has been used to deter-mine eligibility for ART, it has been estimated that nearly80% of those with newly-identified HIV infection, whowere ART-eligible, were not initiated on ART in a timelyfashion. Unfortunately, these missed opportunities forengagement in the HIV care cascade has ultimately trans-lated to preventable high mortality [2, 5].Although it seems obvious, it is critically important thatpeople living with HIV (PLHIV) step through all the mile-stones on the HIV care cascade in order for to fully realizethe benefits of ART—they must become aware of theirHIV-positive status, receive a diagnosis and clinical assess-ment, attend counseling, enroll in care and initiate ART,and adhere to their treatment regimens and follow-upschedules [1, 6]. However, merely accessing these services,and sequentially stepping through the process, may not beenough. Increasing evidence has highlighted the import-ance of the timing in which these services are accessed.Two large, recently-completed clinical trials, TEMPRANOand START, have provided evidence of benefit in not wait-ing to treat HIV infection until evidence of disease progres-sion is present in the form of symptoms or low CD4 count[7, 8]. This benefit of reduced clinical disease progression,as well as decreased overall mortality, has since been con-firmed in other settings, including China [2]. In response,the World Health Organization (WHO) modified its guide-lines in 2015, recommending immediate ART initiation forall PLHIV regardless of symptoms or CD4 count [9].The strong evidence for the clinical benefit of immediateART for PLHIV [7, 8], taken together with the previously-identified public health benefit in the form of reducedtransmission rates [10, 11], meant that optimization of theHIV care cascade was even more important than previouslythought. A number of studies in a variety of settings havebeen designed to examine a range of interventions intendedto improve the HIV care cascade. Thus far, majority ofthese types of studies in low- and middle-income settingshave been in Sub-Saharan Africa [12]. However, in general,they have not followed participants beyond the pre-ARTperiod and have not evaluated the effectiveness of multipleinterventions implemented as a package [13–19]. One suchstudy has recently been completed and published in China,providing evidence that a simplified test-and-treat interven-tion can improve both HIV care cascade retention and timeto ART initiation [2].Objective and outcomesThe objective of the present clinical trial was to evaluate astreamlined pathway for patients from screening HIV-positive to initiating ART, testing the effectiveness of astructural intervention—called the one-stop (“One4All”)strategy. This experimental strategy consisted of a newalgorithm incorporating rapid, point-of-care (POC) HIVscreening and CD4 testing, and in-parallel plasma viral load(VL) testing to promote fast and complete HIV diagnosisand staging, followed by immediate counseling and ARTinitiation for eligible patients. The One4All interventionwas compared to the current SOC (control arm), and fouroutcomes were assessed:1. Testing completeness within 30 days of HIV-positivescreening (primary)2. ART initiation within 90 days of HIV-positivescreening (secondary)3. Viral suppression (≤200 copies/mm3) at 12 monthsafter HIV-positive screening (tertiary)4. All-cause mortality at 12 months after HIV-positivescreening (tertiary)MethodDesignA cluster-randomized trial was designed as diagrammed inFig. 1. Hospitals were the unit of randomization and twelvesimilar county-level hospitals were allocated one-to-one tothe intervention and control arms of the study. The efficacyof the intervention was determined by comparison to thecurrent standard of care (SOC, control arm) at 30 days,90 days, and 12 months.SettingThe setting for this study was county hospitals in GuangxiZhuang Autonomous Region (hereafter referred to asGuangxi). In 2011, Guangxi reported the second highestcumulative number of HIV/AIDS cases in China, thehighest number of newly-reported HIV/AIDS cases, thehighest number of newly-reported AIDS cases, and thehighest number of AIDS-related deaths [5]. As shown inFig. 2, a broad range of characteristics of potential county-level hospitals in Guangxi were examined. The hospitalsMao et al. BMC Health Services Research  (2017) 17:397 Page 2 of 9included in the study were selected based on their homo-geneity across these characteristics.Allocation and blindingThe twelve selected study hospitals were stratified byhistorical rates of testing completeness followed by post-test counseling during the first six months of 2013 priorto being allocated to the intervention or control studyarms. Eight hospitals were within the low historical testingcompletion stratum of <20%, and these eight hospitalswere randomized, four into the intervention arm and fourinto the control arm. Four hospitals were within the highhistorical testing completion stratum of ≥20%, and thesefour hospitals were randomized, two into the interventionarm and two into the control arm. All patients whoscreened HIV-positive in the six intervention hospitalsreceived the One4All intervention and all patients whoscreened HIV-positive in the six control hospitals duringthe same study period received SOC services. There wasno blinding required since the unit of randomization wasthe study hospital and each hospital only offered the HIVcare cascade it was assigned during randomization—SOCor One4All.Participants and recruitmentStudy eligibility criteria were: [a] being 18 years of age orolder, [b] having a positive result on an HIV enzymeimmunoassay (EIA) screening test, [c] seeking care in astudy hospital, either inpatient or outpatient, and [d] resid-ing or intending to reside within the study catchment area.Study exclusion criteria were: [a] having previously receivedconfirmation of HIV infection in any setting, [b] being aprisoner or detainee at time of screening, or [c] being apregnant woman.No special strategies to ensure adequate participantenrollment were employed. All patients who were foundto be eligible were enrolled in the study. After enrollment,basic demographic information was collected. Informationon self-reported risk behavior and route of HIV acquisi-tion were also collected.EthicsThe protocol and consent process were reviewed andapproved by the US National Institutes of Health (NIH),National Institute on Drug Abuse (NIDA) ProtocolReview Board and Data and Safety Monitoring Board, aswell as respective institutional review boards (IRBs) of theUniversity of California, Los Angeles (UCLA) and theCounty Hospitals (N=12)RandomizationControl ArmStandard of Care (SOC)County Hospitals (n=6)Target Enrollment = 30 participants per hospital ×6 hospitals = 180 participantsIntervention ArmOne4All Test InterventionCounty Hospitals (n=6)Target Enrollment = 30 participants per hospital ×6 hospitals = 180 participantsAssessment:• 30 days• 90 days• 12 monthsBaseline AssessmentFig. 1 Diagram of study design. This diagram illustrates the cluster-randomized controlled study design with one-to-one allocation oftwelve county-level hospitals to either the control arm (currentstandard of care [SOC]) or the intervention arm (the “One4All” strategyintended to streamline the HIV care cascade)Fig. 2 List of Hospital Characteristics. The eighteen differentcharacteristics listed here were examined in the process of assessingcandidate county-level hospitals in Guangxi, China. The final twelvestudy hospitals were selected based upon homogeneity acrossthese characteristicsMao et al. BMC Health Services Research  (2017) 17:397 Page 3 of 9National Center for AIDS/STD Control and Prevention(NCAIDS), China Center for Disease Control and Preven-tion (CDC).Individual informed consent was not obtained. Studyinformation was shared with participants and participantswere given opportunities ask questions. While participantscould refuse to participate, or to share their data, theirenrollment still counted toward the total number of studyparticipants. No study-related reimbursement for partici-pants was provided.All patient data were protected and kept confidentialaccording to standard procedures present within China’shealthcare system. Participant data extracted from hospitalrecords, and from China’s National HIV/AIDS Comprehen-sive Response Information Management System (CRIMS,which has been described elsewhere) [20], were assignedtrial-specific participant identification numbers and thenotherwise de-identified prior to use to protect participantprivacy.Sample sizeA minimum sample size of 180 participants per arm, across12 clusters (or 30 participants per cluster) was selected. Thissample size was expected to achieve 93% power based on aone-sided test (at alpha = 0.05) to detect a difference be-tween the group proportions of 0.28, where under the alter-native hypothesis the One4All arm proportion was assumedto be 0.50 and the control arm proportion was assumed tobe 0.22. This calculation assumed an intra-class correlation(ICC) within hospitals of 0.082 based on preliminary data.Although 30 participants per cluster (or hospital) was thetarget, it was expected that some hospitals would enrollparticipants more quickly than others. Therefore, a finalstudy population of greater than 180 participants per armwas anticipated as enrollment at all hospitals remained openuntil the last hospital met its 30-participant goal.InterventionA high-level overview of the HIV care cascade in SOC(control arm) and One4All (intervention arm) hospitals isdepicted in Fig. 3. The One4All test intervention includedrapid, POC HIV EIA screening and CD4 testing, with in-parallel VL testing. This strategy was intended to promoterapid and complete diagnostic assessment and acceleratetime to ART initiation for those deemed ART eligible.The ART eligibility threshold at the time of this study wasCD4 count ≤350 cells/mm3. Each of the steps in the HIVcare cascade up to assessment of ART eligibility aredescribed in detail below for both the SOC condition andthe One4All condition. In both care cascades, all partici-pants received post-test counselling each and every timethey received test results.HIV ScreeningIn SOC study hospitals, the Wantai Screening HIV (1 + 2)Ag&Ab EIA (Beijing Wantai Manufacturer of InfectiousDiseases Diagnostics) was used as the initial screening testfollowed by two different EIAs that varied between sites.In One4All study hospitals, the first HIV screening testwas the also the Wantai Screening HIV (1 + 2) Ag&AbEIA. The second and the third EIAs were the DetermineHIV-1/2® rapid test (Abbott Laboratories) and the InTecHIV rapid test (Xiameng InTec Products).HIV ConfirmationIn SOC study hospitals, an additional blood sample wascollected immediately following screening or on a subse-quent study hospital visit. The sample was sent to the localcity CDC laboratory for Western blot (WB) confirmatorytesting according to usual practice. As confirmatory testingrequired approximately 10 to 15 days to be completed,participants typically had to be contacted and asked toreturn to the study hospital to receive their results.In One4All study hospitals, WB was not used for con-firmatory testing. Rather, VL testing served as the confirma-tory test (see below).CD4 CountIn SOC study hospitals, participants were asked to returnto the study hospital for another blood draw for CD4testing after they were notified that the result of theirconfirmatory testing was positive. Blood specimens wereagain sent to the city CDC laboratory, this time for CD4testing. CD4 testing also required approximately 10 toFig. 3 Illustration of Intervention and Control Conditions. Thisillustration depicts the differences in the HIV care cascade betweencontrol (standard of care [SOC]) and intervention (“One4All” strategy)study arms. In the SOC condition, HIV care cascade steps 2, 3, 4, 5, and7 are all consolidated into a single step 2 in the One4All conditionMao et al. BMC Health Services Research  (2017) 17:397 Page 4 of 915 days to complete. Once CD4 test results were available,participants were again asked to return to be notified.In One4All study hospitals, POC CD4 testing was per-formed in the study hospital with a POC Pima™ CD4Analyzer (Alere Healthcare, USA) using whole bloodsamples. Results were available within 30 min. Partici-pants were notified of their results as soon as possible,but no later than the next day if they had left the studyhospital during the test.Viral loadIn SOC study hospitals, VL testing was conducted ap-proximately one year after ART initiation. Participantswere asked to return to the study hospital for bloodsample collection. Blood samples were sent to theprovincial CDC laboratory for plasma VL testing, whichrequired 10 to 15 days to complete. Participants werecontacted and asked to return to the study hospital toreceive VL testing results once they were available.In One4All study hospitals, blood samples for VL testingwere collected immediately following receipt of positivescreening results, at the same time as the blood draw forCD4 testing (see above). VL testing was conducted againafter one year. Plasma VL testing itself was conducted inthe same manner as in SOC study hospitals—samples weresent to the provincial CDC laboratory, plasma VL testingrequired 10 to 15 days, and participants asked to return tothe study hospital to receive VL test results.This intervention condition posed no specific safety riskto participants. In comparison to SOC study hospitals, theHIV care cascade in One4All study hospitals involvedfewer blood draws and there was no increase in thevolume of blood drawn or the typical risk associated withblood draws.OutcomesThe primary outcome measure was the proportion ofparticipants who achieved testing completeness and re-ceived test results and post-test counseling within 30 daysof positive screening. Testing completeness was defined ascompletion of three required components: 1) initial HIVscreening (one to two tests in One4All and two to four inSOC), 2) CD4 testing, and 3) confirmatory HIV test-ing—WB in the SOC arm, or VL in the One4All arm.“Success” was defined as completion of all three requiredtests and notification of test result along with receipt ofcounseling after each test. Success was compared at 30 daysafter initial HIV-positive screening. The secondary outcomemeasure was the proportion of participants who initiatedART within 90 days from the date of HIV-positive screen-ing. ART-initiation was defined as the receipt of the firstART prescription. Tertiary outcomes included the propor-tion of participants who achieved viral suppression (≤200copies/mm3) and mortality at 12 months. Mortality wasassessed as the number of reported deaths divided by thenumber of patients enrolled.Aside from these main outcome measures, three othermeasures were pre-specified. For the primary outcome, twofurther analyses were planned: time from initial positiveHIV screen to success, not restricted to the 30-day window,compared between control and intervention arms, and suc-cess at 45, 60, and 90 days, also compared between controland intervention arms. For the secondary outcome, onefurther analysis was planned: time from initial positive HIVscreen to ART initiation, not restricted to the 90-day win-dow, compared between control and intervention arms.TimelineA general study timeline is depicted in Fig. 4. Significantpreparation prior to trial commencement was requiredand included completion of the IRB processes at multipleinstitutions, baseline hospital assessment, selection, andallocation, and staff training and site preparation. Thisphase was expected to take approximately 14 months.Once the study began, implementation and enrollmentwere anticipated to take 9 months, and follow-up a further12 months. After trial completion, data analysis andreporting was planned for the following 12 months.Data collection, management, and monitoringBaseline assessment information, including participantcontact and demographic information, laboratory screeningresults and dates, test result notification dates, and post-testcounseling dates were all collected using case reportingforms (CRFs). CRF data were stored in a web-baseddatabase constructed specifically for the study. Thisdatabase served as the centralized location for electronicstorage of all study-related data. Study-related data fromeach hospital system was retrieved weekly and uploaded tothis study database.To supplement the data contained in the study-specificdatabase, records were extracted from CRIMS [20]. Eachcounty hospital nationwide, including all participating studyhospitals, have unique identification codes in CRIMS tofacilitate extraction of all data on HIV cases identified atstudy hospitals.A centralized Data and Statistics Team (DST) wasresponsible for the validation of the web-based study data-base, retrieval of study hospital data, data extraction fromCRIMS, merger of the two datasets into one, assurance ofdata integrity and security, and development and deliveryof training for participating hospital and CDC staffmembers on applicable data management procedures.Data analysisTo adjust for both the clustering effect of hospitals (withhospital as a random effect to account for the ICC) andbaseline participant-level and hospital-level confoundingMao et al. BMC Health Services Research  (2017) 17:397 Page 5 of 9factors, analysis of the primary outcome measure wasperformed using G-side GLIMMIX modeling. Participant-level factors included in the model were age, gender,ethnicity, education, occupation, marital status, transmis-sion route, and treatment setting (inpatient/outpatient),and hospital-level factors included baseline test comple-tion rate prior to randomization. To assess the consistencyof the primary outcome results, a Chi-square test adjustedfor clustering was also used to measure the associationbetween treatment arm and the primary outcome [21],although this method does not account for baselineconfounders. Secondary analyses of the primary outcomemeasure were performed as a waiting time analysisexamining time from initial positive screen to success. Forthis waiting time analysis, participants who did not meetthe testing success criteria were censored at their lastfollow-up dates. Kaplan-Meier curves were used to displaydifferences over time from screening to testing and coun-seling completeness.In comparing differences in baseline characteristicsbetween the two arms, a Chi-square test adjusted for cluster-ing was performed for binary variables. When baselinecategorical variables had more than 2 responses, an adjustedp-value was calculated by referring the observed (un-adjusted) Chi-square value to the distribution of Chi-squarevalues obtained by randomly permuting observed treatmentsover hospitals, with the adjusted p-value defined as theproportion of 'permuted' Chi-square values at least as largeas observed. To test between-arm differences in continuousdata, a t-test adjusted for clustering was performed [21].Analytical methods and the adjustment for clusteringand covariates effects used in the analysis of the primaryoutcome were also applied to the ART initiation second-ary outcome and viral suppression tertiary outcome. Aswith the secondary analysis of the primary outcome,Kaplan-Meier analyses were also performed for time toinitiation of ART, outside the 90-day window, and time todeath. In addition, for the mortality outcome, a randomeffects Cox (shared frailty) model accounting for the hos-pital clustering effect was used to calculate the interven-tion effect on the hazard ratio (HR) of death while alsocontrolling for other covariates. Adjustments to p-valuesand confidence intervals for multiple testing and multiplecomparisons were not performed. All data analyses wereperformed using SAS software (SAS Institute, Inc., USA).DisseminationThe trial is registered with ClinicalTrials.gov (#NCT02084316) and the original protocol has been published onCTNDisseminationLibrary.org. The dataset is availableupon request to the corresponding author.DiscussionTo our knowledge this is the first controlled clinical trial toexamine a bundle of interventions intended to improvemultiple points in China’s HIV care cascade. A notablestrength of this study lies in the rigor with which the effect-iveness of the One4All strategy was tested. We expectedthat it would provide strong evidence in support of astreamlined HIV care cascade that accelerates diagnosis,Fig. 4 Timeline of the Study. The study was expected to require approximately 14 months for pre-implementation preparation (i.e. IRB preparation,submission, review, and approval; baseline hospital assessment, selection, and allocation; and staff training and site preparation). Implementation andenrollment were expected to take a further 9 months, and follow-up required a subsequent 12-month period. Finally, data analysis and reporting wereexpected to require another 12 months. (institutional review board [IRB], antiretroviral therapy [ART])Mao et al. BMC Health Services Research  (2017) 17:397 Page 6 of 9complete clinical assessment, and treatment initiationhaving a significant clinical benefit over the current SOC.With growing global evidence of the clear benefits ofimmediate ART for all PLHIV, both for the purposes oftreatment of HIV infection and of prevention of HIV trans-mission [7, 8, 10], HIV/AIDS programs around the worldmust place renewed focus on optimization of the HIV carecascade. Worldwide, three out of every five PLHIV havenot accessed ART [22]. Steps must be taken to eliminatelosses to follow-up in the pre-ART period, and to ensurethat all PLHIV receive immediate treatment.At the time the present study was conducted, China stillhad a CD4 count ≤350 cells/mm3 ART eligibility criterionin place, with exceptions allowed only for PLHIV whowere pregnant or in serodiscordant relationships [23]Thus, we expected that some participants in our studycould not initiate ART despite being successfully retainedin the pre-ART HIV care cascade. The WHO has sinceelevated the CD4 count threshold for ART to CD4 count≤500 cells/mm3 in 2013 [24], and then abolished it in2015, now recommending that all PLHIV receive immedi-ate ART regardless of CD4 count [9]. Two studies inChina have recently examined the benefit of ART initi-ation for PLHIV who did not meet the ≤350 cells/mm3threshold. One prospectively examined immediate ARTfor all participants regardless of CD4 count [2], the otherretrospectively investigated ART given to PLHIV withCD4 counts >500 cells/mm3 on an exception basis or as apart of the NCAIDS-sponsored “Prevention and Treat-ment of Major Infectious Diseases” project launched in2012, which consisted of a series of cluster-randomizedcontrolled trials among key populations. Both studiesfound significant improvements in mortality. We envisionthe combination of this evidence with the outcome of thepresent study supporting the complete replacement ofChina’s current SOC HIV care cascade with a new,streamlined cascade that accelerates diagnosis, completeclinical assessment, and ART initiation for all PLHIV.Several important challenges were anticipated prior totrial commencement. For example, there was knownvariability in HIV SOC offered between different countiesin Guangxi. A major cause of poor testing completenessand treatment initiation in rural Guangxi is disjointed HIVservices as a result of structural and cultural constraints.County hospitals and CDCs struggle to follow nationalguidelines, mainly because of limited laboratory capacityand complex referral systems between hospitals and CDCs.The study team debated whether to provide strict guide-lines and training to control arm study hospitals in an effortto standardize their current practice and improve homo-geneity. However, we ultimately decided against tamperingwith the control condition, as we determined that a controlcomparator that represented the reality of the currentstandard HIV care cascade was more valuable than anartificial ideal control environment we would have createdonly for the benefit of the study. Therefore, during ourbaseline hospital investigations, the investigators engagedwith hospital leaders and key personnel involved in HIVcare and collected the key characteristics of the routinepractice adopted at each county. A detailed testing totreatment flowchart was developed for each of the hospi-tals, documenting the key deviations from national guide-lines. Then, CRFs were adjusted in order to accuratelyreflect the reality of testing and linkage to care proceduresin each hospital.Another key challenge was in the size of the study. Thestudy was originally designed to include 24 hospitals (orclusters), 12 each in the control and intervention arms.This would have provided adequate power to detect differ-ences between arms in all four outcomes. However, due toboth the known high degree of variability observed inSOC services across counties in Guangxi and historicallylow numbers of patients screening HIV positive in manyhospitals, not to mention the inherent difficulties withimplementing such a large study, we decided to scale backthe number of hospitals to 12. Although we suspected thatwe would still find statistically significant differences inseveral of the outcomes, the study was truly only ad-equately powered to detect between-arm differences inthe primary outcome of testing completeness by 30 days.As the study included no intervention related to ARTregimen adherence, we anticipated the possibility that wemay find no between-arm differences in achievement ofviral suppression. We chose to define viral suppression as≤200 copies/mL based on the equipment used for thestudy in local laboratories, recognizing that this target wasaggressive. Although above the <50 copies/mL definitionof viral suppression commonly recognized in high-incomesettings, it is well below the <1,000 copies/mL thresholdrecognized by the WHO as achievement of viral suppres-sion in low- and middle-income countries such as China.It is also below the 300 to 500 copies per/mL target com-monly used by research teams in studies in low- andmiddle-income settings [1, 25]. Future studies will need toexamine interventions aimed at improving adherence inorder to promote achievement of viral suppression.Additionally, with increased numbers of PLHIV on ART,it will be necessary for China’s public health and health-care communities to be vigilant in monitoring patients fortreatment failure. This will require substantial improve-ment of VL testing infrastructure and China shouldstrongly consider investment in the development and im-plementation of new POC VL testing technologies, whichfurther contribute to a streamlined HIV care cascade.ConclusionThis first-ever, cluster-randomized controlled trial of abundle of interventions intended to streamline the HIVMao et al. BMC Health Services Research  (2017) 17:397 Page 7 of 9care cascade in China (the One4All strategy) wasevaluated on four outcomes: testing completeness within30 days, ART initiation within 90 days, viral suppressionat 12 months, and all-cause mortality at 12 months. Weexpect that the outcome of this trial will provide strongevidence for the benefit of accelerating diagnosis, thor-ough clinical assessment, and ART initiation via an opti-mized HIV care cascade such as the One4All strategy.Furthermore, we anticipate that this evidence taken to-gether with the strong, new evidence of both treatmentand prevention benefits of immediate ART for all PLHIVregardless of CD4 count [7, 8, 10], will compel carefulconsideration by policymakers in China and other, simi-lar low- and middle-income countries. HIV care cascadeoptimization is a critical strategy for meeting theUNAIDS 90-90-90 targets [26], and will be required forthe eventual eradication of the HIV epidemic.AbbreviationsAIDS: Acquired immunodeficiency syndromes; ART: Antiretroviral therapy;CDC: Center for disease control and prevention; CRF: Case reporting form;CRIMS: Comprehensive response information management system; DST: Dataand statistics team; EIA: Enzyme immunoassay; HIV: Humanimmunodeficiency virus; HR: Hazard ratio; ICC: Intra-class correlation;IRB: Institutional review boards; NCAIDS: National Center for AIDS/STDControl and Prevention; NIDA: National Institute on Drug Abuse;NIH: National Institutes of Health; One4All: The one-stop strategy interven-tion; PLHIV: People living with HIV; POC: Point-of-care; SOC: Standard-of-care;UCLA: University of California, Los Angeles; UNAIDS: The Joint United NationsProgramme on HIV/AIDS; VL: Viral load; WHO: World Health OrganizationAcknowledgementsThe authors would like to thank Betty Tai and Shicheng Yu for their help indeveloping the protocol.FundingThis study was co-funded by the National Institute on Drug Abuse, US NationalInstitutes of Health (Grant #CTN0056) and the National Health and FamilyPlanning Commission, People’s Republic of China (Grant #131-13-000-105-01).Scientists from the National Institute on Drug Abuse were involved in studydesign and the writing of this manuscript.Availability of data and materialsNot applicable.Authors’ contributionsYM, ZW, DL, DG, LE, WL, PV, RD, AH, RL, JSGM, ZT and YZ contributed to thestudy design and authored the study protocol. YM, ZW, JMM and DG wrotethe initial draft of this manuscript. All authors reviewed all manuscriptrevisions. ZW had final responsibility for the decision to submit themanuscript for publication and will have access to the final trial dataset. Allauthors read and approved the final manuscript.Competing interestsPaul VanVeldhuisen is a statistician and Robert Lindblad is a medical officerfrom the Emmes Corporation. The Emmes Corporation, USA (https://secure.emmes.com/emmesweb/about) is a health-related research institute.It was contracted by the US National Institute on Drug Abuse (NIDA) withopen competition for Data Center to provide support for data managementand data analysis for NIDA’s supported Clinical Trial Network studies on drugabuse. Emmes is the Data Center for our trial. NIDA provided funding toEmmes to support both PV and RL and their team for technical support inthe design of data collection forms, data management, data analysis, and in-terpretation of results for the trial. The rest of the authors declare that theyhave no competing interests.Consent for publicationNot applicable.Ethics approval and consent to participateThe trial protocol and consent process have been reviewed and approvedby the Protocol Review Board and Data and Safety Monitoring Board of theNational Institute on Drug Abuse, US National Institutes of Health and theInstitutional Review Boards of both the University of California, Los Angelesand the National Center for AIDS/STD Control and Prevention, China CDC.All the ethics committees had approved this study specifically waived theneed to obtain individual written participant consent.DisclaimerThe views and opinions expressed herein are entirely those of the authors,and do not represent the official policy, or endorsement of the NationalCenter for AIDS/STD Control and Prevention, China CDC, nor the NationalInstitute on Drug Abuse, US National Institutes of Health.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1The National Centre for AIDS/STD Prevention and Control, Chinese Centerfor Disease Control and Prevention, 155 Changbai Road, Changping District,Beijing 102206, China. 2National Institute on Drug Abuse, US NationalInstitutes of Health, Bethesda, USA. 3Integrated Substance Abuse Programs,University of California at Los Angeles (UCLA) School of Medicine, LosAngeles, USA. 4Emmes Corporation, Rockville, USA. 5Department ofEpidemiology, UCLA Fielding School of Public Health, Los Angeles, USA.6British Columbia Centre for Excellence in HIV/AIDS, University of BritishColumbia, Vancouver, Canada. 7Guangxi Centre for Disease Control andPrevention, Nanning, China.Received: 3 March 2017 Accepted: 19 May 2017References1. MacCarthy S, Hoffmann M, Ferguson L, Nunn A, Irvin R, Bangsberg D,Gruskin S, Dourado I. The HIV care cascade: models, measures and movingforward. Journal of the International AIDS Society. 2015;18:19395.2. Wu Z, Zhao Y, Ge X, Mao Y, Tang Z, Shi CX, Chen C, Li Y, Qiu X, Nong G, et al.Simplified HIV Testing and Treatment in China: Analysis of Mortality RatesBefore and After a Structural Intervention. PLoS Med. 2015;12(9):e1001874.3. Gu D, Mao Y, Tang Z, Montaner J, Shen Z, Zhu Q, Detels R, Jin X, Xiong R,Xu J, et al. Loss to Follow-Up from HIV Screening to ART Initiation in RuralChina. PLoS One. 2016;11(10):e0164346.4. Tang H, Mao Y, Shi CX, Han J, Wang L, Xu J, Qin Q, Detels R, Wu Z. BaselineCD4 cell counts of newly diagnosed HIV cases in China: 2006-2012. PLoSOne. 2014;9(6):e96098.5. National Center for AIDS/STD Control and Prevention. China CDC: Analysisof HIV/AIDS testing and treatment program in Guangxi. Beijing: NationalCenter for AIDS/STD Control and Prevention, China CDC; 2012.6. Gardner EM, McLees MP, Steiner JF, Del Rio C, Burman WJ. The spectrum ofengagement in HIV care and its relevance to test-and-treat strategies forprevention of HIV infection. Clin Infect Dis. 2011;52(6):793–800.7. Danel C, Moh R, Gabillard D, Badje A, Le Carrou J, Ouassa T, Ouattara E,Anzian A, Ntakpe JB, Minga A, et al. A Trial of Early Antiretrovirals andIsoniazid Preventive Therapy in Africa. N Engl J Med. 2015;373(9):808–22.8. UNAIDS. Implications of the START study data: questions and answers.Geneva: UNAIDS; 2015.9. WHO. Guideline on when to start antiretroviral therapy and pre-exposureprophylaxis for HIV. Geneva: World Health Organization; 2015.10. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, KumarasamyN, Hakim JG, Kumwenda J, Grinsztejn B, Pilotto JH, et al. Prevention of HIV-1infection with early antiretroviral therapy. N Engl J Med. 2011;365(6):493–505.11. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, KumarasamyN, Hakim JG, Kumwenda J, Grinsztejn B, Pilotto JH, et al. Antiretroviral Therapyfor the Prevention of HIV-1 Transmission. N Engl J Med. 2016;375(9):830–9.Mao et al. BMC Health Services Research  (2017) 17:397 Page 8 of 912. Govindasamy D, Meghij J, Kebede Negussi E, Clare Baggaley R, Ford N,Kranzer K. Interventions to improve or facilitate linkage to or retention inpre-ART (HIV) care and initiation of ART in low- and middle-incomesettings–a systematic review. J Int AIDS Soc. 2014;17:19032.13. Wanyenze RK, Hahn JA, Liechty CA, Ragland K, Ronald A, Mayanja-Kizza H,Coates T, Kamya MR, Bangsberg DR. Linkage to HIV care and survivalfollowing inpatient HIV counseling and testing. AIDS and behavior. 2011;15(4):751–60.14. Jani IV, Sitoe NE, Alfai ER, Chongo PL, Quevedo JI, Rocha BM, Lehe JD, PeterTF. Effect of point-of-care CD4 cell count tests on retention of patients andrates of antiretroviral therapy initiation in primary health clinics: anobservational cohort study. Lancet. 2011;378(9802):1572–9.15. Faal M, Naidoo N, Glencross DK, Venter WD, Osih R. Providing immediateCD4 count results at HIV testing improves ART initiation. J Acquir ImmuneDefic Syndr. 2011;58(3):e54–59.16. Larson BA, Schnippel K, Ndibongo B, Xulu T, Brennan A, Long L, Fox MP,Rosen S. Rapid point-of-care CD4 testing at mobile HIV testing sites toincrease linkage to care: an evaluation of a pilot program in South Africa. JAcquir Immune Defic Syndr. 2012;61(2):e13–17.17. Labhardt ND, Motlomelo M, Cerutti B, Pfeiffer K, Kamele M, Hobbins MA,Ehmer J. Home-based versus mobile clinic HIV testing and counseling inrural Lesotho: a cluster-randomized trial. PLoS Med. 2014;11(12):e1001768.18. Leon N, Mathews C, Lewin S, Osler M, Boulle A, Lombard C. A comparison oflinkage to HIV care after provider-initiated HIV testing and counselling (PITC) versusvoluntary HIV counselling and testing (VCT) for patients with sexually transmittedinfections in Cape Town, South Africa. BMC Health Serv Res. 2014;14:350.19. Rosen S, Maskew M, Fox MP, Nyoni C, Mongwenyana C, Malete G, Sanne I,Bokaba D, Sauls C, Rohr J, et al. Initiating Antiretroviral Therapy for HIV at aPatient's First Clinic Visit: The RapIT Randomized Controlled Trial. PLoS Med.2016;13(5):e1002015.20. Mao Y, Wu Z, Poundstone K, Wang C, Qin Q, Ma Y, Ma W. Development ofa unified web-based national HIV/AIDS information system in China. Int JEpidemiol. 2010;39 Suppl 2:ii79–89.21. Donner A, Klar N. Analysis of binary outcomes Design and Analysis ofCluster Randomization Trials in Health Research. Chichester: John Wiley &Sons, Ltd; 2000. p. 84–6.22. UNAIDS. The GAP Report. Geneva: UNAIDS; 2014.23. Panel E. China National Free Antiretroviral Threapy Handbook. 3rd ed.Beijing: The People's Medical Publishing House Co., LTD; 2012.24. WHO. Consolidated guidelines on the use of antiretroviral drugs for treatingand preventing HIV infection—recommendations for a public healthapproach. Geneva: World Health Organization; 2013.25. McMahon JH, Elliott JH, Bertagnolio S, Kubiak R, Jordan MR. Viral suppressionafter 12 months of antiretroviral therapy in low- and middle-income countries:a systematic review. Bull World Health Organ. 2013;91(5):377–385E.26. UNAIDS. 90-90-90 an ambitious treatment target to help end the AIDSepidemic. Geneva: UNAIDS; 2014.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Mao et al. BMC Health Services Research  (2017) 17:397 Page 9 of 9


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items