New AIDS & TB PROGRAMME 2013catalogue.safaids.net/sites/default/files/publications... · 2014. 5. 12. · AIDS & TB PROGRAMME 2013 v Foreword The 2012 ANC sentinel surveillance report

AIDS & TB PROGRAMME 2013 i

National Survey of HIV and Syphilis Prevalence among Women attending Antenatal Clinics in Zimbabwe 2012ii

ContentsForeword iAcknowledgements iiiExecutive Summary ivAcronym v1. INTRODUCTION 1 1.1CountryProfile 1 1.2 Background 22: OBJECTIVES 6 2.1 Broad Objectives 6 2.2Specificobjectives 63: METHOD 3.1 Survey design 6 3.2 Sampling 7 3.2.1 Sentinel population 7 3.2.2 Selection of survey population 7 3.2.3 Selection of sentinel surveillance sites 7 3.3 Sample size determination 7 3.6 Summary of Survey Operational Procedures 9 3.7 Laboratory procedures 10 3.7.1 Laboratory Methods 10 3.7.3 Quality Assurance of HIV Testing 11 3.8 Data entry and management 12 3.9 Data analysis 12 3.10 Overall quality assurance 124. RESULTS 13 4.1 Population characteristics 13 4.2DistributionofHIVPrevalenceinthefifty-threesentinelsites 17 4.3TrendsinHIVPrevalence2002-2012,ForAllWomen15-49yearsin NineteenSentinelSites2002-2012 29 4.4HIVPrevalenceandtrends,Womenage15-24yearsin NineteenSentinelSites,2002-2012 36 4.5ResultsofSyphilis 425. DISCUSSION AND CONCLUSIONS 44 5.1LessonsLearnt 466. RECOMMENDATIONS 46References 47

APPENDICES 48 Appendix1:2012SentinelSurveySitesProvinceandSentinelSiteClassification 48 Appendix 2: 2012 ANC Surveillance Form 50 Appendix 3: Laboratory HIV Parallel Testing Algorithm 53 Appendix 4: ANC Sentinel Surveillance Sites Since 2002 54

AIDS & TB PROGRAMME 2013 iii

List of Tables and FiguresTablesTable 1: Summary of Zimbabwe Population in 2012 1Table2:HealthandSocio-EconomicIndicators 1Table3:Distributionofclientsbyresidentialclassification 13Table 4: Distribution of client characteristics 16Table5:HIVPrevalencebySentinelSite,2012 18Table 6: Distribution of HIV prevalence by exposure to PMTCT services 26Table 7: Overall HIV prevalence statistics 29

Figures Figure1:OverallHIVpooledprevalenceforwomen15-49years,2012 17Figure2:HIVprevalencebysentinelsitelocation,2012 20Figure3:MedianHIVprevalencebyprovince,2012 20Figure4:HIVprevalencebyage-groupamongANCattendees,2012 21Figure5:Urban-ruralcomparisonofHIVprevalenceamongANCattendees15-49years,2012 21Figure6:HIVprevalencebyeducationlevel,2012 22Figure8:HIVPrevalencebyoccupation,2012 24Figure9:HIVPrevalencebygravidityamongANCattendees,15-49years 25Figure10:HIVPrevalencebynumberofabortionsandstill-births,2012 25Figure11:PrevalenceofHIVbyhistoryandpresenceofGUD 26Figure 12: HIV prevalence by partner’s occupation 27Figure 13: HIV prevalence by partner’s place of residence 28Figure 14: HIV prevalence by partner’s level of education 28Figure15:OverallHIVprevalencetrends 29Figure16:HIVprevalencetrendsbyage-group 30Figure 17: HIV prevalence trends by level of education 31Figure 18 : HIV prevalence trends by occupation 32Figure 19: Trends in HIV prevalence by sentinel site location 33Figure 20: Trends in HIV prevalence by occupation 33Figure 21: HIV prevalence trends by province 2012 34Figure22:OverallHIVprevalencetrendsbyyoungwomen(15-24years) 36Figure23:HIVprevalencetrendsinyoungwomen15-24yearsbysite 37Figure24:HIVprevalenceinyoungwomenbyage-group 39Figure25:TrendsinHIVprevalencebyeducationlevelinyoungwomen15-24years 39Figure26:HIVprevalencetrendsbysentinellocation15-24years 40Figure27:HIVprevalencetrendsbyprovince15-24years 41Figure28:PercentageRPRPositiveamong,women,15-49years,and2012 42Figure29:PercentageRPRPositivebyAge-Group,AmongANCAttendees,15-49years,2012 42Figure30:PrevalenceofRPRpositivebygravidity 43

National Survey of HIV and Syphilis Prevalence among Women attending Antenatal Clinics in Zimbabwe 2012iv

AppendicesAppendix1:SentinelSitesClassification 59

Appendix 2: 2012 ANC Surveillance Form 61

Appendix 3: Laboratory Testing Algorithm 63

Appendix 4: Parallel Testing Algorithm 64

Appendix5:ReportonEvaluationofRapidHIVtestKitsusingELISAHIVtestKits 68

Appendix 6: Comparison of 2008 PMTCT data and 2009 ANC Site Prevalence data 75

Appendix7:TrendsinHIVPrevalenceallwomen15-49years,GenescreenHIVTest 76

Appendix 8: List of ANC Survey Participants 77

AIDS & TB PROGRAMME 2013 v

ForewordThe 2012 ANC sentinel surveillance report is a follow up on the 2009 ANC survey. HIV prevalencehasdeclinedfrom16.1%in2009to15.9%in2012amongpregnantwomen15-49years old. Whileprevalenceinmostsentinelsiteshascontinuedtodecline,somesentinelsitesregisterednotable increases in HIV prevalence. High HIV infection rates were observed among women 35-39(26.7%)and40-44(26.0%)yearsofage. HIVprevalencehasremainedhighatsitesclassifiedas‘Other’,(growthpoints,borderposts,miningandresettlementfarms)comparedto the urban and rural areas.

Althoughtheobserveddecreasingtrendisencouraging,overallHIVsero-prevalenceamongwomen attendingANC in Zimbabwe remains high. Although there has been a significantdecreaseinHIVprevalenceinthe15-24yearagegroup(11.6%in2009to9.85%in2012),increased efforts to scale up prevention campaigns targeting youth are still needed.

ThepositivesignsinourfightagainstHIVandAIDSshouldspureveryZimbabweantore-double their efforts and commitment to further reduce the burden of HIV and AIDS.

Brigadier General (Dr.) G. GwinjiPermanentSecretary,MinistryofHealthandChildCareZimbabwe

National Survey of HIV and Syphilis Prevalence among Women attending Antenatal Clinics in Zimbabwe 2012vi

This project has been supported by the President’s Emergency Plan forAIDSRelief (PEPFAR) throughCooperative between the Centers for Disease Control and Prevention and the University of ZimbabweDepartment of Community Medicine SEAM Project under the terms of Cooperative Agreement Number: 1U2GGH000315-01

AIDS & TB PROGRAMME 2013 vii

AcknowledgementsThe Ministry of Health and Child Care would like to extend its gratitude to all the onsite health careworkers,includingthesentinelsitestaff,midwives,aswellastheirprovincialandnationalsupervisors who participated in the Antenatal Clinic Survey for the year 2009. Their hard work and dedication ensured the success of the survey.

We would also like to acknowledge the NationalAIDS Council (NAC) for collaboratingwiththeAIDSandTBUnitintraininganddataabstractionandtotheNationalMicrobiologyReferenceLaboratory(NMRL)fortestingandprovidingqualityassuranceforallthesurveysamples.

Wewish to express our gratitude to: Central StatisticalOffice (CSO),Centers forDiseaseControlandPrevention,Zimbabwe(CDCZimbabwe),ImperialCollegeLondon,JointUnitedNations Programme on HIV/AIDS (UNAIDS), United Nations Children Emergency Fund(UNICEF),UnitedNationsPopulationFund (UNFPA),WorldHealthOrganization (WHO)andElizabethGlazierPaediatricAIDSFoundation (EGPAF) supporting the conduct of thesurvey and providing assistance with data analysis and report writing.

Finally,wewouldliketoexpressourappreciationandgratitudetothepregnantwomenwhoparticipatedanonymouslyinthisstudy,withoutwhomthissurveywouldnothavetakenplace.

Dr. Owen MugurungiDirector AIDS and TB Unit

National Survey of HIV and Syphilis Prevalence among Women attending Antenatal Clinics in Zimbabwe 2012viii

Executive SummaryIn2012,theMinistryofHealthandChildCare(MOHCC),AIDSandTBunitconductedHIVsero-prevalencesurveyofwomenattendingantenatalclinics(ANC)tomonitorthelevelandtrendsinHIVprevalence.Atotalof54sentinelsiteswereselectedtoparticipateinthesurveyusing probability proportional to size sampling method to represent 10 provinces in the country. Atotalof18,437ANCclientswereconsecutivelyenrolledinthesurveyoveraperiodoffourmonths. Leftover blood specimens collected from pregnant women for routine screening at theirfirstANCvisitwereusedforHIVantibodytesting.Allpersonalidentifierswereremovedfrom the specimens ensuring that testing was unlinked and anonymous.

The highest HIV prevalence was observed in Matabeleland South 23.8% while the lowest was inHarare10.8%.Theprevalenceintheothereightprovinceswas;MatabelelandNorth19.6%,Mashonaland East 17.6%, Midlands 17.2%, Bulawayo 17.1%, Mashonaland West 15.6%Masvingo14.6%,Manicaland13.9%andMashonalandCentral12.4%.ThenationalmedianHIVprevalencewas15.9%.HIVprevalenceamongwomen15-49yearswasslightlyhigheramongwomenwhoaccessedservicesinurbansentinelsites(16.1%)ascomparedtotheruralsentinelsites15.7%.

TherewasagradualincreaseinHIVprevalencewithagegroupfrom5.6%inthe15-19yearspeakingto26.7%in the35-39yearoldgroupfollowedbyadecline to23.1%in the45-49year old group. HIV prevalence was highest among women who had not received any form of education,thosewhowerenotmarried,thoseemployedintheinformalsectorandthosewithprevious history of sexual transmitted infection.

TheHIVprevalenceamongpregnantwomen15-49yearshasdeclinedfrom16.1%in2009to15.9%in2012.Asimilartrendwasobservedinthe15-24agegroupwhereHIVdeclinedfrom11.6%in2009to9.85%in2012.Significantdeclineswereatthefollowingprovinces:MashonalandCentral(25.5%in2002to12.4%in2012),MashonalandWest(27.1%in2002to15.6%in2012)andManicaland(25.6%in2002to13.9%in2012).Thereweresomeincreasesin prevalence inMatabeleland South,Bulawayo,Midlands andMashonalandEast in 2012compared to 2009.

TheANCsurveywassupposetotestforcurrentsyphilisinfectionusingRPRscreeningtestsandtheTPHAconfirmatorytestbuthoweveronlyRPRwasdoneduetotheshortageofTPHAkits. This survey was therefore unable to provide the prevalence of syphilis among pregnant women.

Inconclusion,HIVprevalenceamongpregnantwomen15-49yearsstillremainsunacceptablyhigh. The prevalence of HIV has however shown a decline in 2012 as compared to previous years showing that the current preventive efforts are bearing fruits. There is a need to invest in the scale up of prevention strategies targeting all women of child bearing age in order to reduce the incidence of HIV infection.

AIDS & TB PROGRAMME 2013 ix

AcronymAcronym Full name

ANC Antenatal Clinic

AIDS Acquired Immune Deficiency Syndrome

CDC-Zimbabwe Centers for Disease Control and Prevention Zimbabwe

DBS Dried Blood Spot

EIA Enzyme Immunoassay

EQA External Quality Assurance

HIV Human Immunodeficiency Virus

NMRL National Microbiology Reference Laboratory

MLS Medical Laboratory Scientist

MOHCC Ministry of Health and Child Care

PCR Polymerase Chain Reaction

PITC Provider Initiated Testing and Counseling

PNO Provincial Nursing Officer

PMLS Provincial Medical Laboratory Scientist

PMTCT Prevention of Mother to Child Transmission of HIV

RPR Rapid Plasma Reagin

TPHA Treponema Pallidum Haemaglutination Assay

VCT Voluntary Counseling and Testing

ZINQAP Zimbabwe National Quality Assurance Program

National Survey of HIV and Syphilis Prevalence among Women attending Antenatal Clinics in Zimbabwe 2012x

AIDS & TB PROGRAMME 2013 1

1.0 INTRODUCTION1.1 Country ProfileZimbabwe has a male to female ratio of 93:100 (Table 1)

Table 1: Summary of Zimbabwe Population in 2012

Number

Male 6 234 931

Female 6 738 877

Total 12 973 808

Zimbabwe Population Census 2012 The country reported low socio-economic indicators with only 66% of births attended by skilled health worker in the last 5 years (Table 2)

Table 2: Health and Socio-Economic Indicators

Indicator Value

Total Fertility rate (ZDHS 2010) 4.1%

Expected live births, 2011 (Health Information, Ministry of Health and Child Care Zimbabwe) 412, 122

Percentage of women who received ANC care from a skilled health personnel for a live birth in the last 5 years (ZDHS 2010)

90%

Births attended by a skilled health worker for a birth in the last 5 years (ZDHS 2010) 66%

1.2 Background

Zimbabwe is experiencing a declining HIV epidemic although HIV prevalence remains high. This decline has been noted in biannual antenatal clinic surveillance (ANC) conducted in 2002, 2004, 2006 and 2009 and the HIV and AIDS estimates from modeling. These findings were corroborated by population based estimates from the Zimbabwe Demographic and Health Surveys of 2005/6 and 20101.

In line with international standards for monitoring the HIV epidemic Zimbabwe has used both population based and ANC sentinel surveillance to collect critical information for informing policy makers and planners of the status of the HIV epidemic. Surveillance of pregnant women who routinely attend ANC sentinel sites has provided valuable information about the burden of HIV and trends in HIV prevalence2,3 ANC sentinel surveillance (ANC HSS) is conducted using unlinked anonymous testing (UAT), in which leftover blood from routine testing at ANC clinics (usually syphilis testing) is stripped of all personal identifiers and used for HIV surveillance. In UAT-based ANC HSS, informed consent is not obtained from pregnant women and test results

National Survey of HIV and Syphilis Prevalence among Women attending Antenatal Clinics in Zimbabwe 20122

are not returned, thus eliminating a potential source of selection bias due to non-consent. When testing is based on informed consent individuals refusing testing may be at lower or higher risk of HIV infection than consenters. The main advantages of using ANC data include lower cost of data collection compared to population surveys and accessibility of populations. The lack of universal coverage of ANC services in developing countries, exclusion of non-pregnant women and exclusion of men tend to make these data less representative of the general population and this has been documented by Brockmeyer and others4. On the other hand, HIV testing in Demographic and Health Surveys (DHS) and AIDS Indicator Surveys (AIS)provide nationally representative estimates of HIV prevalence in the general population and have the advantage of linking socio-demographic and behavioral data to Sero-status of individuals. These surveys, however, are affected by non-response bias, exclude non-household populations, and are relatively more expensive and not conducted annually.

Over the last decade, expanded access to HIV prevention, care and treatment interventions has highlighted ethical issues inherent in UAT-based ANC HSS. The conduct of ANC HSS using UAT raises ethical concerns given that this methodology does not obtain informed consent from pregnant women, provide them with their HIV testing results or refer them to HIV care, treatment, and prevention interventions if test results are positive5,6.[8, 9]. A growing consensus has reported that, in the context of ART and PMTCT expansion, alternative surveillance methods and data sources are increasingly available and should be explored to address ethical issues associated with UAT-based ANC HSS.

1.3 Antenatal clinic and PMTCT services in Zimbabwe

Zimbabwe provides antenatal care in all its 1643 public health facilities. According to the ZDHS, 90% of women age 15- 49 years who gave birth in the five years preceding the survey received antenatal care from a skilled provider during pregnancy of their most recent birth. There was a slight decline from the 94% reported in 2005/6 to 90% in 2010 ZDHS and this was attributable to high user fees personal socio-economic factors such as high transport costs.

The Ministry of Health and Child Care has mandated all facilities in Zimbabwe to offer Provider Initiated Testing and Counseling (PITC) services. PMTCT services are integrated in ANC services at all sites. The PMTCT programme has rapidly expanded since its nationwide roll-out in 2002; with 1560 out of the 1643 facilities (95%) in the public sector offering PMTCT services. An increase in the proportion of PMTCT sites offering comprehensive services was noted from 77% (1200) in 2010 to 89% (1390) in 2011.

Pregnant women are encouraged to visit the ANC/PMTCT clinic at 12 weeks of pregnancy. The initial visit involves recording of demographic characteristics of patient in the ANC register and the issuance of a personal booking card. The women are physically examined and screening tests such as for syphilis, blood grouping are conducted. With integration of PMTCT services to ANC, in theory, all women are offered individual counseling and testing, but can “opt- out”. Women who refuse HIV testing at their first booking can access these services on a voluntary basis at subsequent visits. HIV positive pregnant women are prioritized for access to antiretroviral therapy.PMTCT sites provide a short course of Nevirapine and dual combination therapy of Zidovudine and Lamivudine prophylaxis in addition to other routine PMTCT services.Of the 412, 122 estimated pregnancies in 2011, 403 938 (96%) presented for first ANC booking; and out of these, 341 725 (85%) accepted HIV test in PMTCT7 (PMTCT REPORT 2011).


1.4 ANC Sentinel Surveillance

Since 1989, Zimbabwe has conducted HIV surveillance of pregnant women who routinely attend ANC sentinel sites to provide valuable information about the burden of HIV and trends in HIV prevalence. ANC sentinel surveillance (ANC HSS)is conducted using unlinked anonymous testing (UAT), in which leftover blood from routinesyphilis testingis stripped of all personal identifiers and used for HIV surveillance. Pregnant women presenting for the first time with their current pregnancy at the participating ANC sites during the survey period are enrolled in the anonymous unlinked Sero-survey. A minimum set of data are extracted from antenatal clinic booking cards and entered into the ANC survey form. The use of informed consent is not obtained and test results are not provided to participating women.ANC surveys were conducted in 2000, 2001, 2002, 2004, 2006 and 2009. Nineteen sentinel sites have participated since 2002 to 2009 and the testing algorithm was standardized. The HIV prevalence among pregnant women 15-49 years has declined from 25.7 (95% CI 24.7 -26.7) in 2002 to 16.1% (95% CI 15.3-17.0) in 2009. The decline was significant (p <0.001). A similar trend was observed in the 15-24 age-group where HIV declined from 20.8% (95% CI 19.8 -21.8) in 2002 to 11.6% (95% CI 10.6 -12.6%) (p<0.001) in 2009.

In order to provide provincially representative estimates of HIV prevalence, the surveillance group conducted the 2012 HIV surveillance in 54 provincially representative ANC/PMTCT sites. These sentinel ANC sites include clinics in rural areas, growth points, commercial farms, mining areas, and urban areas.

Laboratory testing using ELISA tests kits and confirmatory Western Blot were performed at the National Microbiology Reference Laboratory (NMRL) due to lack of capacity in some of the district laboratories.

The data from the 2012 ANC Survey were used by the Ministry of Health and Child Care (MOHCC) to establish the prevalence and trends of HIV and syphilis infection among antenatal clinic attendees at the sentinel sites. In addition, the ANC Surveillance data and other available HIV prevalence data were used by the MOHCC and the Estimates Working Group to generate national HIV and AIDS estimates for Zimbabwe using the Spectrum software.

Although ANC sentinel surveillance utilizing this method of unlinked anonymous testing (UAT) is currently the mainstay for analyzing HIV trends over time in many Sub-Saharan African countries, there are ethical considerations inherent in this method. Unlinked anonymous testing does not provide women with HIV counseling,or inform women of their test result or provide linkages to care and treatment for women who test positive. Further, as a separate and additional data collection activity involving ELISA, RPR and TPHA HIV testing, UAT-based ANC sentinel surveillance can consume significant resources. In the light of a growing PMTCT program in Zimbabwe, this survey also evaluated the utility of using PMTCT data in replacing anonymous unlinked Sero-surveillance and results are reported elsewhere(Y).


2.0 OBJECTIVES2.1 Broad ObjectivesThe broad objective of the 2012 ANC survey was to establish the HIV and syphilis prevalence among antenatal clinic clients in order to monitor the HIV and AIDS epidemic and to develop and plan interventions for HIV and AIDS prevention and control in Zimbabwe.

2.2 Specific objectives• To determine the prevalence and incidence of HIV and syphilis among women (15-49

years) attending sentinel antenatal clinics in Zimbabwe;• To describe HIV and syphilis infection among sub-populations of these women;• To describe the recent trends in the prevalence of HIV among women attending sentinel

antenatal clinics; • To provide information for planning integrated activities for prevention and control of HIV

and AIDS for groups with HIV infection and AIDS and those not infected; and• To store data and specimens for additional studies related to understanding the HIV AND

AIDS epidemic in Zimbabwe.


3.0 METHODS3.1 Survey designThe national antenatal HIV and syphilis prevalence was an anonymous, unlinked, cross-sectional survey targeting pregnant women attending antenatal clinics in the public health sector. This survey expanded its sample size to 18 500 pregnant women recruited from 54 health facilities which formed the primary sampling units (PSU) compared with 7 363 women recruited from 19 sites in 2009. This has expanded geographic coverage considerably to include clients from all the ten provinces, making it possible to generate national and provincial HIV estimates using the EPP spectrum model. The survey was conducted over a period of sixteen weeks starting in June of 2012. 3.2 Sampling3.2.1 Sentinel populationPregnant women (15-49 years) attending ANC services at public health facilities for the first time with current pregnancy were used as the target population to minimize the chances of obtaining information from the same woman more than once.

3.2.2 Selection of survey populationInclusion criteria

All pregnant women aged 15 – 49 years attending antenatal care services for the first time during their current pregnancy were eligible for inclusion.

Exclusion criteria

Pregnant women who had previously visited the ANC clinic during their current pregnancy during the survey period were excluded from participation on the basis of their known status

3.2.3 Selection of sentinel surveillance sitesThe basic goal was to select 55 national sentinel surveillance sites representative for the population size estimate of the province while maintaining the 19 traditional sites which are used in the trend analysis. The 36 additional sentinel sites were selected using the ‘Probability Proportional to size’ (PPS) method based on the reported 2011 facility list first ANC booking to give a total of 55 sites.

3.3 Sample size determination

Sample size was calculated to be 18,500 using the Fleiss formula at 95% confidence interval, with a precision of 0.05, statistical power (β) of 80% and a significance level (α)of P<0.05 determining a statistically significant decline in HIV prevalence of 10% from 16.1% to 14.5%.

Sample size for all other sites except for city sites was calculated to be 330 per site to ensure that the true prevalence will be captured 95% of the time. Sample size for the cities i.e. Kuwadzana, Nkulumane, and St. Mary’s Clinics was higher at 550 in order to allow for estimation of HIV prevalence among urban young women below the age of 25 years in an attempt to assess recent infection, since there were few urban clinics in the sample.


3.4 Training and Supervision

All survey staff (sentinel supervisors, midwives and laboratory staff) were trained at a central place. Members of the technical working group participated in the training of staff in the survey procedures. Supervisory visits were conducted during the survey period. Special support was provided to sites that were facing challenges.

3.6 Summary of Survey Operational Procedures

• History taking on all ANC booking cards• Selection of pregnant women on first antenatal booking for inclusion in the survey• Transfer of demographic information from ANC booking card to ANC survey form• Drawing of blood for routine antenatal tests, i.e., ABO blood grouping, Rhesus typing, or

syphilis testing and labeling for routine tests. Remnant blood was placed in a survey tube and labeled using a bar code

• Preparation and labeling of dried blood spots • All DBS specimens were left to dry for at least 3 hours in a horizontal position then placed

in a zip lock bag with a desiccant pack and stored at 4oC • Samples were initially sent to a separating laboratory, spun and separated. Survey serum

was placed into 1.8 ml cryo-tubes and stored in the freezer (-13oC), until ready for transport to the testing laboratory.

• All samples were transported to the testing laboratory, the National Microbiology Reference Laboratory (NMRL), in Harare, while maintaining the cold chain.

3.7 Laboratory procedures

3.7.1 Laboratory Methods

The sentinel site ANC nurse collected five milliliters (5ml) of whole blood by venipuncture. From the 5 ml tube, 2.5 ml was transferred to a plain tube for routine blood typing and other tests. The remaining 2.5 ml, five (5) drops were placed on blotting paper to be stored as dried blood spots (DBS) for the survey. The test tube and DBS and corresponding forms were labeled with preprinted barcode stickers to ensure correct linkage of test results with survey data collection forms. All DBS specimens were left to dry for at least 3 hours in a horizontal position then placed in a zip lock bag with desiccant pack to maintain integrity before and during transportation.

The completed survey forms were sent with the DBS to the NMRL using EMS envelopes or site-specific transportation logistics to ensure timely transportation. A tally sheet was sent with the specimens to ensure all survey specimens were accounted for. All forms were scanned into an electronic database at the NMRL.


3.7.2 Laboratory testingSamples were tested using two Enzyme Immuno-Assay (EIA) test kits, AniLabsytems (AniLabsystems Ltd, Finland)1 which has been used since 2002 and the newly introduced Enzygnost Anti HIV 1/ 2 plus (Siemens Healthcare products, Munich Germany)2. Discordant samples results were re-tested using the same test kits. Samples that remained discordant were then tested using Western Blotting, New Lav Blot 1 (Biorad, France). The Western Blot was used to resolve discordant results. In order for a person to be considered HIV positive, using the Western blot test they needed to have either antibodies against one of the envelope proteins and one of the core proteins, or against one of the enzymes. The proteins are listed below3

• Proteins from the HIV envelope: gp41, and gp120/gp160.• Proteins from the core of the virus: p17, p24, p55• Enzymes that HIV uses in the process of infection: p31, p51, p66Polymerase chain reaction by ROCHE4 was then used resolve the indeterminate samples. The full testing algorithm is referenced in Appendix 3.

3.7.3 Quality Assurance of HIV TestingInternal quality assurance was conducted at the NMRL. External quality assurance on 10% of the samples was conducted by Zvitambo. All samples passed the external quality assurance and all results from the NMRL were accepted for analysis.

The results of the parallel testing algorithm were used to describe the overall and detailed trends from 2002 through 2012.

3.7.4 Testing for Syphilis

The Immutrep Rapid Plasma Reagin (RPR) testing kit5 was used to screen all women for syphilis.Samples that were reactive for RPR were further tested with the Omega Diagnostics Immutrep Treponema pallidumhaem-agglutation (TPHA) testing kit6. Samples that were reactive for both RPR and TPHA were considered positive for current infection with syphilis.

The RPR kit is a non-treponema method for the serological detection of syphilis. The antigen is a particulate carbon suspension coated with lipid complexes that agglutinate in the presence of plasma reagins. Reagins are antibodies present in the sera of syphilitic patients. Visible agglutination in the form black clumps which was viewed macroscopically, and this indicated the presence of such antibodies in the sample that was being tested. The Omega Diagnostics Immutrep TPHA comprises T. pallidum sensitized formolised tanned fowl erythrocytes; unsensitisedformolised tanned fowl erythrocytes, diluent and control sera. When diluted positive samples are mixed with sensitized erythrocytes, antibody to the sensitizing antigen causes agglutination of the cells. The cells form a characteristic pattern of cells in the bottom of a micro-titration plate well. In the absence of antibody, they form a compact button in the well and provide conclusive results for the presence of current syphilis infection.

1 http://www.anilabsystems.com/anilabsystems/pdfs/HIVFebr2008.pdf2 http://healthcare.siemens.com/clinical-specialities/infectious-disease/infectious-diseases-hiv3 http://www.bio-rad.com/prd/en/US/CDG/PDP/M4T5SX15/GS-HIV-1-Western-Blot4 http://molecular.roche.com/About/pcr/Pages/PCRProces.aspx5 http://www.biokits.com/field/201/Medical%20Science/Rapid-Plasma-Reagin-RPR-Test-Kits6 http://www.eugene-chen.com.tw


3.8 Data entry and management

Data was first collated by the NMRL database officer. Forms with all laboratory results were forwarded to the data entry team. Data collection forms received from the NMRL were checked for completeness, errors and inconsistencies as part of the data cleaning process.

Forms were entered into a database using teleform scanning software by site. The scan system created an access database. All scanned entries were checked by another data entry clerk. The dataset were cleaned and an analysis dataset constructed using Microsoft Access. All outlying data were verifiedusing the original data collection form. An electronic data set was stored on Ministry of Health and Child Care servers for security of data storage. After completion of data entry, the forms were stored for safekeeping in a locked room at the Ministry of Health and Child Care, AIDS and TB Unit.

3.9 Data analysis

The primary outcome of the ANC data analysis was to determine HIV prevalence of all women attending sentinel antenatal clinics in Zimbabwe. All analysis was conducted in STATA Intercooled Version 10. Confidence intervals and p values were calculated to test for significance.

One secondary outcome of the data analysis was to generate data for the national estimates process. ANC site level data were used with other national survey and monitoring data to estimate the current impact of the HIV epidemic. These estimates are presented in the HIV Estimates Report of 2013.

3.10 Overall quality assurance

The overall oversight of the 2012 ANC Survey was provided by the ANC Technical Working Group in consultation with the national surveillance officer who provided coordination and communication between the different people responsible for completing the survey. This group with assistance from the national survey team, coordinated data collection, analysis and report production.


4.0 RESULTS4.1 Population characteristics

A total of 18437 ANC clients contributed to the 2012 national Sero-surveillance. Most women visiting a sentinel site reported that they resided in the clinic’s catchments area (Table 3). Most of the ANC attendees resided in towns and rural areas (39.8% each). The other attendees resided in farming areas 8.26%, growth points 4.3%, mining areas 1.71% and a minority had their place of residence classified as missing 0.986%


Table 3: Distribution of clients by residential classificationProvince Site name Farm

n (%)Growth Pointn (%)

Minen (%)

Ruraln (%)

Town or Cityn (%)

Missingn (%) Total

Bulawayo Nkulumane 1 (0.2) 2 (0.3) 0 (0.0) 7 (1.2) 581 (96.8) 9 (1.5) 600Chitungwiza St Marys 5 (0.8) 3 (0.5) 0 (0.0) 9 (1.5) 582 (97.0) 1 (0.1) 600Harare Hatcliffe 5 (1.5) 6 (1.8) 1 (0.3) 9 (2.7) 317 (93.2) 1 (0.3) 340 Glenview 0 (0.0) 1(0.3) 0 (0.0) 0 (0.0) 328 (98.2) 5 (1.5) 334 Kuwadzana 4 (0.7) 0 (0.0) 0 (0.0) 0 (0.0) 593 (98.8) 3 (0.5) 600Manicaland Hauna 23 (10.3) 17 (7.6) 8 (3.6) 165 (73.7) 10 (4.5) 1 (0.5) 224 Murambinda 5 (1.5) 37 (10.9) 2 (0.6) 283 (83.5) 11 (3.2) 1 (0.3) 339 Mutambara 4 (1.3) 10 (3.2) 0 (0.0) 282 (90.7) 4 (1.3) 11 (3.5) 311 Nyanga 21 (7.6) 2 (0.7) 0 (0.0) 167 (60.7) 62 (22.6) 1 (0.36) 275 Rusape 25 (21.1) 3 (2.5) 1 (0.8) 17 (14.3) 51 (42.9) 1 (0.84) 119 Sakubva 0 (0.0) 0 (0.0) 0 (0.0) 3 (0.9) 337 (99.1) 0 (0.0) 340Mashonaland Central Bindura 133 (39.7) 1(0.3) 15 4.5) 77(23.0) 108(32.2) 1(0.3) 335 Chimhanda 2(0.6) 16 (4.8) 0 (0.0) 308 (93.1) 2 (0.6) 3 (0.9) 331 Concession 206 (61.1) 23 (6.8) 21 6.2) 2 (0.6) 83 (24.6) 2 (0.6) 337 Guruve 51 (15.1) 40 (11.9) 2 (0.6) 233 (69.1) 3 (0.9) 3 (0.9) 337 Karanda 1 (0.3) 6 (1.8) 0 (0.0) 326 (95.9) 4 (1.2) 2 (0.6) 340 Shamva 57 (16.9 9 (2.7) 29 8.6) 162 (47.9) 54 (16.0) 8 (2.8) 338 St Alberts 50 (15.9) 3 (1.0) 0 (0.0) 206 (65.6) 27 (8.6) 1 (0.3) 314Mashonaland East Epworth clinic 0 (0.0) 2 (0.6) 0 (0.0) 1(0.3) 336 (98.8) 1(0.3) 340 Hwedza 34 (10.1) 30 (8.9) 0 (0.0) 223 (66.0) 9 (2.6) 1 (0.3) 338 Makumbe 60 (18.1) 5 (1.5) 0 (0.0) 216 (65.1) 42 (12.7) 1 (0.3) 332 Marondera 83 (24.6) 8 (2.4) 0 (0.0) 117 (34.7) 127 (37.7) 2 (0.6) 337 Murehwa 28 (9.3) 58 (19.3) 0 (0.0) 171 (57.0) 7 (2.3) 2 (0.7) 300 Mutoko 14 (4.1) 116 (34.1) 1 0.30) 181 (53.2) 4 (1.2) 3 (0.9) 340 Sadza 18 (5.3) 7 (2.1) 0 (0.0) 122 (36.0) 110 (32.5) 2 (0.6) 339Mashonaland West Banket 236(70.6) 0 (0.0) 0 (0.0) 89(26.7) 1(0.3) 8(2.4) 334 Chinhoyi 29 (8.7) 2 (0.6) 17 5.1) 2 (0.6) 268 (80.0) 3 (0.9) 335 Kadoma 46 (13.7) 7 (2.1) 16 4.8) 7 (2.1) 224 (66.7) 7 (2.1) 336 Kariba 14 (4.1) 0 (0.0) 0 (0.0) 126 (37.2) 195 (57.5) 2 (0.6) 339 Karoi 111 (32.7) 2 (0.6) 0 (0.0) 21 (6.2) 198 (58.2) 1 (0.3) 340Matebeleland North Binga 0 (0.0) 16 (4.8) 0 (0.0) 313 (93.4) 4 (1.2) 2 90.6) 335 Chinotimba 1(0.3) 4 (1.2) 0 (0.0) 28 (8.3) 295 (87.0) 0 (0.0) 339 Inyathi 17 (5.0) 7 (2.0) 49 (14.3) 223 (65.0) 8 (2.3) 4 (1.17) 343 Nkayi 0 (0.0) 45 (17.5) 0 (0.0) 203 (79.0) 1 (0.4) 1 (0.4) 257 Nyamandlovu 63 (18.6) 20 (5.9) 0 (0.0) 165 (42.8) 3 (0.9) 2 (0.6) 338Matebeleland South Beitbridge 5(1.5) 14 (4.2) 1(0.3) 38 (11.4) 252 (75.7) 22 (6.6) 333 Filabusi Hospital 11 (3.3) 41 (12.2) 10 (2.8) 200 (59.5) 5 (1.5) 3 (0.9) 336 Gwanda 8 (2.4) 0 (0.0) 68 (20.1) 63 (18.6) 191 (56.3) 1 (0.3) 339 Plumtree 4 (1.2) 6 (1.8) 0 (0.0) 201 (60) 119 (35.5) 2 (0.6) 335Masvingo Chiredzi 72 (21.7) 3 (0.9) 0 (0.0) 19 (5.7) 178 (53.6) 4 (1.2) 332 Chivi 1(0.3) 32 (9.5) 0 (0.0) 286 (84.6) 7 (2.1) 4 (1.2) 338 Gutu 3 (0.9) 91 (26.8) 0 (0.0) 198 (58.2) 4 (1.2) 0 (0.0) 340 Morgenster 7 (2.1) 6 (1.8) 9 (2.7) 125 (37.0) 183 (54.1) 3 (0.9) 338 Neshuro 1 (0.3) 14 (4.1) 2 (0.6) 265 (77.9) 1 (0.3) 7 (2.1) 340 Ndanga 11 (3.3) 0 (0.0) 0 (0.0) 222 (65.7) 3 (0.9) 3 (0.9) 338 Silveira 2 (0.6) 26 (7.7) 1 (0.3) 296 (87.3) 5 (1.5) 6 (1.8) 339Midlands Gokwe 2 (0.6) 13 (3.9) 0 (0.0) 250 (74.4) 70 (20.8) 1(0.3) 336 Gweru 7 (20.6) 3 (0.9) 0 (0.0) 6 (1.8) 321 (94.4) 0 (0.0) 340 Kwekwe 12 (3.5) 1 (0.3) 22 (6.5) 9 (2.7) 291 (85.8) 0 (0.0) 339 Musume 0 (0.0) 8 (2.4) 1 (0.3) 295(87.02) 4 (1.8) 6 (1.7) 339 Mvuma 11 (3.3) 9 (2.7) 9 (2.7) 146 (43.2) 62 (18.3) 0 (0.0) 338 Shurugwi 11 (3.3) 21 (6.2) 30 (8.9) 216 (64.1) 45 (13.4) 0 (0.0) 337 Zvishavane 6 (1.8) 2 (0.6) 1 (0.3) 39 (11.5) 284 (84.0) 0 (0.0) 338 Total 1522 (8.3) 799 (4.3) 316 (1.7) 7338 (39.8) 7338 (39.8) 158 (0.9) 18,131


Table 4: Distribution of client characteristicsVariables n=18131 % Variables n=18131 % Variables n=18131 %

Age group Level of education Currently on ART

<15 17 0.1 None 113 0.6 Yes 765 4.2

15-19 3508 19.3 Primary 4049 22.3 No 17037 94.0

20-24 5502 30.3 Secondary 13535 74.7 Missing 329 1.8

25-29 4342 23.9 Tertiary 358 2.0 Women offered PMTCT HIV test - Current visit

30-34 2824 15.6 Missing education 76 0.4 Yes 17206 94.9

35-39 1399 7.7 Place of usual residence No 891 4.9

40-44 280 1.5 Growth point 790 4.4 Missing 34 0.2

45-49 22 0.1 Mine 313 1.7 Women accepted PMTCT HIV test - Current visit

<15 17 0.1 Resettlement farm/farm 2451 13.5 Yes 16914 93.3

15-24 9010 49.7 Rural 7209 39.8 No 122 0.7

15-29 9117 50.3 Town or city 6974 38.5 N/A 712 3.9

15-49 18113 99.9 Missing residence 156 0.9 Missing 383 2.1

Missing 18 0.1 Site classification Obstetric Information

Marital status Rural 8594 47.4 Gravida

Married 16754 92.4 Urban 9300 51.3 0 32 0.2Divorced/Separated 127 0.7 Missing 237 1.3 1 5443 30.0

Single 831 4.6 History of GUD in the last 12 months 2-3 8586 47.4

Widowed 43 0.2 Yes 321 1.8 4+ 3674 20.3

Missing 139 0.8 No 17435 96.2 Missing 159 0.9

Type of occupation Missing 217 1.2 Still births

Student 147 0.8 Presence of GUD 0 17289 95.4

Housewife 14130 77.9 Yes 136 0.8 1+ 393 2.2

Unemployed 1526 8.4 No 16976 93.6 Missing 449 2.5

Employed (Formal/Informal) 2289 12.6 Missing 861 4.7 Abortions

Missing 39 0.2 Previously HIV tested and received results 0 16350 90.2

Parity Yes 10115 55.8 1+ 1347 7.4

0 5967 32.9 No 987 5.4 Missing 434 2.4

1-2 8385 46.2 Missing 6871 37.9

3+ 3199 17.6

Missing 23 0.1


Partner information Distribution of clients by Sites

Partner’s place of residence Banket 333 1.8 Distribution of clients Sites continued...

Growth point 107 0.6 Beitbridge 331 1.8 Murehwa 300 1.7

Mine 106 0.6 Bindura 333 1.8 Musume 278 1.5

Resettlement farm/farm 414 2.3 Binga 334 1.8 Mutambara 299 1.6

Rural 792 4.4 Chimhanda 324 1.8 Mutoko 340 1.9

Town or city 1267 7.0 Chinhoyi 332 1.8 Mvuma 332 1.8

Partner’s type of occupation Chinotimba 339 1.9 Ndanga 337 1.9

working 1924 10.6 Chiredzi 319 1.8 Neshuro 331 1.8

not working 779 4.3 Chivi 323 1.8 Nkayi 265 1.5

Student 4 0.0 Concession 334 1.8 Nkulumane 596 3.3Missing type of

occupation 0 0.0 Epworth 340 1.9 Nyamandlovu 334 1.8

N/A 0 0.0 Filabusi 334 1.8 Nyanga 274 1.5

Partner’s Education Level Glenview 333 1.8 Plumtree 334 1.8

None 19 0.1 Gokwe 336 1.9 Rusape 114 0.6

Primary 369 2.0 Guruve 330 1.8 Sadza 339 1.9

Secondary 2250 12.4 Gutu 340 1.9 Sakubva 336 1.9

Tertiary 95 0.5 Gwanda 331 1.8 Shamva 337 1.9

Distribution of clients by Province Gweru 340 1.9 Shurugwi 321 1.8

Bulawayo 596 3.3 Hatcliffe 339 1.9 Silveira 326 1.8

Chitungwiza 600 3.3 Hauna 224 1.2 St Alberts 314 1.7

Harare 1271 7.0 Hwedza 337 1.9 St Marys 600 3.3

Manicaland 1586 8.7 Inyathi 339 1.9 Zvishavane 338 1.9

Mashonaland Central 2311 12.7 Kadoma 332 1.8 Total 18131 100.0

Mashonaland East 2319 12.8 Karanda 339 1.9

Mashonaland West 1672 9.2 Kariba 336 1.9

Masvingo 2314 12.8 Karoi 339 1.9

Matebeleland North 1611 8.9 Kuwadzana 599 3.3

Matebeleland South 1330 7.3 Kwekwe 340 1.9

Midlands 2284 12.6 Makumbe 326 1.8

Marondera 337 1.9

Morgenster 338 1.9

Murambinda 339 1.9


The majority of the women, 74.5% (13495) were in the 15-29 year age group. Nearly all women (93.7%) were married and unemployed (77.9%). The majority of the women (94.0%) had ever received testing for HIV in PMTCT settings. Slightly more than half (56.3%) had received their results in post- test counseling but a significant proportion (38.2%) had missing information on this variable. A very low proportion 4.31 % of women reported having received antiretroviral therapy (Table 4).

About 40.5% of pregnant women stayed in rural areas, while 45.6% of partners resided in towns or cities. Most partners (68.8%) were employed and a high proportion (81.7%) had at least secondary level education.

4.2 Distribution of HIV Prevalence in the fifty-three sentinel sites

Figure 1: Overall HIV pooled prevalence for women 15-49 years, 2012

The overall median HIV prevalence for women 15-49 years was 15.9% (IQR: 12.6%- 19.3%).


Table 5: HIV Prevalence by Sentinel Site, 2012Province Site Total (N) Total

positive Prevalence C.I

Bulawayo Nkulumane Clinic 596 102 17.1 14.1-20.1Chitungwiza St Marys Clinic 600 67 11.2 8.6-13.7

HarareKuwadzana Clinic 599 70 11.7 9.1-14.3Hatcliffe Clinic 339 33 9.7 6.5-12.9Glenview 333 34 10.2 6.9-13.5

Manicaland

Murambinda District Hospital 339 59 17.4 13.3-21.5Mutambara Mission Hospital 299 26 8.7 5.4-11.9Rusape District Hospital 114 25 21.9 14.2-29.6Sakubva Clinic 336 48 14.3 10.5-18.0Hauna District Hospital 224 23 10.3 6.2-14.3Nyanga District Hospital 274 40 14.6 10.3-18.8

Mashonaland Central

Shamva District Hospital/Wadzanayi Clinic 337 53 15.7 11.8-19.6Bindura Chipadze Clinic 333 62 18.7 14.4-22.8St Alberts Hospital 314 26 8.3 5.2-11.3Guruve Hospital/Bepura Clinic 330 38 11.5 8.0-14.9Concession District Hospital / Mvurwi Clinic 334 53 15.9 11.9-19.8Karanda Mission Hospital 339 24 7.1 4.3-9.8Chimhanda District Hospital 324 31 9.6 6.3-12.7

Mashonaland East

Sadza District Hospital 339 50 14.8 10.9-18.5Makumbe Mission Hospital 326 63 19.3 15.0-23.6Hwedza District Hospital 337 58 17.2 13.2-21.3Marondera Hospital /Chihota Hospital 337 65 19.3 15.1-23.5Murehwa District Hospital / Nhowe Mission hospital 300 52 17.3 13.0-21.6Epworth clinic 340 67 19.7 15.5-24.0Mutoko District Hospital 340 52 15.3 11.4-19.1

Mashonaland West

Karoi District Hospital 339 59 17.4 13.3-21.5Kadoma District Hospital 332 50 15.1 11.2-18.9Kariba Hospital/ Nyamhunga Hospital 336 49 14.6 10.8-18.4Chinhoyi Provincial Hospital 332 47 14.2 10.4-17.9Banket District Hospital 333 55 16.5 12.5-20.5

Matebeleland North

Binga District Hospital 334 27 8.1 5.1-11.0Inyathi Mission Hospital 339 69 20.4 16.0-24.6Chinotimba Clinic 339 80 23.6 19.1-28.1Nkayi District Hospital 265 58 21.9 16.8-26.9Nyamandlovu District Hospital 334 82 24.6 19.9-29.2

Matebeleland South

Beitbridge District Hospital 331 74 22.4 17.8-26.9Plumtree District Hospital 334 78 23.4 18.8-27.9Gwanda Provincial Hospital 331 86 26.0 21.2-30.7Filabusi Hospital 334 78 23.4 18.8-27.9

Masvingo

Silveira/Mashoko Mission Hospital 326 41 12.6 8.9-16.2Chiredzi District Hospital 319 53 16.6 12.5-20.7ChiviHospital / Gundu RHC 323 39 12.1 8.5-15.6Gutu Mission Hospital 340 54 15.9 11.9-19.8Morgenster Mission Hospital 338 52 15.4 11.5-19.3Neshuro Hospital 331 43 13.0 9.3-16.6Ndanga Hospital 337 56 16.6 12.6-20.6

Midlands

Mvuma District Hospital 332 64 19.3 15.0-23.4Gokwe District Hospital 336 38 11.3 7.9-14.7Gweru Provincial Hospital 340 66 19.4 15.1-23.6Kwekwe Hospital / Mbizo B 340 50 14.7 10.9-11.5Musume Mission Hospital 278 51 18.4 13.7-22.9Zvishavane District hospital 338 72 21.3 16.9-25.7Shurugwi District Hospital 321 52 16.2 12.1-20.2


There were variations in HIV prevalence in fifty- three sentinel sites ranging from 7.08 (Karanda) to 26% (Gwanda Provincial Hospital. Other sites that had notably high prevalence figures were Nyamadhlovu 24.6%, Filabusi 23.4%, Plumtree 23.4% and Beitbridge 22.4%.

Figure 2: HIV prevalence by sentinel site location, 2012

HIV prevalence among women 15-49 years was slightly higher among women who accessed services in urban sentinel sites (16.1%) as compared to the rural sentinel sites 15.7%.

Figure 3: Median HIV prevalence by province, 2012

The highest HIV prevalence was observed in Matabeleland South 23.8% while the lowest Harare 10.8%.The national median HIV prevalence was 15.9 (Q1:12.6; Q3=19.3)


Figure 4: HIV prevalence by age-group among ANC attendees, 2012

There was a gradual increase in HIV prevalence with age group peaking in the 35-39 year old group followed by a decline. Despite this, the prevalence remained significantly high in the older age groups.

Figure 5: Urban-rural comparison of HIV prevalence among ANC attendees 15-49 years, 2012

Overall, HIV prevalence was highest, 16.2% among women accessing services in urban sentinel sites. HIV prevalence in the rural areas was highest in the 35-39 years age group 27.8% while the HIV prevalence in urban areas was highest in the 40-44 years age group 29.9%.


Figure 6: HIV prevalence by education level, 2012

HIV prevalence was highest among women who had not received any form of education 21.7% (n=113) while it was lowest among those who had tertiary education.

When comparing HIV positive status with marital status we found that the HIV prevalence was lowest among those women who were married (15.5%) as compared to those who were not married (22%).

Figure 7: HIV Prevalence by occupation, 2012


The highest HIV prevalence was among women that reported their occupation as being employed in the informal sector 20.3%, and the lowest was among students 11%.

Figure 8: HIV Prevalence by gravidity among ANC attendees, 15-49 years

There was a corresponding increase in HIV prevalence with increasing gravidity.

Figure 9: HIV Prevalence by number of abortions and still-births, 2012

There was a higher HIV prevalence for women who had 1+ abortions (18.9%) compared to those who had no previous abortions 15.8%. There was a higher HIV prevalence 18.5% among women who had 1+ still births compared to those who had no previous stillbirths 15.9%.


Figure 10: Prevalence of HIV by history and presence of GUD

HIV prevalence among women reporting history or presence of genital ulcer disease (GUD)was almost three times higher than that of prevalence among women without GUD.

Table 6: Distribution of HIV prevalence by exposure to PMTCT services

History Total (N)

Total positive Prevalence C.I

Previously HIV tested and received results Yes 10115 1872 18.5 17.8 -19.3No 987 133 13.5 11.3 - 15.6Missing 6871 845 12.3 11.5 – 13.1

Currently on ART Yes 755 714* 91.6 89.7 - 93.6No 17037 2124 12.5 12.0 - 13.0Missing

Women who accepted PMTCT HIV test - Current visit

Yes 16914 2132 12.6 12.1 - 13.1No 122 74 60.7 51.9 - 69.5N/A 712 583 81.9 79.1 - 84.7

* It is expected that all 755 clients who reported to be on ART should be HIV positive

The majority of women had previously been tested and received results for an HIV test. Among these women, the HIV prevalence was 18.5%. At least 41 patients who reported being currently on ART had an HIV negative result when tested at the Reference laboratory and confirmed by DNA Polymerase Chain Reaction Test.. Only 122 women did not accept an HIV test in their current PMTCT visit.


4.2: HIV prevalence by partners’ characteristics among ANC attendees, 15-49 years in Zimbabwe

Figure 11: HIV prevalence by partner’s occupation

The lowest HIV prevalence was among women whose partners were still students whereas the highest HIV prevalence was among women whose partners were informally employed.

Figure 12: HIV prevalence by partner’s place of residence

The prevalence was highest among women whose partners resided in the mining areas 24.6% and lowest among women whose partners resided in growth points.


Figure 13: HIV prevalence by partner’s level of education

The HIV prevalence among 19 clients who reported that their partners had no primary education was 20.7%. HIV prevalence decreased with higher level of partners’ education.

4.3 Trends in HIV prevalence 2002-2012, for all women 15-49 years in nineteen sentinel sites 2002 -2012

Nineteen sites have contributed to ANC trend data since 2002 (Appendix 4). Trends determined in this section are based on pooled HIV prevalence from the 19 sites. Since 2002, HIV prevalence is based on results from the parallel testing algorithm

Figure 14: Overall HIV prevalence trends

HIV prevalence among all pregnant women 15-49 years declined sharply between 2002 and 2006.


Table 7: Overall HIV prevalence statistics

Year Positive CI2002 25.7 24.7 - 26.72004 21.3 20.4 -22.32006 17.7 17.1 - 18.82009 16.1 15.3 - 17.02012 15.9 15.0-16.9

Between 2009 and 2012, there is no significant decline in HIV prevalence.

Figure 15: HIV prevalence trends by age-group

HIV prevalence was lowest in the 15-19 years age group across all the years in the period 2002 -2012. HIV prevalence generally increased with increase in age up to 35-39 years. HIV prevalence in older women remained significantly high across all the years with a marked increase in prevalence of 44.4% for the age group 45-49 years in 2012.

Figure 16: HIV prevalence trends by level of education


There was a general trend of decline in the HIV prevalence for women who had attained primary, secondary and tertiary education over the years 2002-2012. Although a general trend of decline occurred in HIV prevalence among women without an education from 2002-2006, a sharp increase occurred in 2009 (30%) followed by a marked prevalence decline in 2012 (13.9%).

Figure 17: HIV prevalence trends by occupation

HIV prevalence continued to decline across all years for the employed and housewives, however for the students there was a slight increase in prevalence from 2009- 2012 although a general trend of decline had been observed in the previous years. For the unemployed the HIV prevalence has been showing a swinging trend with the highest prevalence being noted in 2002 (25.6%) followed by 2012 (20.3%).

Figure 18: Trends in HIV prevalence by sentinel site location


HIV prevalence among women 15-49 years was highest among women who accessed services in sentinel sites classified as other across all survey years 2002 -2012. Sites from growth points, border posts, and mining & resettlement farms are classified as ‘other’.

Figure 19: Trends in HIV prevalence by occupation


Figu

re 2

0: H

IV p

reva

lenc

e tr

ends

by

prov

ince

201

2

Sinc

e 20

02, m

ost p

rovi

nces

hav

e ex

perie

nced

a d

eclin

e in

HIV

pre

vale

nce

that

con

tinue

d fr

om 2

002

to 2

012.

Sig

nific

ant d

eclin

es w

ere

at th

e fo

llow

ing

prov

ince

s:

Mas

hona

land

Cen

tral (

25.5

% in

200

2 to

12.

4% in

201

2), M

asho

nala

nd W

est (

27.1

% in

200

2 to

15.

6% in

201

2) an

d M

anic

alan

d (2

5.6%

in 2

002

to 1

3.9%

in 2

012)

. Th

ere

wer

e so

me

incr

ease

s in

prev

alen

ce in

Mat

ebel

elan

d So

uth,

Bul

away

o, M

idla

nds a

nd M

asho

nala

nd E

ast i

n 20

12 c

ompa

red

to 2

009.

4.4

HIV

pre

vale

nce

and

trend

s, w

omen

age

15-

24 y

ears

in n

inet

een

sent

inel

si

tes,

2002

-201

2


Figure 21: Overall HIV prevalence trends by young women (15 -24 years)

There was a continued decline in trends of HIV prevalence among women age 15-24 years from 29.6% in 2002 to 9.85% in 2012.

Table 8: Overall HIV Trends Statistics, young women 15-24 years

Year Percentage HIV positive CI

2002 20.8 19.8 - 21.8

2004 17.4 16.3 - 18.5

2006 12.5 12.3-14.3

2009 11.6 10.6-12.6

2012 9.9 8.2 – 12.0

The decline in HIV prevalence among young women 15 -24 years is statistically significant.


Figure 22: HIV prevalence trends in young women 15-24 years by site

Site 2002 2004 2006 2009 2012Banket 19.9 25 19.3 15.3 13.0Beitbridge 24.5 17.1 19.2 16.3 18.4Bindura 25.3 18.6 11.8 9 16Binga 20 8.2 5.6 5 3.7Chiredzi 24.8 14.4 12.5 21.5 11.5Gutu 16.3 23.6 15.3 10.7 12.2Gwanda 25.7 21 18.9 12.4 22.8Kadoma 17.9 14.6 19.6 10.4 13.2Karanda 14.6 6.9 5.6 6 4.5Kuwadzana 14.8 18 9.3 9.7 8.9Mkoba 17.1 13.6 10.8 9.7 15.7Murambinda 15.6 13.6 11.1 17.7 9.9Musume 18.8 21.6 15.6 12.9 14.4Mutoko 15.2 14.9 11.5 10.2 11.5Nkulumane 21.9 16.2 10.4 9.6 14.7Sadza 35.2 19.2 13.6 8 12.8Sakubva 15.3 13.1 11.7 9.7 13.4St. Mary’s 22 18.8 9.3 8.2 10.3Victoria Falls 30.8 30.5 15.9 19.5 17.4

Across the four survey years and across all ages, HIV prevalence shows a general trend of decline young women except in the 23 year age-group where there is a slight increase from 14.3% to 14.6 %.

Figure 23: HIV prevalence in young women by age-group

HIV prevalence was highest in the 20-24 years age-group.


Figure 24: Trends in HIV prevalence by education level in young women 15-24 years

HIV prevalence increased among women who reported not having any education (26.5 % in 2002 to 33.3% in 2009); however there was a decline in 2012 to 14.9%.There was a general trend of decline in HIV prevalence for women who had attained primary, secondary and tertiary level education.

Figure 25: HIV prevalence trends by sentinel location 15-24 years

There was decline in HIV prevalence among young women residing in both urban and rural areas from the years 2004 to 2009. However there was an increase in prevalence for women residing in both rural and urban areas from 2009 to 2012.


21.9

22

14.8

15.5

20.3

24.9

18.9

20.5

25.8

25.1

18

16.2

18.8

18

13.3

13.2

16.9

19.4

19

19.9

18.9

17.4

10.4

12.7

9.3

11.4

8.9

12.6

19.4

13.9

11

19.1

13.2

9.6

8.1

9.7

13.7

7.5

9.2

12.6

16.8

13.1

14.3

11.5

11.3

7.8

5.8

9.3

7.7

10.2

9.6

9.9

8.1

18.2

9.6

051015202530

Percentage HIV -Positive

Prov

ince

s

Perc

enta

ge H

IV-P

ositi

ve b

y Pr

ovin

ce A

mon

g A

NC

Att

ende

es 1

5-24

yea

rs, Z

imba

bwe

2002

-201

2

2002

2004

2006

2009

2012

Figu

re 2

6: H

IV p

reva

lenc

e tre

nds b

y pr

ovin

ce 1

5 -2

4 ye

ars

Mos

t sen

tinel

site

s (12

/19)

show

ed a

gen

eral

incr

ease

in th

e H

IV p

reva

lenc

e fr

om 2

009

to 2

012.

Not

able

incr

ease

s occ

urre

d in

the

follo

win

g si

tes:

Gw

anda

12.

4%

to 2

2.8

%, N

kulu

man

e 9.

6% to

14.

7% a

nd m

arke

d de

clin

e oc

curr

ed in

Chi

redz

i 21.

5% to

11.

5% a

nd M

uram

bind

a 17

.7%

to 9

.9%

.


4.5 Results of syphilis

Figure 27: Percentage RPR Positive among, women, 15-49 years, and 2012

The Rapid RPR percentage positive was 1.8% among all women. TPHA confirmatory test were not done by sites due to the absence of the testing kit.

Figure 28: Percentage RPR Positive by Age-Group, Among ANC Attendees, 15-49 years, 2012

The prevalence of syphilis rose with increasing age with a peak in 45-49 years age group. The lowest prevalence was found among the 15-19 years age group.


Figure 29: Prevalence of RPR positive by gravidity

Women who had had four or more pregnancies had the highest RPR positive prevalence (2.9%) while those who were in their first pregnancy had the lowest prevalence 1.0%.


5.0 Discussion and ConclusionsThe HIV prevalence among pregnant women 15-49 years has slightly declined from 16.1% in 2009 to 15.9% in 2012 in Zimbabwe. However the difference is not statistically significant. A similar trend was observed in the 15-19 age group where HIV prevalence declined from 6.8% in 2009 to 5.6% in 2012 which may reflect a reduction in the HIV incidence. Though minor this decline in HIV prevalence among female adolescents may suggest changes in sexual behaviors in the 15-19 years age group. Comparison with the findings of the ZDHS 2010-11 we noted that there was a decline in HIV prevalence in Zimbabwe in keeping with the findings of this survey. This decline has been attributed to intensification of HIV prevention strategies and behavioral change in communities at large.

There was a notable increase in the prevalence of HIV in women aged between 35-49 years in 2012 compared to 2009. Similarly the HIV prevalence among those aged 25 years and older was rising in South Africa from 2010 to 2011 peaking in the 35-39 years age group. The increase in this age group could be due to increased fertility in HIV positive women on treatment whose physiological function is improving or reduced adherence to dual contraception by those already infected. The main feature of the HIV Sero-prevalence trends in ANC attendees has been the decline in HIV prevalence in the young pregnant women, with slightly increasing rates of HIV infection in older age groups suggesting change in sexual norms and behaviors in the groups.

Generally provinces in the southern region; Matabeleland South, Matabeleland North, Bulawayo and Midlands had a higher prevalence of HIV than provinces in the northern region. This difference may be due to the high number of mobile populations in the Southern region which increases risk of acquiring HIV. Mobility has been shown to be correlated HIV infection in many countries.

Mobile populations were shown to be at high risk of HIV in a study conducted in nine African countries including Zimbabwe1. An ANC HIV survey done in 2011 in South Africa showed that the prevalence of HIV was high in the Limpopo Province which is on the Zimbabwean border with South Africa. This prevalence rose from 21.4% in 2010 to 22.1% in 20115. Migrants in South Africa were reported to be having multiple sexual partners6.

Those staying in urban areas had a higher HIV prevalence (16.6%) as compared to those in rural areas (15.2%). This is consistent with the previous ANC HIV Sero-survey in 2009 where the prevalence was slightly higher in urban areas compared to rural areas2. A study in Manicaland showed a higher prevalence of HIV among urban dwellers as compared to rural dwellers10.

A similar observation was observed in the Tanzania ANC study where the prevalence was higher in the city clinics, compared to rural clinics3. This difference in prevalence maybe alluded to life style differences between the two regions. Changing protective cultural practices may be a factor in behavioural changes between rural and urban residence. Since the difference is minor it may point towards urbanization of the rural communities.

Lower levels of educations were associated with higher prevalence of HIV in this survey. According to a studies conducted by Michelo et al and Fylkesnes et al higher educational level was associated with reduced risk for HIV infection7, 8. In essence women with higher levels of


education are more likely to delay in sexual debut, to understand HIV prevention strategies better and to make better informed sexual decisions than those with lower or no education.

Married women had a lower prevalence of HIV as compared to those who were not married. The ZDHS of 2010-11 had similar findings9. These findings may be due to reason that women who are not married are more likely to engage in unstable risky behaviours as compared to their married counterparts.

This is in keeping with results of similar ANC surveillance from rural Malawi where being in a stable first marriage appeared to be protective against HIV infection for young women4.These finding are different from the previous similar study done in 2009 which showed that prevalence among married women was not significantly different from those who were not married1 . This may spells the need for specific approaches targeting these women who are not married.

Women who are informally employed had the highest positivity rate. Lifestyle issues which include long separation with spouses may predispose them to HIV. This may make it difficult to target these women who may not be accessible as compared to employed women in one place.

There was no significant difference in the prevalence of HIV among women with the first pregnancy and those with the forth and subsequent pregnancies. This suggests need to strengthen prevention strategies in all women of childbearing age and to promote all women to know their HIV status before falling pregnant.

The study revealed those who were HIV positive were more likely to have had a history of miscarriage and still births than those who were HIV negative. This relationship between HIV and bad obstetric outcomes has been shown in other studies. In contrast an ANC study in Tanzania stillbirth was independent of HIV status3.

The positive correlation between STIs and HIV is further strengthened by findings in this study. Genital Ulcer Disease has been shown to predispose one to HIV hence the need for health workers to fully examine pregnant women at every visit to exclude genital ulcer disease and to strengthen STI prevention strategies.

In conclusion the prevalence of HIV has shown a decline in 2012 as compared to previous years. High prevalence of HIV was found among women who were older than 25 years, living in the southern provinces, living in urban areas, with lower levels of education, informally employed and with history of STIs.

5.1 Lessons Learnt

1. Training all health workers working in FCH and ANC on methodology of the survey enables continuity of the study in the face of staff rotation and absenteeism.

2. Lack of or inadequate cascading of clear policy on HIV testing and retesting guidelines in ANC might leave some facilities retesting women with a known HIV status at booking.


6.0 Recommendations1. Intensify health education on behavioural change targeting older women for safer sexual

practices – District Health Promotion Officers (DHPO).

2. There is need for strengthening of HIV prevention strategies among women of childbearing age with special emphasis on those; 35– 39 age group, women employed in informal sector, lower educational status and those with genital ulcer disease, i.e. high risk groups - DHPO.

3. There is need to training health workers on site on how the survey would be conducted. This allows all staff on site to be trained thus enabling continuity of survey in the absence of the key staff and also consistence of data – Data Quality Specialist, Surveillance Officer.

4. Further research to determine factors affecting the prevalence of HIV in older women – Public Health Officers.


References1. National survey of HIV and syphilis prevalence among women attending antenatal clinics

in Zimbabwe, 20092. Changalucha J,Kimaro D, Kumogola Y et al, 2010. Trends in HIV & syphilis prevalence and

correlates of HIV infection: results from cross- sectional surveys among women attending ante-natal clinics in Tanzania. BioMed Central Public Health, 10-553

3. Crampin A, Jahn A, Kondowe M et al, 2008. Use of ANC surveillance to assess the effect of sexual behaviour on HIV prevalence in young women in Karonga district, Malawi. Journal of Acquired Immuno Deficiency Sydromes; 48:196-202

4. Harriss K, Marston M, Slaymaker E, 2008. Non-response bias in estimates of HIV prevalence due to the mobility of absentees in national population-bases surveys: a study of nine national survey. Sex Transmission Infection;84(suppl 1): i71-i77

5. Dept of Health. The 2011 National Antenatal Sentinel HIV and Syphilis Prevalence Survey in South Africa. 2012

6. Hargreaves JR, Bonell CP, Morison LA, Kim JC, Phetla G, Porter JDH, Watts C, Pronyk PM. Explaining continued high HIV prevalence in South Africa: socioeconomic factors, HIV incidence and sexual behaviour change among a rural cohort, 2001-2004. AIDS. 2007; 21(S7): S39-48

7. Michelo C, Sandoy IF, Fylkesnes K. Marked HIV prevalence declines in higher educated young people: evidence from population-based surveys (1995-2003) in Zambia. AIDS. 2006; 20: 1031-1038

8. Fylkesnes K, Musonda RM, Sichone M, Ndhlovu Z, Tembo F, Monze M. Declining HIV prevalence and risk behaviours in Zambia: evidence from surveillance and population based surveys. AIDS. 2001; 15: 907-916

9. Zimbabwe National Statistics Agency (ZIMSTAT) and ICF International. Zimbabwe Demographic and Health Survey 2010-11. 2012. Calverton, Maryland, ZIMSTAT and ICF International Inc.

10. Marsh KA, Nyamukapa CA, Donnelly CA, Garcia-Calleja, Mushati P, Garnett GP, Mpandaguta E, Graccly NC, Gregson S. Monitoring trends of HIV prevalence among young people, aged 15 to 24 years, in Manicaland, Zimbabwe. Journal of the International AIDS Society. 2011; 14: 27


7.0 APPENDICESAppendix 1: 2012 Sentinel Survey Sites Province and Sentinel Site Classification

Province District SiteSite classification Rural/Urban

Bulawayo Bulawayo Nkulumane Clinic Urban

Chitungwiza Chitungwiza St Marys Clinic Urban

Harare Harare

Kuwadzana Clinic Urban

Hatcliffe Urban

Glenview Urban

Manicaland

Buhera Murambinda Hospital Rural

Chimanimani Mutambara Hospital Rural

Makoni Rusape Hospital urban

Mutare Sakubva Clinic urban

Mutasa Hauna Hospital rural

Nyanga Nyanga Hospital rural

Mashonaland Central

Shamva Shamva Hospital/Wadzanayi Clinic rural

Bindura Bindura Chipadze Clinic urban

Centenary St Alberts Hospital rural

Guruve Guruve Hospital/Bepura Clinic rural

Mazowe Concession Hospital/Mvurwi clinic rural

Mt Darwin Karanda Mission Hospital rural

Rushinga Chimhanda District Hospital rural

Mashonaland East

Chikomba Sadza District Hospital rural

Goromonzi Makumbe Mission Hospital rural

Hwedza Hwedza District Hospital rural

Marondera Marondera Provincial Hospital/Chihota Hospital urban

Murehwa Murehwa District Hospital/ Nhowe Mission hospital rural

Seke Epworth clinic urban

Mutoko Mutoko District Hospital rural

Mashonaland West

Hurungwe Karoi District Hospital urban

Kadoma Kadoma District Hospital urban

Kariba Kariba Hospital/ Nyamhunga Hospital urban

Makonde Chinhoyi Hospital urban

Zvimba Banket District Hospital rural


Matebeleland North

Binga Binga District Hospital rural

Bubi Inyathi Mission Hospital rural

Hwange Chinotimba Clinic urban

Nkayi Nkayi District Hospital rural

Umguza Nyamandlovu District Hospital rural

Matebeleland South

Beitbridge Beitbridge District Hospital urban

Bulilima/Mangwe Plumtree District Hospital urban

Gwanda Gwanda Provincial Hospital urban

Insiza Filabusi District Hospital rural

Masvingo

Bikita Silveira/Mashoko Hospital rural

Chiredzi Chiredzi District Hospital urban

Chivi Chivi Hospital / Gundu RHC rural

Gutu Gutu Mission Hospital rural

MasvingoMorgenster Hospital rural

Masvingo Provincial Hospital urban

Mwenezi Neshuro Hospital rural

Zaka Ndanga Hospital rural

Midlands

Chirumanzu Mvuma District Hospital rural

Gokwe Gokwe District Hospital urban

Gweru Gweru Provincial Hospital urban

Kwekwe Kwekwe Hospital/Mbizvo B urban

Mberengwa Musume Mission Hospital rural

Zvishavane Zvishavane District Hospital urban

Shurugwi Shurugwi Hospital urban


Appendix 2: 2012 ANC Surveillance Form



Appendix 3: Laboratory HIV Parallel Testing Algorithm

Concordant(Agreement)

Discordant(Disagreement)

Western Blot

Polymerase Chain Reaction

REPORT

Concordant(Agreement)

Indeterminate

2 EIA Kits


Appendix 4: ANC Sentinel Surveillance Sites Since 2002

Site 2002 2004 2006 2009 2012Banket 19.9 25 19.3 15.3 13.0Beitbridge 24.5 17.1 19.2 16.3 18.4Bindura 25.3 18.6 11.8 9 16Binga 20 8.2 5.6 5 3.7Chiredzi 24.8 14.4 12.5 21.5 11.5Gutu 16.3 23.6 15.3 10.7 12.2Gwanda 25.7 21 18.9 12.4 22.8Kadoma 17.9 14.6 19.6 10.4 13.2Karanda 14.6 6.9 5.6 6 4.5Kuwadzana 14.8 18 9.3 9.7 8.9Mkoba 17.1 13.6 10.8 9.7 15.7Murambinda 15.6 13.6 11.1 17.7 9.9Musume 18.8 21.6 15.6 12.9 14.4Mutoko 15.2 14.9 11.5 10.2 11.5Nkulumane 21.9 16.2 10.4 9.6 14.7Sadza 35.2 19.2 13.6 8 12.8Sakubva 15.3 13.1 11.7 9.7 13.4St. Mary’s 22 18.8 9.3 8.2 10.3Victoria Falls 30.8 30.5 15.9 19.5 17.4


New AIDS & TB PROGRAMME 2013catalogue.safaids.net/sites/default/files/publications... · 2014. 5. 12. · AIDS & TB PROGRAMME 2013 v Foreword The 2012 ANC sentinel surveillance report

Documents