Improving Retention in the Early Infant Diagnosis of HIV Program in Rural Mozambique by Better Service Integration

Ciampa, Philip J MD, MPH*†; Burlison, Janeen R MPH, MSN, WHNP-BC*; Blevins, Meridith MS*; Sidat, Mohsin MD, PhD‡; Moon, Troy D MD, MPH*§; Rothman, Russell L MD, MPP†; Vermund, Sten H MD, PhD*

JAIDS Journal of Acquired Immune Deficiency Syndromes: 1 September 2011 - Volume 58 - Issue 1 - pp 115-119
doi: 10.1097/QAI.0b013e31822149bf
Brief Report: Epidemiology and Prevention

Low mother/infant retention has impeded early infant diagnosis of HIV in rural Mozambique. We enhanced the referral process for postpartum HIV-infected women by offering direct accompaniment to the location of exposed infant testing before discharge. Retrospective record review for 395 women/infants (September 2009 to June 2010) found enhanced referral was associated with higher odds of follow-up (adjusted odds ratio = 3.18, 95% confidence interval: 1.76 to 5.73, P < 0.001); and among those followed-up, earlier infant testing (median follow-up: 33 days vs. 59 days, P = 0.01) compared with women receiving standard referral. This simple intervention demonstrates benefits gleaned from attention to system improvement through service integration without increasing staff.

From the *Vanderbilt Institute for Global Health, Nashville, TN; †Department of General Internal Medicine and Public Health, Vanderbilt University, Nashville, TN; ‡Department of Community Health, Universidade Eduardo Mondlane, Maputo, Mozambique; and §Friends in Global Health, LLC, Maputo, Mozambique.

Received for publication January 5, 2011; accepted April 22, 2011.

The HIV/AIDS Treatment and Care program in Zambézia is supported by President's Emergency Plan for AIDS Relief through the CDC Global AIDS Program (grant U2GPS000631). Dr P.J.C. was supported by the Veteran's Affairs Quality Scholars Fellowship Program.

Funders had no role in study design, data collection, analysis, decision to publish or preparation of this article.

The authors have no conflicts of interest to disclose.

Correspondence to: Philip J. Ciampa, MD, MPH, Vanderbilt Institute for Global Health, 2525 West End Ave. Suite 725 Nashville, TN 37203 (e-mail:

Article Outline
Back to Top | Article Outline


Development of a comprehensive strategy to reduce mother-to-child transmission of HIV has the potential to reduce perinatal HIV transmission from a rate of 25% to 5% or less.1-4 Despite successes in many venues, advances have been slow to translate to many field conditions. Experiences with prevention of mother-to-child transmission (PMTCT) programs in many developing countries have fallen short of the results seen in structured trials.5-8 For example, the 2010 PEARL study suggested variable, substantial, and diverse programatic dysfunction in 43 facilities in 4 countries, resulting in an overall rate of prophylactic coverage of 51% for HIV-infected women and exposed infants.9 Similar to PMTCT challenges, attrition of HIV-exposed infants is a significant limitation for programs in many settings, interrupting opportunities for prompt diagnosis of infected infants and for measures to prevent late transmission that require continuity of care.10-13 These measures are even more important in settings where prophylactic coverage is limited. Factors such as the lack of integrated services and poor quality of patient counseling are often cited as important limitations to providing ideal care for HIV-exposed infants.1,5-7,9,10,12,14,15 Development and implementation of effective interventions to address these barriers is an explicit priority of the Global Health Initiative and the President's Emergency Plan for AIDS Relief.16,17

Similar to the experience of others working in sub-Saharan Africa,10-13 the success of the early infant diagnosis (EID) of HIV program in Mozambique has been hampered by maternal/infant program attrition. In a 2009 study, we examined barriers to follow-up at one clinical site in rural Mozambique.18 Only 25% of HIV-infected mothers who gave birth at the health center brought their infants back to the EID service for HIV screening, though nearly 50% followed up for their own HIV care; women receiving combination antiretroviral therapy (cART) were more likely to return with their infant for EID.18 In response to these worrisome program indicators for postpartum HIV-infected mothers and their HIV-exposed infants, we initiated a pilot study to examine operational barriers to follow-up for EID and to implement interventions targeted to improve mother/infant retention. In this report, we evaluate the effectiveness of one such intervention designed to improve retention in rural Mozambique. We hypothesize that a structural intervention to enhance referral to EID will result in a higher proportion of HIV-exposed infants who return for HIV testing and continued monitoring.

Back to Top | Article Outline


Study Context

Women presenting for prenatal care at 2 rural district-level hospitals in Zambézia Province are offered routine HIV screening.19,20 Women who are HIV infected are evaluated for either PMTCT prophylaxis or cART when eligible per national guidelines (Table 1).21 All HIV-infected women are advised to bring their infants for HIV screening by dried blot spot polymerase chain reaction at 1 month of age. At both study hospitals, prenatal, maternity and infant diagnosis services take place at 3 different sites within the same hospital grounds. Standard postpartum counseling of HIV-infected women takes place in the multibed maternity ward by maternity nurses and includes recommendations to return for infant HIV screening and to exclusively breastfeed their infant. A medical record for the infant is generated at the time of infant follow-up (≥1 month of age).

Back to Top | Article Outline

Study Design and Population

Using a retrospective cohort analysis of HIV-infected women/HIV-exposed infant pairs identified through clinic record review, we performed a pilot study in the districts of Alto Molócuè and Namacurra to improve infant retention in the EID program. All HIV-infected women who gave birth to a live infant at a study hospital from September 2009 until June 2010 were included along with their infants. Two cohorts of mother/infant pairs were identified as follows: (1) pairs receiving standard referral to EID at the time of discharge from labor and delivery and (2) pairs receiving the study intervention of enhanced referral. The primary study outcome is the proportion of mother/infant pairs who followed up for EID within 3 months of birth. The number of days to the initial infant EID visit is a secondary outcome.

Back to Top | Article Outline

Study Intervention

All postpartum HIV-infected women were eligible to be offered enhanced EID referral by maternity nurses before discharge. Unlike those who received standard postpartum referral, HIV-infected mothers who received enhanced referral were directly accompanied by a maternity nurse before discharge to the location of EID services within the hospital grounds. In the privacy of the infant testing suite, a mother received HIV-related counseling by a maternal-child health nurse. Although the content of counseling provided by staff during enhanced referral was similar to that given with standard referral, the delivery was enhanced by the privacy of the EID suite. In contrast to the standard referral process, all infants who received enhanced referral had a medical record generated before discharge. The distance between the maternity and EID services at both hospitals was less than 50 meters. We estimate that the process of enhanced referral required an additional 5-10 minutes to complete when compared with standard care due to the time required to offer and complete direct accompaniment to EID services. No additional staff members were hired to deliver enhanced referral.

Back to Top | Article Outline

Data Analysis

We reviewed mothers' charts and abstracted relevant data. Receipt of study intervention was ascertained by review of infant medical records and defined as a record that was initiated at fewer than 7 days of life, indicating that the mother/infant pair had received direct accompaniment to EID services at the time of maternity discharge. Maternal/child follow-up for EID was determined by review of the maternity and EID service rosters. Mother/infant pairs from both registries were matched by maternal name and verified with infant date of birth. An infant was considered to have been followed-up if the visit occurred at ≤3 months of age; this proportion represented the primary outcome. For those infants presenting for EID services, time to the initial infant EID visit (regardless of duration) was calculated and comprised a secondary outcome. Basic maternal demographic data were abstracted from maternity records. Univariate and multivariable regression models were used to compare the odds of follow-up within 3 months of birth for those that received enhanced versus standard referral; covariates were chosen a priori and included the number of children in the family, maternal occupation, distance from maternal home to the hospital, maternal receipt of cART, infant month of birth, and study site. Missing values for these covariates were accounted for using multiple imputation techniques. In this technique, we use other mother/infant characteristics to predict the missing value for each patient, and then we perform regression on the imputed data. We repeated this task 25 times and took the average estimates from multiple logistic regressions.22 For infants presenting for EID, time from birth to any initial infant EID visit was compared for those receiving enhanced versus standard referral by Wilcoxon rank sum. All analyses were done using the STATA statistical software package v11 (College Park, TX). The National Committee of Bioethics for Health in Mozambique and the Institutional Review Board at Vanderbilt University approved this research.

Back to Top | Article Outline


In our 2 sites, 395 HIV-infected women gave birth to a live infant from September 2009 to June 2010 (262 in Namacurra, 133 in Alto Molócuè); all were included in analyses. Of these, 332 women/infant pairs (84.1%) received standard referral and 63 pairs (16.0%) received enhanced referral. The median maternal age (interquartile range) was 22 years (19-26) and most had 1 or 2 children; 75.1% of women had income solely from subsistence agriculture and 87.5% of women lived within 10 kilometers of the hospital. Only 15.7% of women received cART to treat HIV infection as per national guidelines. (350 CD4+ cells/μL threshold). A higher proportion of women who had received enhanced referral were on cART compared with those who had received standard referral (23% vs. 14%, P = 0.09); otherwise, the groups were similar across measured demographic characteristics (Table 2).

EID was sought for 25.6% of the infants in the standard referral and 54.0% of those in the enhanced referral groups. Compared with those women who had received standard referral, women who received enhanced referral had increased odds of successful follow-up for infant EID [unadjusted odds ratio (OR): 3.41, 95% confidence interval (CI): 1.96 to 5.92, P < 0.001]. This association was maintained after adjusting for potential confounders, including maternal occupation, number of children, distance from the hospital, and maternal cART status (adjusted OR: 3.18, 95% CI: 1.76 to 5.73, P < 0.001) (Table 3). In the multivariable model, there were also independent associations between successful follow-up and increasing family size (per child, OR: 1.21, 95% CI: 1.03 to 1.46, P = 0.05) and decreased distance from the hospital (< vs. ≥10 km, OR: 2.54, 95% CI: 1.06 to 6.05, P = 0.04). There was a trend toward an independent association between maternal occupation of student and successful follow-up (student vs. agriculture, OR: 2.02, 95% CI: 0.96 to 4.29, P = 0.07). There was no independent association between maternal cART status, month of infant birth, or study site with the likelihood of infant follow-up for EID in this sample (data not shown). For those infants with an initial EID visit at any point during the study (n = 145), the median time from birth to follow-up (interquartile range) was 59 days (32-86.5) for those who received standard referral and 33 days (30-63) for those that received enhanced referral (reduction of 26 days, P = 0.01).

Back to Top | Article Outline


In 2 rural district hospitals, a simple intervention to enhance the process of discharge referral of HIV-exposed infants was independently associated with a tripling of the odds of follow-up for EID within 3 months. In addition, women who received enhanced referral more often returned with their infant at the ideal time for EID (≈30 days of life). Results from this study suggest that attention to the process of health care delivery may allow for development of targeted interventions to improve PMTCT/EID program outcomes such as patient retention in resource-limited settings.

There are several plausible reasons to explain the substantial benefit demonstrated by the seemingly modest intervention evaluated in this study. First, direct accompaniment of postpartum mothers to the site where EID takes place informs the mother more concretely of the need and venue for infant testing and better prepares her for the infant visit in a month. Linkage between these 2 services improved continuity. Second, mothers who received enhanced referral were counseled in the privacy of the EID testing suite instead of in the open maternity ward, with a possible effect of reducing stigma and promoting continued program participation. Third, mothers who were directly accompanied to the EID suite had the opportunity to enroll their infants that day in EID services and have a medical record started for their newborn. If she were to bring the infant back for preventive vaccinations, this medical record may alert staff to the infant's HIV-exposed status and enable them to reinforce the need for testing to the mother. The infant medical record also facilitates identification and engagement of infants who are lost-to-follow-up in their home communities.

Although the enhanced referral process shows promise in impacting HIV-exposed infant follow-up for EID, only a small fraction of women who were eligible for the study intervention actually received it. Feedback from nursing staff indicated that the main barriers to offering women enhanced referral were structural; 2 common reasons cited were that there was no standard way of recording enhanced referral activities and that because of this, offering these enhanced services to women was often forgotten. High staff turnover was another commonly cited barrier. Though the process of enhanced referral was slightly more time consuming for staff than standard care, added workload was not seen by staff as a barrier to offering more women these services (P.J.C. and T.D.M. (MD), personal communication, Oral Communication, August 2010). We are in the process of seeking to overcome these barriers with better standardization and documentation of discharge activities.

Although the simplicity, low cost, and potential efficacy of this intervention make it particularly promising to scale-up to other settings, enhanced referral alone is not sufficient to reach adequate EID coverage because nearly half of all women who received it failed to return with their infants for follow-up. This pilot work could be complemented by qualitative exploration of barriers to follow-up from the patient's perspective and work to improve the quality of the counseling given to patients before discharge.

Strengths of our study included the complete ascertainment of all hospital deliveries in the period of interest, the simple nature of our enhanced services, and the relevance of our research to implementation of improved PMTCT and EID coverage. A limitation is that these hospital-based results may not apply to women who give birth at home—a group that represents a majority of all births in Zambézia Province (51% in 2010, unpublished data from the Ogumaniha Project Baseline Survey Report, Vanderbilt University). The intervention and control groups were similar across the demographic characteristics included in the study, and differences were accounted for in multivariable models, but our results may be biased by unmeasured confounding or by selection bias in a nonrandomized trial. The name matching technique used to define follow-up is subject to potential misclassification bias, and our study was not designed to account for infant mortality or migration that may be misclassified as lost-to-follow-up. Last, only a small portion of our overall study sample received the study intervention.

Improving PMTCT system performance is key to reducing early mother-to-child transmission.23-25 At the distal end of this PMTCT cascade of activity, postpartum discharge serves as a bridge to EID and to care for the mother herself. Infant attrition at this step may result in postpartum infant medical neglect (if perinatally infected) or subsequent infection through breastmilk, more likely if cART is not given to the mother. This study demonstrates the potential to improve one aspect of system performance by designing a hospital-based intervention to improve EID and subsequent infant care delivery in a real-world setting. In 2011, we will complete pilot work that builds on these results, using quality improvement methodology to implement clinic changes that encourage additional staff and patient participation in enhanced referral services. Initial results are promising. Implementation research can guide how best to integrate PMTCT services with prenatal and infant preventive care to improve EID coverage and mother/infant cART to reduce HIV clinical burden on mothers and infants.

Back to Top | Article Outline


The authors thank Sérgio Patrício Roques and Madalena Vundo for help with staff interviews and data collection.

Back to Top | Article Outline


1. Coovadia H. Current issues in prevention of mother-to-child transmission of HIV-1. Curr Opin HIV AIDS. 2009;4:319-324.
2. Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet. 1999;354:795-802.
3. Vergara AE, Blevins M, Vaz LME, et al. Efficacy of three short-course regimens of zidovudine and lamivudine in preventing early and late transmission of HIV-1 from mother to child in Tanzania, South Africa, and Uganda (Petra study): a randomised, double-blind, placebo-controlled trial. Lancet. 2010;359:1178-1186.
4. Coovadia H, Kindra G. Breastfeeding to prevent HIV transmission in infants: balancing pros and cons. Curr Opin Infect Dis. 2008;21:11-15.
5. Stringer EM, Sinkala M, Stringer JS, et al. Prevention of mother-to-child transmission of HIV in Africa: successes and challenges in scaling-up a nevirapine-based program in Lusaka, Zambia. AIDS. 2003;17:1377-1382.
6. Perez F, Mukotekwa T, Miller A, et al. Implementing a rural programme of prevention of mother-to-child transmission of HIV in Zimbabwe: first 18 months of experience. Trop Med Int Health. 2004;9:774-783.
7. Manzi M, Zachariah R, Teck R, et al. High acceptability of voluntary counselling and HIV-testing but unacceptable loss to follow up in a prevention of mother-to-child HIV transmission programme in rural Malawi: scaling-up requires a different way of acting. Trop Med Int Health. 2005;10:1242-1250.
8. Delva W, Draper B, Temmerman M. Implementation of single-dose nevirapine for prevention of MTCT of HIV-lessons from Cape Town. S Afr Med J. 2006;96:706, 708-709.
9. Stringer EM, Ekouevi DK, Coetzee D, et al. Coverage of nevirapine-based services to prevent mother-to-child HIV transmission in 4 African countries. JAMA. 2010;304:293-302.
10. Braun M, Kabue MM, McCollum ED, et al. Inadequate coordination of maternal and infant HIV services detrimentally affects early infant diagnosis outcomes in Lilongwe, Malawi. J Acquir Immune Defic Syndr. January 10, 2011. E-pub ahead of print.
11. Nyandiko WM, Otieno-Nyunya B, Musick B, et al. Outcomes of HIV-exposed children in western Kenya: efficacy of prevention of mother to child transmission in a resource-constrained setting. J Acquir Immune Defic Syndr. 2010;54:42-50.
12. Ahoua L, Ayikoru H, Gnauck K, et al. Evaluation of a 5-year programme to prevent mother-to-child transmission of HIV infection in Northern Uganda. J Trop Pediatr. 2010;56:43-52.
13. Sherman GG, Jones SA, Coovadia AH, et al. PMTCT from research to reality-results from a routine service. S Afr Med J. 2004;94:289-292.
14. Briggs CJ, Garner P. Strategies for integrating primary health services in middle- and low-income countries at the point of delivery. Cochrane Database Syst Rev. 2006:CD003318.
15. Delvaux T, Konan JP, Ake-Tano O, et al. Quality of antenatal and delivery care before and after the implementation of a prevention of mother-to-child HIV transmission programme in Cote d'Ivoire. Trop Med Int Health. 2008;13:970-979.
16. USAID. Implementation of the Global Health Initiative: consultation document. 2010; Available at: Accessed December 19, 2010.
17. Office of U.S. Global AIDS Coordinator. The U.S. President's emergency plan for AIDS relief: five-year strategy. 2009. Available at: Accessed December 19, 2010.
18. Cook RE, Ciampa PJ, Sidat M, et al. Predictors of successful early infant diagnosis of HIV in a rural district hospital in Zambezia, Mozambique. J Acquir Immune Defic Syndr. 2011;56:e104-e109.
19. Moon TD, Burlison JR, Sidat M, et al. Lessons learned while implementing an HIV/AIDs care and treatment program in rural Mozambique. Retrovirol: Res Treatment. 2010;3:1-14.
20. Audet CM, Burlison J, Moon TD, et al. Sociocultural and epidemiological aspects of HIV/AIDS in Mozambique. BMC Int Health Hum Rights. 2010;10:15.
21. Comité Nacional de Tratamento Antiretroviral. Guia de Tratamento Antiretroviral e Infecções Oportunistas no Adulto, Adolescente e Grávida 2009/2010. 2009. Available at: Accessed April 14, 2011.
22. Little RJA, Rubin DB. Statistical Analysis With Missing Data. New York, NY: Wiley; 1987.
23. Barker PM, Mphatswe W, Rollins N. Antiretroviral drugs in the cupboard are not enough: the impact of health systems' performance on mother-to-child transmission of HIV. J Acquir Immune Defic Syndr. 2011;56:e45-e48.
24. Stringer EM, Chi BH, Chintu N, et al. Monitoring effectiveness of programmes to prevent mother-to-child HIV transmission in lower-income countries. Bull World Health Organ. 2008;86:57-62.
25. Reithinger R, Megazzini K, Durako SJ, et al. Monitoring and evaluation of programmes to prevent mother to child transmission of HIV in Africa. BMJ. 2007;334:1143-1146.

HIV/AIDS; prevention of mother-to-child transmission; sub-Saharan Africa

© 2011 Lippincott Williams & Wilkins, Inc.