ORIGINAL ARTICLE Medical Journal of Zambia, Vol. 41, No. 2: 81-85 (2014) Awareness of Repeat Antenatal HIV Testing in Moers Six Weeks Postnatal in Lusaka, Zambia A. Mtaja, B. Amadi, & S. Wasomwe Department of Paediatrics and Child Heal University Teaching Hospital Lusaka Zambia ABSTRACT Zambia national guidelines on e Prevention of Moer to Child Transmission (PMTCT) recommend at all pregnant women and breastfeeding moers must have 1 HIV tests every 3 mons. However, less an 10% of 2 pregnant women in Lusaka District get retested. Repeat HIV testing identifies women who seroconvert after e first test and allows measures to reduce moer to child transmission to be instituted. Objectives: This cross-sectional survey aimed at investigating e effect of awareness of repeat testing on actual retesting in women attending e 6 week postnatal clinic at Chilenje Heal Centre in Lusaka. The objectives of e study were to determine e proportion of women at were aware of e repeat antenatal HIV testing, e number of women at were retested later in pregnancy or labour, and e number of postnatal moers who seroconverted during e study. Meods: Questionnaires were used to assess awareness and moers eligible for a repeat test were offered a repeat test. 404 moers at e six week postnatal visit were recruited by convenient sampling. Only women at had proof of a negative HIV test result in pregnancy were included in e study. Data was stored on Epidata and analysed using stata. Chi squares were used to make associations between e categorical variables and e primary outcome, repeat testing. Multivariate logistic regressions were used to adjust for confounders. Results: Seventy two percent of e women were aware of e importance of repeat testing but only 36% received a repeat test. None of e women seroconverted during e study. Awareness was significantly associated wi repeat testing (Odds ratio 3.8; 95% CI 2.9 6.9). Conclusion: Women at are aware of repeat antenatal HIV testing are four times more likely to receive a repeat test an ose at are not. Booking in e first trimester increases e chance of being retested two fold. Women at have 5 or more ANC visits have a ree-fold chance of being retested an ose wi fewer visits. INTRODUCTION Moer to child transmission of HIV is e major source of 3 HIV infection in children. An HIV infected child represents a missed opportunity for prevention. The retesting rates of less an 10% noted in clinics in Lusaka District are too low to translate into an effective PMTCT programme at will ensure zero vertical transmission of HIV. Wi ese low retesting rates, women at seroconvert will be missed and pose a high risk of infection to e unborn child. Barriers to uptake of 3-6 PMTCT services were numerous in various studies. Of note were lack of discussion by antenatal care providers and lack of awareness of existing services by clients. It is important to investigate e effect of awareness on repeat testing as is will identify some of e issues related to non implementation of e National Guidelines and allow for appropriate intervention. Meodology Research Design : This was a cross-sectional study Target Population: All postnatal moers attending Maternal Child Heal Clinic at Chilenje Clinic, Lusaka Study Population: Postnatal moers who previously tested negative for HIV, coming for e 6 week visit and have met e eligibility criteria. Study Site: The study will be conducted at Chilenje Clinic, at its Maternal and Child Heal department, in 81
Lusaka, Zambia. Chilenje Clinic is located in e Soueastern part of Lusaka. It covers a catchment area of about 77,142 residents. It offers out patient as well as in patient services (bed capacity is 30). It has e following clinics: Maternal and Child Heal and Antiretroviral clinics. It also has several 'corners' at cater for patients wi sexually transmitted diseases and Tuberculosis. It has a you friendly corner to cater for e you. It has a labour ward at has a capacity of 4 beds. It has a basic laboratory and also has a pharmacy. It keeps its relevance in e community via e community heal Workers who are community volunteers. It has an ambulance to facilitate transfer to higher level hospitals. Until now, Chilenje has operated as a Clinic, but is now earmarked for a first level Hospital where Caesarean Sections and oer minor surgical operations will be 7 done. Sample size and Sampling Procedure 404 moers at e 6 week postnatal visit were recruited by convenient sampling. Only women at had proof of a negative HIV test result in pregnancy were included in e study. Research Instruments Questionnaires were used to assess awareness and moers eligible for repeat testing were offered e test. Research Meods Moers were invited to e study as ey came for eir 6 week post natal clinic visit. Consent was obtained and questionnaires administered to assess awareness and oer parameters. Moers eligible for repeat testing were counselled and tested. Testing was done as described in e Zambia HIV rapid test Algorim. Post-test counselling was also done. Data Analysis Data was stored on Epidata v 3.1 and analysed using stata 11. Chi square tests were used to associate categorical variables wi e primary outcome (repeat testing). Multivariate logistic regressions were used to adjust for confounders. Associations were considered significant at 95% confidence interval if p<0.05. Effect size was measured using odds ratios. Study area The study was limited to Chilenje Heal Centre due to logistic challenges. RESULTS 163 were aware and not retested/97 were not aware and not retested (bo groups (total 260) were offered a repeat test) Pre - test counseling 288 HIV - uninfected result Post - test counseling Figure 1: summary of findings: Flow chart of results 410 Postnatal moers who tested HIV negative during antenatal care seen in MCH clinic were approached about e study 404 Accepted 289 women were aware of repeat antenatal HIV testing 288 HIV antibody tests done in e study (260 of ose not retested antenatally Plus 28 who were retested antenatally but were eligible for a repeat test at time of study) 6 Declined 144 were aware and retested antenatally HIV - infected result: seroconversion : Zero Table 1: Characteristics of study participants; n = 404 Variable Age Mean/percentage 26.3 years Education Secondary 56% Tertiary 29% Employment Unemployed 63% Formal 25% Self 12% Income KR 2000 41% KR 1000 32% Residence Medium cost - 80% High cost 13% Marital status Married 82% Single 18% Parity Multiparous 59% Uniparous 35% Grandmultiparous 6% Booking 2 nd Trimester 76% 1 st Trimester 16% 3 rd Trimester 8% # routine ANC visits 3 to 4 63% 1 to 2 16% 5-21% Place of delivery Clinic 54% Hospital 41% Home 5% Table 2 Outcomes Outcome Frequency Percentage (%) Awareness Aware 289 72% Not aware 115 28% Repeat testing Received retest 144 36% No retest 260 64% Seroconversion nil Nil 82
Table 3 shows how awareness of repeat testing was strongly associated wi actual repeat testing Repeat HIV Test Odds ratio P value 95% conf. Interval awareness 4.16 <0.0001 2.39 7.25 Table 4 - Results after adjusting for confounders Variable Odds ratio Awareness Potential confounding variable 1) Education - Primary 0.40 P value 95% confidence interval - Secondary 0.83 - Tertiary 2.11 0.001 1.34 3.34 2) Gestational age at booking - First trimester 3.3 <0.0001 1.87 5.81 - Second trimester 0.45 - Third trimester 0.7 3) Number of ANC visits - 1 to 2 visits 0.53-2 to 4 visits 0.50 <0.0001-5+ visits 3.8 2.29 6.33 Table 8 - Multivariate analysis rhiv = repeat HIV testing; aw = awareness of repeat testing; ft = first trimester; av5 = 5+ ANC After adjusting for confounders, awareness of repeat antenatal HIV testing remained significantly associated wi repeat testing ( OR 3.89, p < 0.0001; CI 2.19 6.90). Booking in e first trimester and having five and more ante natal clinic visits were independently significantly associated wi repeat testing as shown above. DISCUSSION This study investigated e effect of awareness of repeat antenatal HIV testing on actual repeat testing. Seventytwo percent of women were aware of e importance of repeat antenatal HIV testing while only 36% actually received a repeat test antenatally. All women had received at least one HIV test during pregnancy, and are assumed to have been counselled prior to e test. Therefore, all women were expected to be aware of repeat HIV testing. The women at reported being unaware of e importance of repeat testing were women at probably received inadequate counseling as was seen in e study by Gita et al who found at lack of discussion by antenatal care providers was a major barrier 8 to HIV testing during pregnancy. Unfortunately, not a single counseling session was listened to in order for e counseling to be assessed. However, it was observed at e PMTCT guidelines were present in e clinic and at all but 2 nurses were trained in PMTCT. Women who are aware were 4 times more likely to be retested an ose who were not (p <0.0001, CI 2.39 7.25). It is possible at ese women used e knowledge ey had on e importance of repeat testing to request e HIV test. The results show at an increase in education, five or more antenatal visits and booking in e first trimester are significantly associated wi being aware of repeat HIV testing. This makes sense because more contact wi a heal care provider ensures reinforced information. A higher number of ANC visits increased e chances of being retested 3-fold. This is comparable to Rouzioux et al and Newell et al who found e converse at poor antenatal attendance translated into low rates of repeat 9,10 testing. After controlling for confounding factors, awareness still remained significantly associated wi repeat HIV testing (OR 3.11; p<0.0001, CI 2.39 7.25). This is line wi what Gita et al found where lack of awareness of services offered at e ANC contributed significantly to e low 8 uptake of ese services. Therefore e converse is true at awareness increases uptake as was found in is study. The study found at 36% of e women were retested during pregnancy. This is higher an e Lusaka district average of 10%. The data was collected over a period of six mons and during is time e heal care providers were sensitized to e study aim which was to actively repeat test e women. This result, erefore, could be a reflection of a ripple effect on oer departments of e 83
clinic like e labour ward. A similar effect was seen by Keiffer et al who observed retesting at 45% in eir 11 intervention sites compared to 14% at e control sites. In is study, e major reason (in 73% of e participants) for not retesting was at repeat testing was not offered to e women by e heal care providers. This was a barrier to a successful PMTCT programme also found by Kinuia et al, who cite a strain on an over-stretched 12 human resource base. Interestingly, unavailability of counsellors, counsellors being too busy, counsellors being rude, moers being told to come back wi partner and long queues at e ANC were not major issues in being reasons why moers did not retest, wi ese parameters accounting for e 'oer' category (7%) of e responses. It was observed, however, at ere were at most 4 nurses in e maternal Child heal clinic for e various activities offered at e Clinic, six short of e ideal. The clinic receives about 40 antenatal and 60 post natal women daily. No posters on PMTCT and retesting were seen on e walls of e clinic. The posters on e walls were ose concerning Child Heal programmes. None of e 292 moers at were retested seroconverted. The study was not powered to detect seroconversion. Also of note is e low prevalence of 7 HIV (12%) in antenatal moers at Chilenje clinic. This is much lower an at found at oer clinics where e prevalence of HIV in pregnant women ranges between 16 to 21%. Nearly all (99%) of e women offered repeat testing during e study accepted e test. This shows at refusal of repeat testing does not contribute to its low levels. Once e test is offered, e probability at a woman will refuse e test is almost nil. CONCLUSION Women at are aware of repeat antenatal HIV testing are four times more likely to receive a repeat test an ose at are not. Booking in e first trimester increases e chance of being retested two fold. Women at have 5 or more ANC visits have a ree-fold chance of being retested an ose wi fewer visits. LIMITATIONS The results obtained from is study may not be generalizable to all clinics in Lusaka as ere may be factors at are specific to Chilenje Clinic. For example, e study found at 29% of e study population attended tertiary education, 82% were married, 80% lived in 'middle class' housing and 95% delivered in a heal centre. This is certainly not representative of e population in Lusaka. Recall bias was e second limitation since moers were asked about an event at had occurred in e past. Several factors can influence maternal awareness of repeat antenatal HIV testing. Bo Maternal and Heal facility based factors should have been explored. The study explored mostly maternal factors, and e data presented is mostly quantitative. The omission of collecting data on heal facility related issues e.g listening in to a counseling session by a PMTCT qualified counselor, interviews wi heal centre staff and e omission of qualitative data from e moers e.g focus group discussions missed out on a richer data set at would have produced a more robust recommendations for policy and practice. RECOMMENDATIONS Counseling in e antenatal period must be strengened to ensure at women are empowered to be able to ask for a repeat test when not offered. Campaigns at encourage early booking and a minimum of four ANC visits as recommended by MOH should be intensified and implemented in all heal facilities. PMTCT guidelines which emphasize repeat testing should be adhered to in all antenatal clinics to ensure full benefit of e programme whose aim among oers is to reduce HIV infection in children. The Ministry of Heal should critically and periodically look at e staffing levels in e local clinics to ensure at e staff are not overwhelmed and stretched to e point where ey compromise service delivery to e community, oerwise we will begin to see a reversal of e gains in PMTCT. 84
ACKNOWLEDGEMENTS This research would not have been possible wiout e support of The Center for Infectious Disease Research in Zambia and e Vanderbilt Institute rough e UNZA- VU Capacity Building Project. I would like to ank my supervisors Dr S. Wa Somwe and Dr B. Amadi for eir guidance. I am also indebted to Dr B. Andrews, Dr V. Mulenga and Dr C. Chabala for eir invaluable help. Thank you for your patience and for allowing me to disturb your work at any time. I ank my hard working assistants Sr P. Sipatonyana and Sr C. Chipanda of Chilenje Clinic. I acknowledge all e faculty in e Department of Paediatrics and Child heal as well as my colleagues for eir input. Last but not least, a special anks to all e study participants who made is research possible. REFERENCES 1. Zambia National Protocol Guidelines for e Integrated Prevention of Moer-to-Child Transmission of HIV, 2010 2. Zambia Electronic Perinatal record system 3. Prevention of Moer-to-Child transmission of HIV: Expert Panel Report and recommendations to e US. Global AIDS Coordinator January 2010 4. Bwirire LD, Fitzgerald M, Zachariah R, et al. Reasons for loss to follow up among moers registered in a Prevention-of-Moer-to-Child Transmission Program in Rural Malawi. Trans R Soc Trop Med Hyg 2008; 102: 1195 1200 5. Medley A, Garcia-Moreno C, McGill S, et al. Rates, Barriers and outcomes of HIV serostatus disclosure among women in developing countries; Implications for Prevention of Moer-to Child Transmission Programs. Bull World Heal Organ 2004:82: 299 307. 6. Kebaabetswe PM. Barriers to Participation in e prevention of Moer-to-Child HIV transmission Program in Gaborone, Botswana, a qualitative approach. AIDS Care. 2007 Mar: 19 (3): 355 60. 7. Chilenje Clinic Records and Interview wi Chilenje Clinic Sister-in-Charge. 8. Gita S, Ashock D, Manisha K. Low Utilization of HIV testing During Pregnancy: What Are e Barriers to HIV testing for women In Rural India? Journal of Acquired Immune Deficiency Syndrome. 1 February 2008 Volume 47 Issue 2 pp 248 252. 9. Rouzioux C, Costagliola D, Burgard M et al.: Timing of moer-to-child HIV-1 transmission depends on maternal status. The HIV Infection in Newborns French Collaborative Study Group. AIDS7(Suppl. 2),S49-S52 (1993). 10. Newell ML, Coovadia H, Cortina-Borja M, et al. Mortality of infected and uninfected infants born to HIV-infected moers in Africa. A pooled analysis. Lancet 2004; 364:1236 43. 11. Kieffer M, Hoffman H, Nlabhats B, et al. Repeat HIV testing in labor and delivery as a standard of care increases ARV provision for women who seroconvert during pregnancy. Program and abstracts of e 17 Conference on Retroviruses and Opportunistic Infections (CROI); February 16-19, 2010; San Francisco, California. 12. Kinuia J, Kiarie J, Farquhar C, et al. Co-factors for HIV incidence during pregnancy and e postpartum period. Program and abstracts of e 17 Conference on Retroviruses and Opportunistic Infections (CROI); February 16-19, 2010; San Francisco, California. 85