RESEARCH ARTICLE Open Access

Acceptability and feasibility of point-of-careCD4 testing on HIV continuum of care inlow and middle income countries: asystematic reviewMinh D. Pham1,2*, Paul A. Agius1, Lorena Romero3, Peter McGlynn2, David Anderson1,4, Suzanne M. Crowe1,5,6

and Stanley Luchters1,2,7

Abstract

Background: CD4 testing is, and will remain an important part of HIV treatment and care in low and middleincome countries (LMICs). We report the findings of a systematic review assessing acceptability and feasibility ofPOC CD4 testing in field settings.

Methods: Electronic databases were searched for studies published in English between 2005 and 2015 that describePOC CD4 platforms. Studies conducted in LMICs and under field conditions outside a laboratory environment wereeligible. Qualitative and descriptive data analysis was used to present the findings.

Results: Twelve studies were included, 11 of which were conducted in sub-Saharan countries and used one POC CD4test (The Alere Pima CD4). Patients reported positively regarding the implementation of POC CD4 testing at primaryhealth care and community level with ≥90 % of patients accepting the test across various study settings. Health serviceproviders expressed preference toward POC CD4 testing as it is easy-to-use, efficient and satisfied patients’ needs to agreater extent as compared to conventional methods. However, operational challenges including preference towardvenous blood rather than finger-prick sampling, frequent device failures and operator errors, quality of training for testoperators and supervisors, and increased staff workload were also identified.

Conclusions: POC CD4 testing seems acceptable and feasible in LIMCs under field conditions. Further studies usingdifferent POC CD4 tests available on the market are required to provide critical data to support countries in selectionand implementation of appropriate POC CD4 technologies.

Keywords: Point-of-care testing, CD4, Pima, acceptability, feasibility, systematic review

BackgroundFor many years, CD4 count testing has been a key diag-nostic tool to identify HIV positive patients eligible forantiretroviral therapy (ART) and monitoring patient re-sponses to treatment. Recently, in light of new evidencesupporting early treatment [1], there is a movement inpolicy recommendation and practice towards CD4-independent ART initiation and a number of countries

have already approved treatment for all HIV infected in-dividuals regardless of their CD4 count. However, in thenear future, CD4 counts still remain an important andpractical part of HIV care particularly in decision makingaround ART initiation, clinical management and treat-ment monitoring in many countries where access to viralload monitoring remains limited [2].Availability and access to CD4 count testing has been

identified as a major barrier for increasing access to ARTparticularly in low-resource settings where laboratorybased CD4 monitoring is not always available or easy toaccess [3]. The lack of reliable and affordable tests in these

* Correspondence: minh.pham@burnet.edu.au1Burnet Institute, Melbourne, VIC, Australia2Department of Epidemiology and Preventive Medicine, Faculty of MedicineNursing and Health Science, Monash University, Melbourne, AustraliaFull list of author information is available at the end of the article

© 2016 The Author(s). 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.

Pham et al. BMC Health Services Research (2016) 16:343 DOI 10.1186/s12913-016-1588-y

settings leads to missed opportunities of early ART initi-ation and significant patient loss-to-follow up [4].Point-of-care (POC) CD4 testing has been introduced

as an approach to overcome this challenge, possessingmajor advantages as compared to standard laboratory-based CD4 testing by flow cytometry in primary health-care settings that lack infrastructure support and ab-sence of well-functioning patient and/or sample referralsystems [5]. Results from recent field studies suggestthat POC CD4 tests are reliable [6–8], can help to in-crease the likelihood of an infected person having theirCD4 T-cell count measured and receive the result [9],reduce time from HIV testing to ART initiation [10, 11],facilitate rapid (same day) ART initiation among patientseligible for treatment [12, 13], reduce loss to follow up[11], and most importantly, provide an immediate CD4count which can significantly improve patient retentionon care [14].However, there is lack of synthesis of available evi-

dence regarding operation and implementation of POCtechnologies in field settings. We conducted a systematicreview to assess acceptability and feasibility of currentlyavailable or prototype commercial POC CD4 tests andto identify evidence gaps from field evaluations with ageographical focus in resource-constrained settings.

MethodsLiterature search strategyThis review was conducted following the requirementsof the preferred reporting items for systematic reviewsand meta-analyses (PRISMA). A literature search usingestablished search terms was first conducted in Medlineto identify any study describing POC CD4 tests con-ducted in low and middle income countries (LMICs)published between Jan 2005 – Jan 2015 in English (seeAdditional file 1). The search strategy was then adaptedby using the appropriate subject thesauri where availableand modifying the search syntax to the relevant softwareplatforms, and was undertaken across other electronic da-tabases including: Embase, CENTRAL, Cinahl, PsycINFO,Biological Abstracts, Scopus and Web of Science. Searcheswere also conducted in grey literature resources andhand-searching of reference lists and citation was per-formed to identify relevant studies.

Study selectionStudy inclusion criteria were defined using PICO (partic-ipants, interventions, comparisons, outcomes) format[15]. Participants (P) included HIV positive, HIV nega-tive and unknown HIV status persons aged ≥ 12 months.For intervention (I), any of the following six commer-cially available POC CD4 testing platforms listed in theUNITAID “2014 HIV/AIDS Diagnosis TechnologyLandscape” report [16] were included: (1) PointCare

NOW™ (PointCare Technology Inc, Marlborough, MA,USA); (2) Pima™ CD4 (Alere Inc, Waltham, MA, USA);(3) Daktari™ CD4 Counter (Daktari Diagnostics Inc,Cambridge MA, USA); (4) CyFlow® CD4 miniPOC(Partec, Munich, Germany); (5) BD FACSPresto™ (BDBiosciences, San Jose, CA, USA); and (6) MyT4™ CD4Test (Zyomyx Inc, Fremont, CA, USA). Comparators/controls (C): Laboratory based CD4 test (Flow Cytome-try) if applicable with outcomes (O) of interest contain-ing information on acceptability, feasibility of POCCD4 testing in field settings (Fig. 1).

Data extraction and data synthesisA pre-constructed electronic data extraction form wasdeveloped, pre-tested and finalized by consensus amongauthors. Data extraction was conducted by one reviewerand verified by the second using the data extractionform with 20 % duplicate extraction (Five papers wererandomly selected and data extracted independently by2 reviewers and compared to identify discrepancies ifany). Quality of included studies was assessed using ei-ther the QUADAS tool for quality assessment of diag-nostic studies included in systematic review [17], theEPHPP quality assessment tool for quantitative studies[18], or a specific tool for quality assessment of qualita-tive studies [19].The following data items were extracted: author (s)

and year of publication, intervention (name of POCCD4 technology, training for test operators where men-tioned), comparison (if applicable), population (age,gender, HIV status), study setting (types of facility, loca-tion/country), study design and sample size, and keystudy outcomes of interest.

Definitions and assessment of acceptability and feasibilityAcceptability and feasibility are closely linked and affecteach other but they are not identical. Acceptability fo-cuses on individual factors (provider and patient’s per-spectives) while feasibility takes other system factors(infrastructure, human resource, policy) into account.Thus, one test could be acceptable to both provider andpatient, but may not be feasible to be deployed in aspecific setting (for example if a test device requiresconstant electricity supply or dedicated, well trainedtechnicians to operate them). To assess acceptabilityand feasibility of POC CD4 testing in field settings, aconceptual framework initially developed to explore thefeasibility, acceptance and use of a rapid diagnostic testfor malaria [20] has been adapted for use in this study.From the service provider’s perspective, acceptability isassessed by the ability to understand how to correctlyperform a POC CD4 test, their willingness to carry outthe test when necessary as part of their daily work andtheir belief that the test is relevant to their work and

Pham et al. BMC Health Services Research (2016) 16:343 Page 2 of 12

test result is accurate. From patient’s perspective, ac-ceptability is influenced by their willingness to have thetest performed on themselves, their belief that the testis convenient to take and relevant in determining theirCD4 count. Feasibility depends on acceptability and thepresence or absence of supporting system factors suchas training, monitoring/supervision, supplies, infrastruc-ture such as space, light, water etc. that enable the imple-mentation of POC CD4 testing in the field.Based on this concept, the following pre-defined mea-

surements were used to assess acceptability (1) uptake ofPOC CD4 testing: the proportion of individuals whoaccept a POC CD4 test (having blood samples drawn forPOC CD4 test) of the total number of individuals whohave been confirmed HIV positive and offered CD4 test-ing; (2) reported attributes of the POC CD4 test relatingto day-to-day field operation from the relevant stake-holders’ perspectives (clients/patients, service providers,health manager/policy makers). For assessing feasibilitythe following measurements were used: (1) reportedsystem factors associated with or having effect on

acceptability and feasibility of POC CD4 test in thefield, and (2) locally specific context and operational is-sues affecting the deployment of POC CD4 testing.

ResultsStudy characteristicsThe search identified 12 studies that reported informa-tion and data on the outcomes of interest (acceptabilityand feasibility) including 11 articles and one abstract, 11of which were conducted in Sub-Saharan countries andused Pima as a single intervention or as part of an inter-vention program to improve HIV testing uptake andlinkage to care (Table 1). Overall, the quality of includedstudies was considered between moderate and strong.The QUADAS score ranged from 7 to 12 (of a max-imum score of 14) with most of the studies scoring be-tween 10 and 12. Using the EPHPP tool, two studieswere ranked as “moderate” and one as “strong”. Onestudy deployed both quantitative and qualitative methods,and addressed most of the essential criteria listed in thetool for appraising the quality of qualitative research.

Fig. 1 Selection process of included study for a systematic review of POC CD4 test

Pham et al. BMC Health Services Research (2016) 16:343 Page 3 of 12

Table 1 Characteristics of studies included in the review

Author (s), Year Study objective Study population/Study setting Study design/Sample size Sample/data collection Intervention

Galiwango, 2014 [26] To conduct a field evaluationassessing the accuracy of Pima

HIV infected patients (pre andexperienced ART persons) atfield clinics of Rakai HealthScience Program in Rakai district,Southern rural Uganda. Studyarea has half a million population.Program is a community-basedresearch organization with afocus on HIV/AIDS andreproductive health

Cross-sectional studyA total of 903 patientswere recruited amongwhich 258 (28.5 %)patients were on ART

Venous blood samples collectedby nursing team. Data wascollected for clinical purposeand analyzed anonymously

Four Pima machines wereused (machines were movedto testing site, located nextto clinics, everyday fromcentral Lab)Prior to daily testing, normaland low controls were runon each machine. Tests wererun by qualified lab technicianswho received Pima usagetraining

van Rooyen, 2013 [21] To conduct assessment oneffect of a home-basedcounseling and testingprogram that included POCCD4 testing on (1) high HIVtesting coverage (2) identifynewly infected or unawareHIV cases (3) reduce barriersto care and (4) increase accessand adherence to ART

Known HIV-positive individualsolder than 18 years in KwaZulu-Natal, South Africa. Studypopulation characterized byhigh unemployment, low percapita income; and very highHIV prevalence (23.5 % amongpeople aged≤ 25). Study areawas within walking distanceof a primary health centerand ART clinic

Prospective cohort studywith one, three and sixmonths follow-ups bylay counselors to evaluateoutcomes281 households enrolled,671 adults consentedand tested. Among 201HIV infected participants,193 had POC CD4 test inaddition to CD4 lab-Cyflow.

POC CD4 testing was conductedat home using finger-prick bloodsample. Venous blood sampleswere collected for BDFACSCalibur

Pima (as part of home-basedcounseling and testing programincluded POC CD4 testing,facilitated counseling andreferrals)POC CD4 testing wasconducted in the home at thesame visit when positive HIVrapid test was obtained. POCtest run by lay counselor/nurseassistant; training of testoperators was not reported

Mtapuri-Zinyowera,2013 [24]

To document experience inimplementation of Pima inmaternal and new-bornchild health setting inZimbabwe

Clients (HIV positive women,lactating mothers, their familiesand other users) of healthfacilities with high-volume ANCvisits of > 100 pregnant womenseen/month located in 7 districts(five in each district) in Zimbabwewith and without Pima machinesKey informants: relevant projectstaffs, MNCH staffs, counselors,lab staffs and ART personnel

Cross-sectional study346 individuals wereinterviewed. Questionnaireswere administered to23 staff members, 62trained POC users.Observation tools wasapplied to 22 trainedusers. Client exitquestionnaire wasadministered to 142clients of POC sites and42 clients of non-POC sites. 1client FGD conducted in each of7 districts. Data of 207 client’srecords was extracted from 45facilities

Primary data was collectedthrough face-to-face interviews,focus group discussion andobservation using audiorecorders and cameras (withverbal consent). Secondary datawas extracted from medicalrecords xxx

Pima Implemented at 35 ANChigh-volume health facilitiesto provide CD4 count to HIVpositive pregnant women andtheir families in hard to reachareas. Health care cadre andtraining of test operatorsnot reported

Larson, 2012 [22] To assess the impact ofmobile HIV counseling andtesting program on theproportion of patientscompleting referral visitwithin 8 weeks of HIVtesting

Adult HIV positive patientsdiagnosed between May andNovember 2010 in a mobileHIV testing program (calledACCESS VCT) with 2 mobileunits (with tents) to conductHCT at sites (taxi rank, shoppingmall) in Gauteng Province,South Africa

Retrospective cohort studyA total of 508 patients werediagnosed with 311 patientswere offered POC CD4 and197 patients were not

Data was drawn retrospectivelyfrom routinely collected medicalrecords kept by the ACCESSVCT program and completedin Feb 2011 allow for 8 weeksfollow-up for all HIV positivepatients

Four Pima devices were usedin the same mobile locationwith each assigned to onenurse. With 6–10 nursespresent during the day oftesting patients wererandomly assigned on afirst-come first-serve basis;

Phamet

al.BMCHealth

ServicesResearch

(2016) 16:343 Page

4of

12

Table 1 Characteristics of studies included in the review (Continued)

training of test operatorsnot reported

Glencross, 2012 [31] To report and compare theperformance of Pima inlaboratory or typical SouthAfrican primary healthHCT clinics

Adult HIV patients attending(1) Hospital based antenatalHCT clinic in Johannesburg-phase II (2) Two Primary healthcare HCT clinic in Limpopoprovince-phase IIIA; and (3) Inner-city primary health care clinic inJohannesburg, South Africa-phase IIIB

Cross-sectional studyPhase II: N = 91Phase IIIA: N = 96Phase IIIB: N = 139

Both venous and capillary bloodsamples were collected

Pima operators (nursingpersonnel) were trained bythe suppliers prior tocommencing testing, accordingto methods defined by themanufacturer. Daily qualitycontrol was performed beforecommencing daily testing

Thakar, 2012 [25] To assess the use of Pimaat 21 ART centers in India

HIV positive patients aged 18–60attending 21 ART centers indifferent parts of India havingminimum (5-10/day) to moderate(25-30/day) patient load.

Cross-sectional studyTotal of 1790 participantswere consecutivelyenrolled in 21 centers(5–10 HIV positivepatients from each centre)

Both venous and capillaryblood samples were collected

Technologists were trained fortwo days for finger pricksample collection & CD4count estimation using Pimaanalyzer including the use ofcalibrators. Samples were runafter low/normal controlcartridge give acceptablevalues

Manabe, 2012 [27] To evaluate performance ofPima in both laboratory andnon-laboratory environment

HIV infected patients at AdultInfectious Diseases Institute Clinicwithin the Mulago HospitalComplex in Kampala, Uganda

Cross-sectional studyN = 206.

Both venous and finger-prickblood samples were collectedby study nurse

CD4 counts were performedusing 4 Pima devices. Duplicatemeasurements were performedon both capillary and venoussamples using 2 differentdevices. Test operator cadreand training not stated

Jani, 2011 [28] To assess the ability of nurseto produce accurate resultswith POC test in primary healthcare settings providing ART

Documented HIV infectedindividuals from general patientpopulation attending 2 primaryhealth care setting providing arange of health services includingART in Maputo, Mozambique

Cross-sectional studyN = 697.

Participants provided finger-prick (for POC tests) andvenous blood (for lab-basedtests)

Pima POC CD4 test operatorswere nurses in primary healthclinics trained by themanufacturer. Manufacturerprovided internal qualitycontrol and all POC instrumentspassed external qulity controlassessment during study period

Mtapuri-Zinyowera,2010 [23]

To evaluate the use of Pimaand the ability of both nursesand laboratory technicians torun POC CD4 test

Newly diagnosed HIV positivepatients at a VCT center at NewAfrica House in Harare, Zimbabwe

Cross-sectional studyN = 165.

Participants provided finger-prick (for POC tests) andvenous blood (for lab-basedtests).

Two Pima devices were used.Nurses and laboratorytechnicians equally run POCCD4 tests (50/50) on eachdevice.All test operators were formallytrained on the Pima device andsample collection methodologyfor half a day

Wade, 2014 [30] To assess performance andoperational characteristicsof Pima

HIV infected patients presentingfor routine CD4 testing atinfectious disease clinic in Dares Salam (Tanzania)

Cross-sectional studyN = 200.

Both capillary blood (Pima)and venous blood(FACSCalibur) were collected

Pima test operator cadre andtraining not reported. Pimatesting procedures were notdescribed

Phamet

al.BMCHealth

ServicesResearch

(2016) 16:343 Page

5of

12

Table 1 Characteristics of studies included in the review (Continued)

Mwau, 2014 [7] To evaluate the technicalperformance of MyT4POC CD4 test

HIV infected patients ≥ 18 yearsold at comprehensive HIV careclinics of 2 health care facilitiesin Busia county of Westernprovince, Kenya

Cross-sectional studyN = 276.

Finger-prick blood samples(for MyT4 test) and venousblood samples for conventionalCD4 tests collected.

All samples were collected andtested using MyT4 POC CD4 bytrained health care staffs (nursesand lab technicians)Training for staff not reported

Arnett N, 2013 [29] To assess healthcare workeracceptance and ability toperform POC CD4 test

HIV infected patients from 5PMTCT and HIV treatment sitesin Dar-es-Salaam, Tanzania

Cross-sectional study1060 patients provided bloodspecimens, 11 HCWs interviewed

Each participant provided 3samples: (1) venous (1) finger-prick directly to PIMA cartridgeand (1) finger-prick collectedinto Microtube

Pima POC CD4 tests run bytrained healthcare workers

HCT: HIV Counseling and Testing; ANC: Antenatal clinic; MNCH: Maternal and new-born child health; ART: Antiretroviral therapy; PMTCT: Prevention of mother to child transmission

Phamet

al.BMCHealth

ServicesResearch

(2016) 16:343 Page

6of

12

Acceptability and feasibility of POC CD4 testOnly one of the included studies, presented as a confer-ence abstract, aimed to assess acceptability of a POCCD4 test from the healthcare workers’ perspective. Allother studies primarily assessed accuracy and/or effectof POC CD4 on HIV program or patient related out-comes, and provided additional data related to accept-ability and feasibility of POC CD4 testing under fieldconditions (Table 2).

Patient perspectivesAvailable data suggest that the Pima test was highly ac-ceptable across study settings. Acceptance rates rangedfrom 90 % to 100 % when offered either at the home, viamobile HIV counseling and testing or in permanent vol-untary counseling and testing settings [21–23]. Patientsappreciated having on-site CD4 testing availability. Oneparticipant reported: “They used to collect our bloodand send it somewhere, hence that will take some daysbut as for now there is a big difference, we now receiveour results there and then” [24]. In terms of preference re-garding the method of blood sampling, the findings of onestudy in India [25] suggested that study participants preferto give venous blood samples versus finger-prick becauseof the requirement for other blood tests (which can onlybe performed with venous blood) and fear of being sub-jected to the discomfort of multiple pricks with more painif sufficient volume is not obtained in one prick.

Provider perspectivesOne study which employed qualitative research methodsin 35 maternal and new-born child health (MNCH)settings in Zimbabwe [24] reported health personnelexpressed a preference toward performing POC CD4testing (Pima) as it was efficient in resource use, user-friendly and responded well to patients’ needs: the ma-chine “used fewer resources (syringes and needles),and less technical expertise was needed” to run the testand it was “catering for all types of clients”; finger-prick sampling was reported as “less traumatizing tothe patients as less blood is taken for immediate use”and “no suspicion that blood was used for other pur-poses”. The Pima machine was reported as portable,compact, easy-to-use, battery operated with 3–4 h backupproviding feasibility for use in non-laboratory settings[21, 25–27] and by non-laboratory technicians [23, 28].However, relatively low throughput capacity, frequenterror codes, cartridge rejection before expiration dateand increased technical breakdown after one year ofoperation at a busy site were also reported as limita-tions of the Pima technology [24]. Healthcare workersalso noted that a full blood count is required to identifytype of antiretroviral drugs patients can be prescribedand this cannot be done with a finger-prick sample; also

problems with test errors may lead to multiple pricksbeing required [24]. In another study conducted at fivePrevention of Mother to Child Transmission (PMTCT)and HIV/AIDS care and treatment clinics in Tanzania,73 % (8/11) of health care workers who were inter-viewed named venous blood as their preferred samplecollection method and 100 % (11/11) trusted Pima-venous test results [29].

Influencing factorsHuman resource shortage is a major factor that influ-ences feasibility and acceptability of POC CD4 in fieldsettings. It was suggested that the introduction of onsitePOC CD4 testing would increase the workload of healthpersonnel at clinical settings and this could be consid-ered the most prominent challenge in terms of feasibilityand acceptability in the context of no financial incentivesor additional staffing could be provided [24, 28]. Train-ing for health care staff on POC CD4 testing was reportedas another factor that might influence acceptability andfeasibility. In addition to training of test operators, trainingfor health managers is also required. Senior health staff re-ported that it was not feasible for them to monitor test op-erators on a test on which they were not trained: “it’sdifficult for me to supervise something I don’t knowabout” and “it’s downgrading to received instructionfrom a junior” [24]. The internal quality control andperformance monitoring were also reported as criticalfactors to ensure on-going reliability and acceptabilityof POC CD4 testing in clinic settings where externalquality control is a challenge because of the remotenessof the sites [27, 30, 31].

DiscussionThe findings of our review show that there is relativelylimited data on acceptability and feasibility of POC CD4testing, with data only available for the Pima test; noother current commercially available POC CD4 tests[32] have published data on acceptability and feasibilityin field settings. For Pima, the available data consist-ently demonstrates that it is feasible to deploy fordecentralization of CD4 testing through differentmodels of service delivery. However, there are some is-sues related to the implementation of Pima that mightinfluence acceptability and feasibility of the test that re-main unanswered. From the patient’s perspective noqualitative data were available to capture the willing-ness of patients to choose a POC CD4 test over astandard flow cytometric CD4 test when offered, or thepatient’s belief in the accuracy or relevance of POCCD4 testing in identifying treatment eligibility or re-sponse to treatment. One study [22] reported 10 % ofpatients declined the offer of POC CD4 testing andchose standard CD4 testing at referral clinics but

Pham et al. BMC Health Services Research (2016) 16:343 Page 7 of 12

Table 2 Acceptability and feasibility of POC CD4 test

Author (s), Year Technologies Proportion of HIV patientsaccepted POC CD4 testwhen offered

Reported attributes of POC CD4test related to day-to-day fieldoperation

System Factors associated with/havingeffect on acceptability/feasibility ofPOC CD4 test

Locally specific context and operationalissues which affect the deploymentof POC CD4 test

Galiwango, 2014 [26] Pima Easy to use; enable same day,on-site immunologicalassessment and resultcommunication

In busy clinic, it requires 2–4machines with additionaltechnician to completepatient testing

van Rooyen, 2013 [21] Pima Highly acceptable at the timeof learning about HIV test result(96 % of identified HIV positiveindividuals accepted, testedand received POC CD4 countresult at a HBCT visit

Feasible to be conducted athomes, as part of home-basedHIV counseling and testingprogram, in a rural SouthAfrican setting

Mtapuri-Zinyowera, 2013 [24] Pima Relatively low throughput,frequent error codes andcartridge rejection beforeexpiration date. Increasedtechnical breakdown after 1year of operation at busy sites;major breakdowns includehardware and alignment andloss of camera focus25 % of users reported havingsome challenges after machineinstallation of which 67 % (ofcases) were resolved by theManufacturer and 33 % by users

Users reported training was usefuland relevant to day-to-day operation.Training for supervisor is needed tomonitor staff performance. Externalquality control was a challengebecause of remoteness of sitesChallenges with finger-pricksampling is also noted including thatit cannot be used for full bloodcount (required to determine typesof ARVs patient can take), high errorrate led to multiple finger-prickexposing patient to more pain

Staff workload was the mostprominent challengereported by users (multipletasks and increased workloadwithout compensation); taskshifting should be consideredgiven prospect of additionalstaff employment is low

Larson, 2012 [22] Pima When offered a rapid POCCD4 test in a routine mobileHCT setting, acceptance amongpatients is high (90 %); only32 /311 (10.3 %) patientsdeclined the offer of POC CD4

Glencross, 2012 [31] Pima Negative impact of (poor) capillaryblood sampling on POC CD4 testperformance: Capillary samplingdemands absolute diligence andstringency of sampling technique.Ongoing dedicated training as wellas implementation of systems formonitoring and evaluation of testingis strongly recommended

Thakar, 2012 [25] Pima Users expressed that PIMA wascompact and hence could fit inthe small space available at thecenters. It is battery operated,showed a battery backup of

Study participants preferred to givevenous blood sample because ofrequirement of blood collection forother investigations using venousblood and a fear of being subjected

Phamet

al.BMCHealth

ServicesResearch

(2016) 16:343 Page

8of

12

Table 2 Acceptability and feasibility of POC CD4 test (Continued)

3–4 h eliminating requirementof continuous electricity

to multiple pricks if sufficient volumeof blood is not obtained in a singleprick

Manabe, 2012 [27] Pima Easy-to-use, portable, relativelyfast device to test CD4+ T cellcounts in the field

Quality control and observedpractical training for test operatorswould be required to ensure thatgood volume and flow of blood(capillary) is obtained

Jani, 2011 [28] Pima Essential WHO-recommendedART staging and monitoringdiagnostic tests can be accuratelyconducted at primary healthcare clinic level by non-laboratorystaff using POC CD4 test

Operators should be trained forfinger prick testing and theirperformance should be regularlymonitored as training and monitoringhas been shown to be essential tothe ongoing reliability of other POCCD4 test

Implementation of POC CD4in primary health care clinicsrequires careful planning. Taskshifting of ART services tocommunity clinics placesadditional strain on theworkloads of nurses and otherhealthcare workers that maybe unsustainable

Mtapuri-Zinyowera, 2010 [23] Pima The offer of POC CD4 testingwithin post-test counselingwas accepted by almost alleligible clients, even withinthe context of a study andthe need to provideinformed consent.

POC CD4 testing can beperformed in non-laboratorysetting by non-laboratorytechnicians (nurses)

It is important to ensure that Pimatest operators are well trained onfinger-prick sample collection.Preliminary observations in this studysuggest that incorrect finger-pricksampling affects the reliability of POCCD4 results

Wade, 2014 [30] Pima Significant contribution of operatorsto variability of POC CD4 test results:dedicated training for test operators,particularly on capillary bloodsampling is required to ensure qualityof POC CD4

Mwau, 2014 [7] MyT4 Relatively high throughput:over 20 tests/6 h health facilityworking day

Implementation would bemost effective by assigning adedicated full time operator

Arnet N, 2013 [29] Pima 100 % (11/11) HCW interviewedtrust Pima venous CD4 results;91 % (10/11) for Pima Microtubeand 82 % (9/11) for Pima direct.The most preferred samplecollection method was Pimavenous 73 % (8/11)

HBCT home-based HIV counseling and testing

Phamet

al.BMCHealth

ServicesResearch

(2016) 16:343 Page

9of

12

reasons for their preference was not studied. From thehealth service provider’s perspective, a number of issuesneed to be addressed in terms of acceptability and feasi-bility to provide insights into future implementation ofthe test. Although the device was reported as beingeasy-to-use, a reported high error rate and frequent testoperator errors raised questions about the healthworkers’ ability to correctly perform the test in theirdaily working environment. Information on the trainingand supervision of test operators that would allow as-sessment of their capacity to follow testing procedureand perform the test correctly is lacking. The prefer-ence of health workers toward venous blood collectionfor CD4 testing also needs further studies to identify ifthis preference is due to the health workers’ belief inthe validity of the test, difficulties involved with capil-lary blood sampling or other logistic issues related toblood testing in field settings. Lastly, human resourcesshortage represents a major challenge. The introduc-tion of POC CD4 testing at primary healthcare levelhas been reported to be associated with increasedwork-load for health staff and while the provision ofthis service may well response to patients’ needs andimprove quality of HIV treatment and care, the ques-tion remains is how to provide effective and sustainablePOC CD4 testing service in busy primary clinic settingswithout overburdening the already strained healthcareproviders. All these are important issues that need tobe studied and factored into the planning process toguide future implementation of the test in remote anddisadvantaged areas.Although data was only available for Pima POC CD4

testing, some lessons could be learned for future plan-ning and implementation of POC CD4 testing in LMICs.First, like other rapid diagnostic tests the validity or ac-curacy of POC CD4 test as perceived by the end users(health care provider and patient), will affect acceptanceand use of the test in the field [20]. As accuracy and effi-cacy of the test are influenced by the test operators’practical skills, the importance of quality training onPOC CD4 testing for health professionals, both test op-erator and supervisor, must be recognized and infra-structure to support this should be adequately addressedbefore introduction and scale up of any POC CD4 tech-nology [33]. This remains true for all POC tests thatcould be considered “quick” and “easy” but often requirecomprehensive training and supervision to ensure diag-nostic test accuracy under field conditions [34, 35].Without standardized quality training packages deliveredto health workers in these clinics, ensuring competencystandards are obtained by all test operators and their su-pervisors, and without on-going internal quality assur-ance systems in place it is impossible to rule out anypotential operator-induced bias. Second, the preference

toward a specific type of blood sample may also influ-ence the acceptability and feasibility of a POC CD4 test.This preference may come from either the health serviceprovider’s or patient’s views including challenges withone type of blood sampling, the need for a large amountof blood for other blood tests, or the health worker’s be-lief in accuracy of the test with one type of blood overthe other. For POC CD4 tests that can be used with ei-ther type of blood sample, this would require furtherstudy to identify technical and programmatic solutionsfor extending test applicability in the field. Third, POCtechnologies are designed to be user friendly and be eas-ily operated by low-to-middle level healthcare cadres. Inaddition to the importance of training for health staff,discussed above, their workload upon introduction ofPOC CD4 testing is an important issue to consider. Thereality is that in some of the highest volume clinics inlow-resource settings, it is the nurses, midwives andcounselors who have the highest workload. Therefore, itis recommended that patient and staff work flow shouldbe thoroughly studied and work load issues appropri-ately addressed through additional incentives, staffing ortask shifting to ensure the effectiveness, efficacy and sus-tainability of POC technology when introduced intobusy clinical settings.The future role and impact of CD4 testing in general

and POC CD4 testing in particular need to be consid-ered from both technical and programmatic perspec-tives. In order to achieve the newly proposed sustainabledevelopment goal of ending the HIV/AIDS epidemic by2030 [36], and with evidence showing positive clinicalimpact of ART for patients with CD4 count > 500 cells/μl [37], a policy shift toward recommendation of ARTinitiation for all HIV infected individuals regardless ofCD4 count has recently been announced by WHO [38].However, scaling up of ART programs in low-resourcesettings may risk low adherence and retention to carerates if critical health system factors such as well-trainedhealth staff, well-functioning patient monitoring and ap-propriate support systems are not in place [39]. A suc-cessfully scaled up treatment program, therefore, willrequire innovative and effective models of service deliv-ery (such as integration and decentralization of care),strong procurement and supply chain management, suf-ficient laboratory and/or point-of-care diagnostic ser-vices to monitor clinical treatment outcomes, HIV viralload and drug resistance and antiretroviral drug toxic-ities. Most of these systems are not currently in place inthose LMICs mostly affected by the HIV epidemic andcould only be achieved through a gradual process ofhealth system strengthening. Therefore, in countrieswhere treatment for all is not currently feasible with theavailable resources and current health system capacity,CD4 testing is required to prioritize treatment eligibility

Pham et al. BMC Health Services Research (2016) 16:343 Page 10 of 12

likely for those patients whose CD4 count is less than350 cells/μl. Going forward, CD4 testing will still play animportant role in identifying patients who present late tocare for clinical management, for initiation and cessationof prophylaxis and for management of patient responsesto ART even in countries where CD4-independent ARTinitiation is being introduced.This review has some limitations that should be con-

sidered in interpretation of the findings. First, assess-ment of acceptability and feasibility was not theprimary objective of any of the included studies and ina number of studies POC CD4 testing was only a partof a more comprehensive program intervention. Thiscarries a risk of introducing bias as impact of the POCCD4 test and its acceptability and feasibility could beinfluenced by other interventions or implementationstrategies. Second, we included only published studiesin English and this inclusion may overlook data fromstudies published in other languages or unpublisheddata from evaluations conducted by government agen-cies, local reference facilities or research institutions.All but one of the included studies reporting outcomesof interest use Pima as the index test and this presentschallenges in terms of generalizing the findings of thereview to other POC CD4 tests, as specific technicaland operational characteristics of each test will signifi-cantly impact feasibility and acceptability of the testfrom both service provider and patient perspectives. Ofnote, these limitations cannot be overcome until moredata from field studies of different POC CD4 technolo-gies are available.

ConclusionsData is available for one of the POC CD4 tests on themarket, the Alere Pima CD4, and suggests that it couldbe feasible to implement point-of-care CD4 testing innon-laboratory settings in low and middle income coun-tries. Further studies using other currently or newlyavailable POC CD4 tests in different geographical re-gions are needed to inform in-country decision makingregarding the selection and adoption of a suitable test.Qualitative studies on feasibility and acceptability areneeded to explore end users’ beliefs on the value of andpreference toward POC CD4 testing. Evidence regardingsupporting system factors such as training, monitoring,supplies, and facility requirements should be availableand inform the planning process for the introductionand scaling up of POC CD4 tests at primary health carelevels in low-resource settings.

Additional file

Additional file 1: Medline search strategy (DOC 50 kb)

AbbreviationsART, Antiretroviral therapy; HBCT, home based counseling and testing; LMICs,Low and middle income countries; LTFU, Loss to follow up; MNCH, Maternaland new born child health; PICO, participants, interventions, compactors/controls, outcomes; POC, point-of-care; PRISMA, Preferred reporting items forsystematic review and meta-analysis

AcknowledgmentsThe authors gratefully acknowledge the contribution to this work of the VictorianOperational Infrastructure Support Program received by the Burnet Institute.

FundingFunding was provided by the National Health and Medical Research Councilof Australia (NHMRC) (Project grant GNT 1063725 and Career DevelopmentFellowship to Stanley Luchters; Principle Research fellowship to SuzanneCrowe). Minh D. Pham received support via an International PostgraduateResearch Scholarship (IPRS) from the Commonwealth of Australia and theVictorian International Research Scholarship (VIRS) from State Government ofVictoria, Australia. The donor had no involvement in the design of the study;collection, analysis, interpretation of data, or the writing and submission ofthis manuscript.

Availability of data and materialsAll data generated or analyzed during this study are included in thispublished article and its supplementary information file.

Authors’ contributionsMP, SL, DA, SC developed review protocol. LR, MP performed literature search.MP, PM, PA performed data extraction and data analysis. MP drafted themanuscript. LR, PM, PA, DA, SC, SL reviewed and commented on initial and finaldrafts of the manuscript. All authors read and approved the final manuscript.

Competing interestsThe authors declare that they have no competing interests.

Consent for publicationNot applicable.

Ethics approval and consent to participateNot applicable.

Author details1Burnet Institute, Melbourne, VIC, Australia. 2Department of Epidemiologyand Preventive Medicine, Faculty of Medicine Nursing and Health Science,Monash University, Melbourne, Australia. 3The Alfred Hospital, The Ian PotterLibrary, Melbourne, VIC, Australia. 4Department of Immunology, Faculty ofMedicine Nursing and Health Science, Monash University, Melbourne,Australia. 5Department of Infectious Diseases, The Alfred hospital Melbourne,Melbourne, Australia. 6Monash School of Medicine, Faculty of MedicineNursing and Health Science, Monash University, Melbourne, Australia.7International Centre for Reproductive Health, Department of Obstetrics andGynecology, Faculty of Medicine and Health Sciences, Ghent University,Ghent, Belgium.

Received: 6 November 2015 Accepted: 27 July 2016

References1. INSIGHTSTART. Initiation of Antiretroviral Therapy in Early Asymptomatic HIV

Infection. N Engl J Med. 2015;373:795–807.2. Ford N, et al. The future role of CD4 cell count for monitoring antiretroviral

therapy. Lancet Infect Dis. 2015;15(2):241–7.3. Zachariah R, et al. Viewpoint: Why do we need a point-of-care CD4 test for

low-income countries? Trop Med Int Health. 2011;16(1):37–41.4. Govindasamy D, Ford N, Kranzer K. Risk factors, barriers and facilitators for linkage

to antiretroviral therapy care: a systematic review. AIDS. 2012;26(16):2059–67.5. Reid S, Fidler S, Cooke G. Tracking the progress of HIV: The impact of point-

of-care tests on antiretroviral therapy. Clin Epidemiol. 2013;5(1):387–96.6. Rathunde L, et al. Evaluation of the Alere Pima™ for CD4+ T lymphocytes

counts in HIV-positive outpatients in Southern Brazi. Int J STD AIDS. 2014;25(13):956–9.

Pham et al. BMC Health Services Research (2016) 16:343 Page 11 of 12

7. Mwau M, et al. Technical Performance Evaluation of the MyT4 Point of CareTechnology for CD4+ T Cell Enumeration. PLoS ONE. 2014;9(9), e107410.

8. Scott LE, et al. A meta-analysis of the performance of the Pima (TM) CD4 forpoint of care testing. BMC Med. 2015;13:168.

9. Wynberg E, et al. Impact of point-of-care CD4 testing on linkage to HIVcare: a systematic review. J Int AIDS Soc. 2014;17:18809.

10. Patten GE, et al. Impact on ART initiation of point-of-care CD4 testing at HIVdiagnosis among HIV-positive youth in Khayelitsha, South Africa. J Int AIDSSoc. 2013;16:18518.

11. Jani IV, et al. Effect of point-of-care CD4 cell count tests on retention ofpatients and rates of antiretroviral therapy initiation in primary health clinics:an observational cohort study. Lancet. 2011;378(9802):1572–9.

12. Black S, et al. Acceptability and challenges of rapid ART initiation amongpregnant women in a pilot programme, Cape Town, South Africa. AIDSCare. 2014;26(6):736–41.

13. Myer L, et al. Pilot programme for the rapid initiation of antiretroviral therapyin pregnancy in Cape Town, South Africa. AIDS Care. 2012;24(8):986–92.

14. Faal M, et al. Providing immediate CD4 count results at HIV testingimproves ART initiation. J Acquir Immune Defic Syndr. 2011;58(3):e54–9.

15. Counsell C. Formulating questions and locating primary studies for inclusionin systematic reviews. Ann Intern Med. 1997;127(5):380–7.

16. UNITAID. 2014 HIV/AIDS Diagnostics Technology Landscape. 4th ed. 2014.http://www.unitaid.eu/en/resources/publications/technical-reports.

17. Whiting P, et al. The development of QUADAS: a tool for the qualityassessment of studies of diagnostic accuracy included in systematic reviews.BMC Med Res Methodol. 2003;3:25.

18. Jackson N, et al. Criteria for the systematic review of health promotion andpublic health interventions. Health Promot Int. 2005;20(4):367–74.

19. Walsh D, Downe S. Appraising the quality of qualitative research. Midwifery.2006;22(2):108–19.

20. Asiimwe C, et al. Early experiences on the feasibility, acceptability, and useof malaria rapid diagnostic tests at peripheral health centres in Uganda-insights into some barriers and facilitators. Implement Sci. 2012;7:5.

21. Van Rooyen H, et al. High HIV testing uptake and linkage to care in a novelprogram of home-based HIV counseling and testing with facilitated referral inKwaZulu-Natal, South Africa. J Acquir Immune Defic Syndr. 2013;64(1):e1–8.

22. Larson BA, et al. Rapid point-of-care CD4 testing at mobile HIV testing sitesto increase linkage to care: an evaluation of a pilot program in South Africa.J Acquir Immune Defic Syndr. 2012;61(2):e13–7.

23. Mtapuri-Zinyowera S, et al. Evaluation of the PIMA point-of-care CD4 analyzerin VCT clinics in Zimbabwe. J Acquir Immune Defic Syndr. 2010;55(1):1–7.

24. Mtapuri-Zinyowera S, et al. PIMA Point of Care CD4+ Cell Count Machinesin Remote MNCH Settings: Lessons Learned from Seven Districts inZimbabwe. Infect Dis (Auckl). 2013;6:51–60.

25. Thakar M, et al. Utility of the point of care CD4 analyzer, PIMA, to enumerateCD4 counts in the field settings in India. AIDS Res Ther. 2012;9(1):26.

26. Galiwango RM, et al. Field evaluation of PIMA point-of-care CD4 testing inRakai, Uganda. PLoS ONE. 2014;9(3), e88928.

27. Manabe YC, et al. Evaluation of portable point-of-care CD4 counter withhigh sensitivity for detecting patients eligible for antiretroviral therapy. PLoSONE. 2012;7(4), e34319.

28. Jani IV, et al. Accurate CD4 T-cell enumeration and antiretroviral drugtoxicity monitoring in primary healthcare clinics using point-of-care testing.AIDS. 2011;25(6):807–12.

29. Arnett N, Chang K, Schmitz M, Lemwayi R, Rwehumbiza P, Mwasekaga M.Healthcare workers’ acceptance and performance of point-of-care CD4testing in Dar es Salaam, Tanzania, 2011. Kuala Lumpur: 7th IAS Conferenceon HIV Pathogenesis and Treatment; 2013.

30. Wade D, et al. WHO multicenter evaluation of FACSCount CD4 and PimaCD4 T-cell count systems: instrument performance and misclassification ofHIV-infected patients. J Acquir Immune Defic Syndr. 2014;66(5):e98–e107.

31. Glencross DK, et al. Performance evaluation of the Pima™ point-of-care CD4analyser using capillary blood sampling in field tests in South Africa. J IntAIDS Soc. 2012;15(1):3.

32. UNAIDS. HIV/AIDS Diagnostics technologies Landscape Semi-AnnualUpdate. 2015. Available at: http://www.aidsdatahub.org/hivaids-diagnostics-technology-landscape-semi-annual-update-murtagh-mm-2015.

33. Fajardo E, et al. Errors generated by a point-of-care CD4+ T-lymphocyteanalyser: a retrospective observational study in nine countries. Bull WorldHealth Organ. 2015;93:623–30.

34. Seidahmed OM, et al. End-user errors in applying two malaria rapid diagnostictests in a remote area of Sudan. Trop Med Int Health. 2008;13(3):406–9.

35. Wolpaw BJ, et al. The failure of routine rapid HIV testing: a case study ofimproving low sensitivity in the field. BMC Health Serv Res. 2010;10:73.

36. Nations, U. Final draft of the outcome document for the UN Summit toadopt the Post-2015 Development Agenda. 2015.

37. TEMPRANO. A Trial of Early Antiretrovirals and Isoniazid Preventive Therapyin Africa. N Engl J Med. 2015;373(9):795–807.

38. WHO. GUIDELINE ON WHEN TO START ANTIRETROVIRAL THERAPY AND ONPRE-EXPOSURE PROPHYLAXIS FOR HIV. Geneva: World Health Organization;2015. p. 78.

39. Unge C, et al. Challenges for scaling up ART in a resource-limited setting: aretrospective study in Kibera, Kenya. J Acquir Immune Defic Syndr. 2009;50(4):397–402.

• 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 research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:

Pham et al. BMC Health Services Research (2016) 16:343 Page 12 of 12