SEARCH

SEARCH BY CITATION

Keywords:

  • human immunodeficiency virus;
  • pregnant women;
  • prevention of mother to child transmission;
  • diagnostic tests – routine;
  • infection transmission – vertical
  • VIH;
  • femmes enceintes;
  • PTME;
  • tests de diagnostic;
  • routine;
  • transmission de l’infection;
  • vertical
  • VIH;
  • mujeres embarazadas;
  • prevención de la transmisión vertical;
  • prueba diagnóstica;
  • rutina transmisión infección;
  • vertical

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objective  To assess the contribution of provider-initiated testing and counselling (PITC) to achieving universal testing of pregnant women and, from available data on components of PITC, assess whether PITC adoption adheres to pre-test information, post-test counselling procedures and linkage to treatment.

Methods  Systematic review of published literature. Findings were collated and data extracted on HIV testing uptake before and after the adoption of a PITC model. Data on pre- and post-test counselling uptake and linkage to anti-retrovirals, where available, were also extracted.

Results  Ten eligible studies were identified. Pre-intervention testing uptake ranged from 5.5% to 78.7%. Following PITC introduction, testing uptake increased by a range of 9.9% to 65.6%, with testing uptake ≥85% in eight studies. Where reported, pre-test information was provided to between 91.5% and 100% and post-test counselling to between 82% and 99.8% of pregnant women. Linkage to ARVs for prevention of mother to child transmission (PMTCT) was reported in five studies and ranged from 53.7% to 77.2%. Where reported, PITC was considered acceptable by ANC attendees.

Conclusion  Our review provides evidence that the adoption of PITC within ANC can facilitate progress towards universal voluntary testing of pregnant women. This is necessary to increase the coverage of PMTCT services and facilitate access to treatment and prevention interventions. We found some evidence that PITC adoption does not undermine processes inherent to good conduct of testing, with high levels of pre-test information and post-test counselling, and two studies suggesting that PITC is acceptable to ANC attendees.

Objectif:  Evaluer la contribution du dépistage et conseil initié par le prestataire (DCIP) au succès du dépistage universel des femmes enceintes et, à partir des données disponibles sur les composants du DCIP, évaluer si l’adoption du DCIP adhère aux recommandations pour les renseignements à fournir avant le test, aux procédures de conseils post-test et aux liens avec le traitement.

Méthodes:  Revue systématique de la littérature publiée. Les critères d’inclusion comprenaient: le report de l’effet de l’offre systématique du dépistage du VIH au niveau du service sur la prise du test; les études menées dans le cadre des services de soins prénataux, l’adoption d’un concept d’étude randomisée et contrôlée ou non randomisée et contrôlée effectuées dans plus d’un site dans chaque bras ou les études de séries chronologiques menées dans un seul site avec des données pré/post-intervention, les études publiées dans la langue anglaise.

Résultats:  Dix études admissibles ont été identifiées. La prise du test pré-intervention variait de 5,5%à 76,6%. Après l’introduction du DCIP, la prise du test a augmenté dans une gamme de 13%à 65,6%, avec une prise du test ≥ 85% dans huit études. Lorsque rapportés, les renseignements pré-test ont été fournis à entre 91,5% et 100% des femmes enceintes et post-test à entre 82% et 99,8%. Les liens avec l’ARV pour la PTME ont été signalés dans cinq études et variaient de 53,7%à 77,2%. Lorsque rapporté, le DCIP a été jugé acceptable par les visiteuses des services de soins prénataux.

Conclusion:  Notre étude fournit la preuve que l’adoption du DCIP au sein des services de soins prénataux peut faciliter les progrès vers le dépistage volontaire universel des femmes enceintes. Cela est nécessaire pour augmenter la couverture des services de PTME et faciliter l’accès aux traitements et aux interventions de prévention. Nous avons trouvé des preuves que l’adoption du DCIP ne sape pas les processus inhérents à la bonne conduite des tests, avec des taux élevés de renseignements fournis en pré-test et du conseil post-test, et deux études suggérant que le DCIP est acceptable pour les femmes fréquentant les services de soins prénataux.

Objetivo:  Evaluar la contribución del aconsejamiento y prueba iniciado por el proveedor (APIP), para conseguir la realización universal de la prueba en mujeres embarazadas, utilizando los datos disponibles sobre componentes del APIP, y evaluar si la adopción del APIP se adhiere a la información pre-prueba, a los procedimientos de aconsejamiento post-prueba y va ligado al tratamiento.

Métodos:  Revisión sistemática de literatura publicada. Los criterios de inclusión incluían: el efecto reportado sobre la aceptación de la prueba del VIH, de un ofrecimiento rutinario de la misma a nivel del centro; el haberse realizado en centros de cuidados prenatales; con un diseño aleatorizado-controlado o no-aleatorizado-controlado con >1 centro en cada brazo o en donde se realizaron estudios con series temporales en un centro con datos pre-/post intervención; estudios publicados en inglés.

Resultados:  Se identificaron diez estudios que cumplían los criterios de inclusión. Las pruebas anteriores a la intervención estaban entre el 5.5% y el 76.6%. Tras la introducción del APIP, la aceptación de la prueba aumentó en un rango del 13% al 65.6%, con la realización de la prueba en ≥85% en ocho de los estudios. Cuando se reportaba, se dio información anterior a la prueba a cerca del 91.5% - 100% y aconsejamiento post-prueba a entre 82% y 99.8% de las mujeres embarazadas. La conexión con TAR para la prevención de la transmisión vertical se reportó en cinco estudios y estaba en un rango de entre el 53.7% y el 77.2%. En aquellos estudios en donde se reportaba, el qAPIP era considerado como aceptable por quienes se visitaban en las clínicas prenatales.

Conclusión:  Nuestra revisión aporta evidencia de que la adopción del APIP dentro de los cuidados prenatales puede facilitar el progreso hacia una realización voluntaria y universal de la prueba para el VIH entre mujeres embarazadas. Esto es necesario para aumentar la cobertura de los servicios de prevención de la transmisión vertical y facilitar el acceso a intervenciones de tratamiento y prevención. Hemos hallado evidencia de que la adopción del APIP no disminuye procesos inherentes a la buena conducción de la prueba, con altos niveles de información antes de la prueba y de aconsejamiento posterior a la misma y dos estudios en los que se sugiere que el APIP es aceptable para aquellos que visitan una clínica de cuidados prenatales.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

HIV testing is a ‘critical gateway’ to treatment, care and support services. Knowledge of HIV status can empower individuals and couples to take measures to prevent HIV acquisition or onward transmission. For those already infected, a positive test result is necessary to access treatment and, in the case of pregnant women, to access prevention of mother to child transmission (PMTCT) services (Higgins et al. 1991, Coates et al. 2000; WHO 2003). Across communities, normalising awareness of HIV status through increasing testing could reduce HIV-related stigma and discrimination (WHO 2003).

However, progress towards universal knowledge of HIV status is inadequate (Granich et al. 2009). In 10 population-based surveys conducted in sub-Saharan Africa in 2007–2009, the median percentage of people living with HIV who knew their status was <40% (WHO 2010). In Africa, the estimated percentage of people who know their HIV status ranges from <10% in Sierra Leone, DRC and Liberia to 40–56% in Kenya and South Africa (WHO 2010). In 2009, an estimated 50% of pregnant women in Eastern and Southern Africa received an HIV test, up from 43% in 2008(WHO 2010).

For many years, HIV testing was delivered through a voluntary counselling and testing (VCT) model. A client-initiated approach, VCT guidelines stressed the importance of confidentiality, expressed voluntarism and written informed consent. VCT relied on personal motivation to seek testing, which is influenced by a number of factors that might act as barriers to widespread testing (Leon et al. 2010b). With <10% of key populations in low- and middle-income countries who may have been exposed to HIV accessing VCT and with evidence that many opportunities to diagnose patients with HIV were lost (Seipone 2004; Fetene & Feleke 2010), VCT was increasingly seen as inadequate (Nieburg et al. 2005; Fetene & Feleke 2010).

To address this problem in 2007 WHO issued guidance recommending the routine offer of HIV testing in clinical settings. Termed ‘provider-initiated testing and counselling’ (PITC), this model supports recommending testing to all patients attending health facilities, including antenatal services, in countries with generalised epidemics and for those attending, for example, antenatal, tuberculosis and sexually transmitted infection services in non-generalised epidemiological contexts (WHO 2007). PITC eliminates the need for lengthy pre-test counselling, replacing it with pre-test information that meets the minimum standard for informed consent (WHO 2007). Opt-out PITC is recommended; however, the guidance suggests that opt-in PITC should be considered for ‘highly vulnerable populations’ (WHO 2007). PITC proponents highlight its potential impact on universal knowledge of HIV status, thereby its potential to reduce mother-to-child transmission (MTCT), HIV-related morbidity and mortality (Bayer & Edington 2009), and its capacity to ‘normalise’ HIV testing (Leon et al. 2010a). However, others have concerns that removing extensive pre-test counselling and written informed consent could lead to coercion and thus threatens the ethics and human rights of HIV testing (Gruskin et al. 2008; Bayer & Edington 2009). Prior to the WHO guidance, numerous countries had adopted the models of PITC in clinical settings, particularly in antenatal clinics (ANC) in response to evidence of the effectiveness of antiretrovirals (ARV) in reducing MTCT and the urgent need to provide access to PMTCT services to pregnant women with HIV (Seipone 2004; Centers for Disease Control & Prevention 2008).

Growing evidence suggests that adopting PITC in ANC reduces lost opportunities to test pregnant women for HIV and increases PMTCT uptake and coverage (Kharsany et al. 2010, Amornwichet et al. 2002, Weigel et al. 2009; Chersich & Luchters 2008). The aim of this systematic review was to determine the extent to which the adoption of guidelines for a PITC-related model in ANC contributes to achieving the goal of universal voluntary testing of pregnant women and increases the coverage of PMTCT, treatment and care services. We also aimed to assess whether PITC adheres to the standards inherent to good conduct of HIV testing, namely that it provides pre-test information and post-test counselling.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We conducted a systematic review of published literature. We searched Medline, Embase and Global Health using the following search terms: provider-initiated, routine, diagnostic, opt-out, HIV, human immunodeficiency virus, AIDS, acquired immune deficiency syndrome, counselling, counseling, testing and screening, antenatal, PMTCT, vertical transmission and prenatal care. The final full search was conducted in December 2010. BH conducted the literature search and, with support from DB and MK, determined whether full references extracted were eligible for inclusion. Studies were considered eligible for inclusion if they were conducted in ANC settings and reported on the impact of adopting a model of PITC on the primary outcome: HIV testing uptake, and were English language studies. We included randomised controlled trials (RCT) and non-randomised studies (NRS) with >1 site per arm as long as they collected both pre- and post-intervention data for each facility. We also included time series studies conducted in one facility if they had collected pre- and post-intervention data. We excluded studies conducted at population level as they provide limited information of changes within individual facilities, thus imposing greater challenges in inferring causation. Our review identified three types of intervention/comparison conditions that satisfied these criteria (Table 1).

Table 1. Comparison and intervention conditions considered eligible for inclusion
 Comparison HIV testing modelIntervention HIV testing model
  1. *These models were defined as those where the emphasis of consent changed from one where, following a routine offer of testing, a patient explicity opted-in to a testing model where the patient had to opt-out should they choose not to test.

AVCTRoutine offer of testing
BDiagnostic testingRoutine offer of testing
CRoutine offer of opt-in testingRoutine offer of opt-out testing*

Data extraction

BH extracted data into two data extraction forms, one for study design (Table 2) and one for data relating to HIV testing uptake and adjusted odds ratios, where reported. We also collated data on counselling and linkage to ARV. We critically appraised the quality of studies using a quality assessment tool adapted from the Effective Public Health Practice Project (EPHPP1). The tool included 22 questions that allowed for yes, no and cannot tell responses. Studies were not excluded on the basis of quality, but given a weak, moderate or strong quality rating (weak n = 2, moderate n = 7 and strong n = 1).

Table 2. Details of studies included in the review
SettingYear of studyStudy designStudy of a policy change or PITC training interventionComparison groupPeriod of follow-upFacility settingNum. of facilitiesPopulationPopulation sizeUrban/RuralOutcome(s)
  1. *Unpublished results from conference abstracts and presentations provided details of the study, including number of women reached, for the period 2004–2005. Results only include data from these years.

  2. †Although 2004–2009 results are available, only 2007–2009 are included in the review as in 2007 private facilities, which are included in 2009 data, started PMTCT.

  3. ‡Results only presented for period March–December 2005, when testing was opt-in, with January–December 2006, before the implementation of free services.

  4. RCT, randomised controlled trial; BA, before/after study; RBA, retrospective (before/after) study; TS, time series

Scotland (Simpson et al. 1998)1996–1997RCTRCT comparing VCT to 4 models of routine testingRandomly selected VCT control10 monthsHospital ANC1Pregnant women3024UUptake and acceptability of HIV testing, women’s knowledge, satisfaction and anxiety
Scotland (Simpson et al. 1999)1998BAPolicy change10 months prior to opt-out testing3 months  Pregnant women2954UUptake and acceptability of HIV testing, testing-related anxiety
USA (Stringer et al. 2001)1998–2000BATraining & policy changeData from year before policy change2 yearsANC8Pregnant women7193UUptake of opt-out testing on HIV testing rates
Kenya (van’t Hoog et al. 2005) 2001–2003RBAImplementation of pilot projectYear prior to pilot project2 yearsHospital ANC1 (6 public health centres)Pregnant women8231UUptake of counselling, testing and nevirapine
Botswana (Creek et al. 2007)2003–2004BATraining & policy changeLast 5 months of VCT1 yearANC4Pregnant women1456UHIV testing uptake, PMTCT intervention rates, and rates of ANC attendance
Zimbabwe (Chandisarewa et al. 2007) 2004–2005BATraining & policy changeLast 6 months of opt-in1 yearANC4Pregnant women & their partners9423UHIV testing uptake, maternal HIV status; perceptions of routine testing
Malawi (Moses et al. 2008, Zimba et al. 2006*)2002–2006TSPolicy changeLast 9 months of opt-in18 monthsANC4Pregnant women30 092URate HIV counselled, HIV testing uptake and HIV positivity
Uganda (Byamugisha et al. 2010) 2002–2009TSTraining & Policy changeHistorical control – hospital records pre/post policy change7 yearsReferral Hospital1Pregnant women & their partners54 429UNumber of new ANC attendances/year, number of ANC patients counselled and tested, number that obtained a positive result, number of male partners counselled and tested and obtained positive result, HIV-positive mothers and infants that received ART prophylaxis
Ethiopia (Mirkuzie et al. 2010)2004–2009TSPolicy changeHistorical control – PMTCT reports5 years†Public and private facilities52 (25 private)Pregnant women and their partners663 603 (from 2004 to 2009)UProportion of women receiving pre-test counselling, HIV testing and post-test counselling; proportion of women and babies who received ARV prophylaxis
Malawi (Kasenga et al. 2009)‡2005–2007TSPolicy changeHistorical control – hospital records pre/post policy change2 yearsHospital ANC1Pregnant women2055RDemand for ANC, HIV testing uptake and hospital delivery

Defining PITC and lost opportunities

For the purpose of this review, we defined PITC as the routine offer of HIV testing, regardless of whether the emphasis of consent was opt-in or opt-out, by healthcare providers to pregnant women attending ANC settings as a standard component of clinical practice. Lost opportunities to test for HIV were defined as the difference between the reported proportion of women accepting testing and universal (100%) testing uptake. Although achieving 100% uptake of testing is an ambitious goal, universal testing is defined as such – the universal uptake of HIV testing to achieve universal knowledge of HIV status. We expected studies to report increased uptake of HIV testing after implementation of PITC; however, our primary research question was to determine the ability of PITC to achieve universal (100%; 0% lost opportunities) HIV testing in ANC. We restricted this review to studies conducted in ANC settings as the majority of studies on PITC were conducted in ANC and the majority of PITC programmes were initially implemented in ANC.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We retrieved 1009 references, of which 831 were excluded following title review. Of the remaining 178, a further 158 were excluded following abstract review or retrieval of the full paper as they did not meet the inclusion criteria (Figure 1). Of the remaining 20 studies, ten targeted the population of interest (Table 2).

image

Figure 1.  Inclusion flow diagram.

Download figure to PowerPoint

Studies that met the inclusion criteria

Tables 2 and 3 present the details and characteristics of the included studies. The majority were NRS (90%) comparing the period before the introduction of a routine testing guideline, implemented either through an intervention at specific clinics or through a national policy change, to the period after. We identified only one RCT (Simpson et al. 1998). In the studies, the number of clinics varied from 1 to 52 and sample sizes varied from 1456 to 54 429 women. Seven studies were conducted in Africa (van’t Hoog et al. 2005; Chandisarewa et al. 2007; Creek et al. 2007; Moses et al. 2008; Kasenga et al. 2009; Byamugisha et al. 2010; Mirkuzie et al. 2010), two in Europe (Simpson et al. 1998, 1999) and one in the USA (Stringer et al. 2001). The outcome of HIV testing uptake was reported in all studies (Table 2).

Table 3. Study characteristics
Study characteristicsFrequency (n)
  1. *For the study conducted in Ethiopia, public facilities adopted opt-out testing but private facilities adopted opt-in testing after VCT.

Study design
 Non-randomised study9
 Randomised controlled trial 1
Year of study initiation
 1995–19961
 1997–19982
 1999–20000
 2001–20022
 2003–20044
 2005–20061
Study direction
 Prospective5
 Retrospective5
Inclusion and exclusion criteria (individuals)
 Reported1
 Not reported9
Reason for selection of clinic(s)
 Reported1
 Not reported9
Results assessors blinded during interpretation
 Reported0
 Not reported10
Intervention group
 Routine offer of testing vs. VCT4
 Routine offer of opt-out testing vs. routine offer opt-in testing 6*
Comparison group
 Before intervention comparison6
 Historical comparison3
 Controlled comparison 1
Study typology (Table 1)
 Group A4
 Group B0
 Group C6

Intervention heterogeneity

The PITC interventions under study varied in a number of ways. Some studies reported extensive healthcare worker training prior to the implementation of PITC (Chandisarewa et al. 2007; Creek et al. 2007; Byamugisha et al. 2010), whilst others did not report on this (Kasenga et al. 2009). Some studies were the evaluations of interventions implemented in facilities, whilst others evaluated the influence of policy change on activities at facility level (Table 2). The information provided to women through pre-test education also varied: one study reported extensive community mobilisation (Chandisarewa et al. 2007), whilst two reported provision of leaflets (Simpson et al. 1998, 1999) (Table 4).

Table 4. Processes of interventions of included studies
Country, year of studyIntervention description (Panel 1)Provision of training for providersInformation on HIV/HIV testingOffer of testingPre-test counsellingDescription of diagnostics
Scotland, 1996–1997 (Simpson et al. 1998)ATraining in offering of testTesting advertised by posters and a letter, women in intervention groups received a leaflet with information about all blood tests or only HIV testsRoutine offer of opt-in testing to the intervention groupIndividual minimal or comprehensive discussionNR
Scotland, 1998 (Simpson et al. 1999)CTraining in offering of testWomen received leaflets prior to ANC appointmentsRoutine offer of opt-out testing to all womenIndividual discussion of benefits of testing using standard protocolNR
USA, 1998–2000 (Stringer et al. 2001)CHIV update for all maternity providers – included information on benefits of HIV screening, current ART for adults, therapy for PMTCT and proposed policy changeWomen referred to written material providing information on HIVRoutine offer of opt-out testingWritten materials available along with one page counselling formNR
Kenya, 2001–2003 (van’t Hoog et al. 2005)ATraining in PMTCTNRRoutine offer of opt-out testingAll first time ANC attendees received individual pre-test counselling from nurse-counsellorRDT
Botswana, 2003–2004 (Creek et al. 2007)ATraining for counsellors on group/individual pre-test counselling with flipchartNRRoutine offer of opt-out testingGroup pre-test information and discussionElisa, return after 1 month; RDT, return 1 week, depending on gestation period
Zimbabwe, 2004–2005 (Chandisarewa et al. 2007)CTwo day training on data collection and interview techniquesCommunity mobilisation to create awareness: skit performed in clinics, colleges, churches and industrial facilitiesRoutine offer of opt-out testing to all women by PMTCT clinic counsellors15 min group educationRDT
Malawi, 2004–2005 (Moses et al. 2008)CNRHealth information & education provided in waiting areaRoutine offer of opt-out testingPre-test counselling in groups of 8–12 womenRDT
Uganda, 2002–2009 (Byamugisha et al. 2010)AOne week refresher training on HIV C&T updates plus 5 day training on RDTNRRoutine offer of opt-out testing as ‘standard of care’Group pre-test information with flip chart to facilitate discussionRDT
Ethiopia, 2004–2009 (Mirkuzie et al. 2010)CNRNRRoutine offer of opt-out testing in public facilities from 2008NRNR
Malawi, 2005–2007 (Kasenga et al. 2009)CNRNRRoutine offer of opt-out testingGroup pre-test counsellingNR

Missed opportunities for testing

Table 5 presents data on HIV testing uptake. Prior to the implementation of PITC, lost opportunities for testing ranged from 94.5% in Scotland (Simpson et al. 1998) to 21.3% in a study from Malawi (Kasenga et al. 2009). Following implementation of PITC, lost opportunities decreased substantially in all settings (Figure 2). Despite a significant decrease in lost opportunities, in Scotland, where the aim of the study was to determine uptake and acceptability of different methods of a universal offer of testing, the lost opportunities amongst women in intervention groups remained high at 65.2% (P < 0.001) (Simpson et al. 1998). The adoption of opt-out PITC further reduced missed opportunities to 11.7% (Simpson et al. 1999). In rural Malawi, implementation of opt-out testing reduced lost opportunities to test by 9.9%; the introduction of free maternity services in October 2006 reduced lost opportunities further to 1.2% (testing uptake 98.8% (n = 2,249/2,277);Kasenga et al. 2009). The lowest proportion of lost opportunities for testing was reported in Zimbabwe, where <0.1% of opportunities to test women were lost (P < 0.001) (Chandisarewa et al. 2007).

Table 5. HIV Testing Uptake
Country, (Author)HIV Testing Uptake Among Pregnant Women Presenting to ANC
PITC modelComparison groupDecrease (%) in lost opportunities to test
  1. *< 0.05.

  2. †OR – Odds ratio; adjusted for midwife, marital status (married vs being single), previous test, age.

  3. ‡OR – Odds Ratio; adjusted for age, education, ethnicity, smoking history, substance use and history of sexually transmitted infections.

  4. §Unpublished results from conference abstracts results: PITC Model: 98.3% (= 14 491/14 749)*; Comparison Model: 76.1% (= 11 689/15 343) (Zimba et al. 2006).

  5. ¶Private facilities offered opt-in testing yet >90% of women attended public facilities, where opt-out testing was offered (personal communication, 2011).

Scotland (Simpson et al. 1998), 1996–199734.8% (= 707/2030)* (OR offered testing vs control 8.4 (6.2, 11.5)†5.5% (= 55/994)29.3%
Scotland (Simpson et al. 1999), 199888.3% (= 816/924)*34.8% (= 707/2030)53.5%
USA (Stringer et al. 2001), 1998–200088%* (OR study year 2.3 (2.1, 2.7))‡75%13%
Kenya (van’t Hoog et al. 2005), 2001–200375.8% (= 3101/4089)*61.6% (= 2551/4142)14.2%
Botswana (Creek et al. 2007), 2003–200494.8% (= 914/964)*76.6% (= 377/492)18.2%
Zimbabwe (Chandisarewa et al. 2007), 2004–200599.9% (= 4547/4551)*62.8% (= 3058/4872)37.1%
Malawi (Moses et al. 2008), 2004–200599%*§73%26%
Uganda (Byamugisha et al. 2010), 2002–200987.6% (= 21 538/24 595)*22% (= 6570/29 834)65.6%
Ethiopia (Mirkuzie et al. 2010), 2004–200985%¶52.2%32.8%
Malawi (Kasenga et al. 2009), 2005–200788.6% (= 879/992)*78.7% (= 837/1063)9.9%
image

Figure 2.  Achieving universal testing among pregnant women in ANC. The contribution of expanded PITC interventions to universal testing compared with pre-intervention testing uptake.

Download figure to PowerPoint

Components of the PITC process

Four studies presented results for the proportion of women provided pre-test counselling/information (van’t Hoog et al. 2005; Chandisarewa et al. 2007; Byamugisha et al. 2010; Mirkuzie et al. 2010) (Table 6). Following implementation of PITC, pre-test counselling/information ranged from 91.5% in a study from Kenya (van’t Hoog et al. 2005) to 100% in Zimbabwe (Chandisarewa et al. 2007). These four studies also presented results on the proportion of women tested and subsequently post-test counselled or received test results (van’t Hoog et al. 2005; Chandisarewa et al. 2007; Creek et al. 2007; Mirkuzie et al. 2010). Prior to the implementation of revised PITC models the proportion of women receiving post-test counselling and test results ranged from 72% who received test results in Botswana (Creek et al. 2007) to 96.9% in Zimbabwe (Chandisarewa et al. 2007). Following implementation of revised PITC models post-test counselling/receipt of results increased significantly to 99.8% (P < 0.001) in Zimbabwe (Chandisarewa et al. 2007) and to 82% (P < 0.001) in Botswana (Creek et al. 2007). Five studies reported results for linkage to ARV for PMTCT (van’t Hoog et al. 2005; Chandisarewa et al. 2007; Creek et al. 2007; Byamugisha et al. 2010; Mirkuzie et al. 2010) with three reporting increased uptake of ARV for PMTCT and four reporting an increase in the absolute number of patients identified as HIV positive and linked to ARV for PMTCT (van’t Hoog et al. 2005; Byamugisha et al. 2010; Mirkuzie et al. 2010). Across all studies, at least 20% of women were not linked to ARV for PMTCT (van’t Hoog et al. 2005; Chandisarewa et al. 2007; Creek et al. 2007; Byamugisha et al. 2010; Mirkuzie et al. 2010).

Table 6. Outcomes of the PITC components
Pre-test counselled/informationPITC modelComparison group
  1. *P < 0.05.

Kenya (van’t Hoog et al. 2005) 91.5% (n = 3743/4089)*77.4% (n = 3206/4142)
Uganda (Byamugisha et al. 2010)98.3% (n = 24 171/24 595)*62.3% (n = 18 583/29 834)
Zimbabwe (Chandisarewa et al. 2007)100% (n = 4551/4551)100% (n = 4872/4872)
Ethiopia (Mirkuzie et al. 2010)94%71.9%
Patients post-test counselled
 Kenya (van’t Hoog et al. 2005)90.3% (n = 2799/3101)89.3% (n = 2278/2551)
 Botswana (Creek et al. 2007)82% (n = 753/914)*72% (n = 272/377)
 Zimbabwe (Chandisarewa et al. 2007)99.8% (n = 4538/4547)*96.9% (n = 2964/3058)
 Ethiopia (Mirkuzie et al. 2010)84.5%50.7%
Linkage to PMTCT for women with positive results
 Kenya (van’t Hoog et al. 2005)70% (n = 471/673)*56.6% (n = 302/534)
 Botswana (Creek et al. 2007)66% (n = 144/220)70% (n = 87/125)
 Zimbabwe (Chandisarewa et al. 2007)73% (n = 663/908)76.3% (n = 372/487)
 Uganda (Byamugisha et al. 2010)77.2% (n = 885/1147)*55.8% (n = 316/566)
 Ethiopia (Mirkuzie et al. 2010)53.7%33.8%

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Our review suggests that implementation of PITC as a standard component of clinical care in ANC settings decreased lost opportunities to test pregnant women for HIV. In 3/10 studies, lost opportunities were <10% following introduction of PITC, and in 8/10, lost opportunities were <20%. Adopting PITC in ANC closes the gap towards achieving universal voluntary HIV testing of pregnant women and consequently increases the opportunities for pregnant women to access PMTCT and appropriate treatment and prevention interventions. Many key components of VCT were retained such as pre- and post-test counselling, although we were not able to assess directly the quality of these components or the extent to which patients felt coerced into testing. Linkage to ARV for PMTCT was low, with at least 20% of HIV-infected women not linked to ARVs, highlighting the need for improved linkage to PMTCT services.

The studies included in the review were of variable quality and had significant heterogeneity, creating challenges in data synthesis and interpretation, particularly of the outcomes of the components of the PITC process, for which there was limited data. Consequently, our review may be subject to limitations. First, we included only published studies, and published research is more likely to present statistically significant findings than unpublished research (Dwan et al. 2008). Further, despite systematically searching databases to minimise bias, no data from Latin America and the Caribbean or Asia were identified, which may be due to restricting our review to English language studies or a lack of published data from these regions despite the adoption of PITC in ANC in these settings. Second, the studies included in our review are subject to limitations. Only ten studies were identified for inclusion, the majority being NRS. Challenges associated with reporting and synthesising evidence from NRS are well documented (Deeks et al. 2003). A potential weakness of included studies is selection bias: there may have been differences between the intervention and ‘control’ groups at baseline, yet only one of the NRS controlled for these in analyses and only the study from Botswana compared groups to determine whether they had any significant differences in characteristics, including social and demographic characteristics (Deeks et al. 2003; Creek et al. 2007; Stringer et al. 2001). Hence, study results may overestimate the impact of PITC, as has been shown by other studies that compared evidence generated from NRS with results from RCTs (Deeks et al. 2003). In addition, NRS are vulnerable to time trends; to minimise the impact of these trends, we restricted our review to studies conducted within the same clinical settings with pre- and post-data as we felt that changes over time within the same facilities would most likely be a result of the interventions described. Finally, ambiguity in the use and differences in the interpretation of the terms opt-in, VCT and client-initiated may have implications on our classification of the model of testing in the control period of the studies.

Notwithstanding these limitations, we attempted to synthesis the findings of studies on similar primary outcomes assessed through similar study designs and in similar (ANC) settings. Most studies were from African countries with generalised epidemics. The overall direction and size of the effect provide evidence of the impact of this strategy on universal voluntary testing in ANC in similar contexts. National testing figures report substantial increases in HIV testing coverage where the testing model is PITC (WHO 2008, 2009, 2010): WHO 2010 Universal Access Report documents high coverage of HIV testing amongst pregnant women in Europe and Central Asia at 75% and in Latin America and the Caribbean at 57% (WHO 2010). In Eastern and Southern Africa, the region with the highest HIV prevalence, HIV testing amongst pregnant women reached over 50%, an increase from 43% in 2008 (WHO 2010). In Western and Central Africa, coverage increased from 16% to 21% between 2008 and 2009, with 54% [40–84%] of pregnant women with HIV in sub-Saharan Africa receiving ARVs for PMTCT compared with 45% [37–58%] in 2008 (WHO 2010). Consequently, although attributing reduced lost opportunities for testing to PITC is challenging in the absence of RCTs, as testing uptake is likely to be influenced by multiple correlated factors including the availability of rapid testing kits, increased availability of ART, healthcare providers expectations and attitudes, and the introduction of free services (Moses et al. 2008; April et al. 2009), the consistency and strength of association suggest that these results are reliable in terms of direction and magnitude.

A number of important questions remain. Reports of high HIV testing coverage and a reduction in lost opportunities where testing is routinely offered have been met with caution by some as concerns remain that, in the absence of written informed consent following individual pre-test information, intentional and unintentional coercion may confound patient choice (Gruskin et al. 2008; Becker et al. 2009). The process by which individuals decide to undergo testing is complex, being influenced not only by available information, but also by cultural, societal and individual beliefs and values and the potential individual and social implications of a positive test result (Gruskin et al. 2008). The routine offer of HIV testing in ANC settings might, however, simplify the decision-making process, through the normalisation of testing by removing fear of stigma, discrimination and moral judgements where testing is client-initiated (Oosterhoff et al. 2008). PITC has also raised ethical and human rights concerns in relation to testing and counselling and concerns regarding potential psychological consequences of a positive result in the absence of extensive pre-test counselling (Koo et al. 2006). Concerns exist that PITC will fail to adhere to the principles of consent, confidentiality and counselling inherent to VCT. Yet, anecdotal evidence suggests that where the testing strategy is VCT, pre- and/or post-test counselling is sometimes of poor quality or lacking (Paxton et al. 1999; Ayarza & Reyes 2002, Chopra et al. 2005). In our review, we found that opt-out PITC adhered to the standards inherent to counselling and testing, with data indicating that women who were tested also receive pre-test counselling/information and that a majority of the women tested receive post-test counselling. The quality of these components was, however, not assessed, and additional research of the quality of the PITC components is warranted.

Although PITC closes the gap towards universal testing, a minority of women continue to refuse testing. An increased understanding of the factors associated with opt-out, such as age, ethnicity and ≥12 years of education in the USA (Stringer et al. 2001), and marital status in Scotland (Simpson et al. 1998), as suggested by evidence from studies included in our review, would ensure opportunities to test specific subpopulations within ANC settings are addressed (Homsy et al. 2007). Similarly, further research regarding the acceptability of services within ANC settings is required as, although we present evidence of the adherence to the standards inherent to good HIV testing conduct, our review provides limited evidence of women’s perception and acceptability of the strategy. Studies included in our review suggest that the routine offer of testing is ‘helpful’ to women in Zimbabwe, with a majority of women in Scotland in favour of the availability of testing to all pregnant women (Simpson et al. 1998; Chandisarewa et al. 2007). In-depth personal perceptions regarding PITC processes were, however, not examined in the studies identified, and additional research is warranted.

Despite evidence that implementation of a PITC intervention may improve linkage to ARVs for PMTCT, at least 20% of women testing positive across five studies failed to access ARVs for PMTCT; in two studies conducted in Botswana and Zimbabwe, the proportion of women linked to ARVs for PMTCT decreased, albeit insignificantly. There is limited current published evidence of why a large proportion of women fail to be linked to ARV for PMTCT and of the effectiveness of linkage to ART, and subsequent adherence to ART regimens, for the management of the mother’s infection for women testing in ANC and requiring treatment for their own health. The benefits of testing, in the absence of effective linkage to PMTCT, treatment and care services, are limited. Understanding these linkages, and their weaknesses, is crucial to ensuring the benefits of testing are maximised. The studies included in this review focus primarily on HIV testing uptake. The majority are early PITC studies, conducted during a transitional period when community and political support and healthcare provider confidence in conducting PITC may have been limited. Additional research of recent studies, which focus on PITC and the effectiveness of linkages to ARVs and ART in the era of ART scale-up and of increased support for PITC, through the availability of WHO guidance and national policies recommending PITC, is required. In addition, the costs associated with the introduction of PITC need to be assessed alongside assessments of any impact PITC has on overburdened services and service providers. For individuals accepting the offer of testing, sexual behaviours following a negative test result also need to be monitored and researched as concerns exist that a negative result may result in sexual disinhibition (Kiene et al. 2009). Finally, our review was restricted to the contribution of PITC to universal voluntary testing for pregnant women in ANC settings; additional research is needed to better understand how PITC might contribute to universal voluntary testing amongst different populations within different clinical contexts.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Dr. Delia Boccia and Dr. Mishal Khan who were involved in determining eligibility and critically appraising studies identified for review. This study was funded by WHO and CDC.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Amornwichet P, Teeraratkul A, Simonds RJ et al. (2002) Preventing mother-to-child transmission. The first year of Thailand’s national program. JAMA288, 245248.
  • April M, Walensky R, Chang Y et al. (2009) HIV testing rates and outcomes in a South African community, 2001–2006: implications for expanded screening policies. Journal of Acquired Immune Deficiency Syndrome51, 310316.
  • Ayarza R & Reyes B (2002) Changes in quality of life of people living with HIV/AIDS in Ecuador. XIV International AIDS Conference. Barcelona, Spain.
  • Bayer R & Edington C (2009) HIV testing, human rights, and the global AIDS policy: exceptionalism and its discontents. Journal of Health Politics, Policy and Law34, 301323.
  • Becker J, Tsague L, Sahabo R & Twyman P (2009) Provider Initiated Testing and Counseling (PITC) for HIV in resource-limited clinical settings: important questions unanswered. Pan African Medical Journal 3, 15.
  • Byamugisha R, Tylleskar T, Kagawa MN, Onyango S, Karamagi CA & Tumwine JK (2010) Dramatic and sustained increase in HIV-testing rates among antenatal attendees in Eastern Uganda after a policy change from voluntary counselling and testing to routine counselling and testing for HIV: a retrospective analysis of hospital records, 2002–2009. BMC Health Services Research10, 290.
  • Centers for Disease Control & Prevention. (2008) Provider-initiated HIV testing and counseling of TB patients – Livingstone District, Zambia, September 2004–December 2006. MMWR – Morbidity & Mortality Weekly Report57, 285289.
  • Chandisarewa W, Stranix-Chibanda L, Chirapa E et al. (2007) Routine offer of antenatal HIV testing (“opt-out” approach) to prevent mother-to-child transmission of HIV in urban Zimbabwe. Bulletin of the World Health Organization 85, 843850.
  • Chersich M, Luchters SMF, Othigo MJ et al. (2008) HIV testing and counselling for women attending child health clinics: an opportunity for entry to prevent mother-to-child transmission and HIV treatment. International Journal of STD & AIDS19, 4246.
  • Chopra M, Doherty T, Jackson D & Ashworth A (2005) Preventing HIV transmission to children: quality of counselling of mothers in South Africa. Acta Paediatrica94, 357363.
  • Coates T, Grinstead O, Gregorich S et al. (2000) Efficacy of voluntary HIV-1 counselling and testing in individuals and couples in Kenya, Tanzania, and Trinidad: a randomised trial. The Voluntary HIV-1 Counseling and Testing Efficacy Study Group. The Lancet 8, 103112.
  • Creek TL, Ntumy R, Seipone K et al. (2007) Successful introduction of routine opt-out HIV testing in antenatal care in Botswana. Journal of Acquired Immune Deficiency Syndromes 45, 102107.
  • Deeks J, Dinnes J, D’Amico R et al. (2003) Evaluating non-randomised intervention studies. Health Technology Assessment 7, 1173.
  • Dwan K, Altman DG, Arnaiz JA et al. (2008) Systematic review of the empirical evidence of study publication bias and outcome reporting bias. PLoS ONE 3, e3081.
  • Fetene NW & Feleke AD (2010) Missed opportunities for earlier HIV testing and diagnosis at the health facilities of Dessie town, North East Ethiopia. BMC Public Health 10, 362.
  • Granich R, Gilks C, Dye C, De Cock K & Williams B (2009) Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model. The Lancet.373, 4857.
  • Gruskin S, Ahmed S & Ferguson L (2008) Provider-initiated HIV testing and counseling in health facilities – what does this mean for the health and human rights of pregnant women? Developing World Bioethics 8, 2332.
  • Higgins DL, Galavotti C, O’Reilly KR et al. (1991) Evidence for the effects of HIV antibody counseling and testing on risk behaviors. JAMA266, 24192429.
  • Homsy J, King R, Malamba S et al. (2007) The need for partner consent is a main reason for opting out of routine HIV testing for prevention of mother-to-child transmission in a rural Ugandan hospital. Journal of Acquired Immune Deficiency Syndromes44, 366369.
  • Kasenga F, Byass P, Emmelin M & Hurtig A-K (2009) The implications of policy changes on the uptake of a PMTCT programme in rural Malawi: first three years of experience. Global Health Action2, 17.
  • Kharsany ABM, Karim QA & Karim SSA (2010) Uptake of provider-initiated HIV testing and counseling among women attending an urban sexually transmitted disease clinic in South Africa – missed opportunities for early diagnosis of HIV infection. AIDS Care 22, 533537.
  • Kiene S, Bateganya M, Wanyenze R et al. (2009) Provider-initiated HIV testing in health care settings: should it be client-centered counselling?Journal des Sociaux du VIH/SIDA6, 115119.
  • Koo DJ, Begier EM, Henn MH, Sepkowitz KA & Kellerman SE (2006) HIV counseling and testing: less targeting, more testing. American Journal of Public Health 96, 962964.
  • Leon N, Colvin CJ, Lewin S, Mathews C & Jennings K (2010a) Provider-initiated testing and counselling for HIV – from debate to implementation. South African Medical Journal 100, 220221.
  • Leon N, Naidoo P, Mathews C, Lewin S & Lombard C (2010b) The impact of provider-initiated (opt-out) HIV testing and counseling of patients with sexually transmitted infection in Cape Town, South Africa: a controlled trial. Implementation Science5, 8.
  • Mirkuzie AH, Hinderaker SG & Morkve O (2010) Promising outcomes of a national programme for the prevention of mother-to-child HIV transmission in Addis Ababa: a retrospective study. BMC Health Services Research 10, 267.
  • Moses A, Zimba C, Kamanga E et al. (2008) Prevention of mother-to-child transmission: program changes and the effect on uptake of the HIVNET 012 regimen in Malawi. AIDS 22, 8387.
  • Nieburg P, Cannell T & Morrison J (2005) Expanded HIV Testing: Critical Gateway to HIV Treatment and Prevention Requires Major Resources, Effective Protections. The Center for Strategic International Studies, Washington D.C.
  • Oosterhoff P, Hardon AP, Nguyen TA, Pham NY & Wright P (2008) Dealing with a positive result: routine HIV testing of pregnant women in Vietnam. AIDS Care 20, 654659.
  • Paxton S, Diaz J, Junga R, Pillai A & Thomas J (1999) APN+ Human Rights Initiative: participatory research into HIV/AIDS-related human rights violations in 10 Asian countries. 5th International Congress on HIV/AIDS in Asia and the Pacific. Malaysia.
  • Seipone K (2004) Introduction of routine HIV testing in prenatal care – Botswana, 2004. MMWR. Morbidity and Mortality Weekly Report 53, 10831086.
  • Simpson WM, Johnstone FD, Boyd FM, Goldberg DJ, Hart GJ & Prescott RJ (1998) Uptake and acceptability of antenatal HIV testing: randomised controlled trial of different methods of offering the test. BMJ 316, 262267.
  • Simpson WM, Johnstone FD, Goldberg DJ, Gormley SM & Hart GJ (1999) Antenatal HIV testing: assessment of a routine voluntary approach. BMJ 318, 16601661.
  • Stringer EM, Stringer JS, Cliver SP, Goldenberg RL & Goepfert AR (2001) Evaluation of a new testing policy for human immunodeficiency virus to improve screening rates. Obstetrics and Gynecology 98, 11041108.
  • van’t Hoog AH, Mbori-Ngacha DA, Marum LH et al. (2005) Preventing mother-to-child transmission of HIV in Western Kenya: operational issues. Journal of Acquired Immune Deficiency Syndromes 40, 344349.
  • Weigel R, Kamthunzi P, Mwansambo C, Phiri S & Kazembe PN (2009) Effect of provider-initiated testing and counselling and integration of ART services on access to HIV diagnosis and treatment for children in Lilongwe, Malawi: a pre- post comparison. BMC Pediatrics 9, 80.
  • WHO. (2003) The Right to Know. New Approaches to HIV Testing and Counselling. WHO, Geneva.
  • WHO. (2008) Towards Universal Access: Scaling Up Priority HIV/AIDS Interventions in the Health Sector. WHO, Geneva.
  • WHO. (2009). Towards Universal Access: Scaling Up Priority HIV/AIDS Interventions in the Health Sector. WHO, Geneva.
  • WHO. (2010). Towards Universal Access: Scaling Up Priority HIV/AIDS Interventions in the Health Sector. WHO, Geneva.
  • WHO, UNAIDS. (2007) Guidance on Provider-Initiated HIV Testing and Counselling in Health Facilities. WHO, UNAIDS, Geneva.
  • Zimba C, Kamanga E, Chilongozi D et al. (2006) Impact of routine HIV counseling and testing with an opt-out strategy compared to voluntary counseling and testing in the implementation of PMTCT services, Lilongwe, Malawi 16th International AIDS Conference. Toronto, Canada.
  • Zimba C, Kamanga E, Chilongozi D et al. (2010) The Impact of Routine Counseling and Testing With an Opt-Out Strategy Compared to Voluntary Counseling and Testing in the Implementation of PMTCT Services in Lilongwe, Malawi. UNC Project, Lilongwe.