University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Implementation of a Programme for the Prevention of Mother-to-Child Transmission of HIV in a Ugandan Hospital over Five Years: Challenges, Improvements and Lessons Learned
Global Health Sciences Literature Digest
Published May 9, 2007
Journal Article

Magoni M, Okong P, Bassani L, Kituka Namaganda P, Onyango S, Giuliano M. Implementation of a Programme for the Prevention of Mother-to-Child Transmission of HIV in a Ugandan Hospital over Five Years: Challenges, Improvements and Lessons Learned. Int J STD AIDS. 2007 Feb;18(2):109-13.


To identify factors that may limit acceptance of HIV testing and enrollment in a prevention of mother-to-child transmission (PMTCT) program.

Study Design

This was a retrospective cohort study of pregnant women offered voluntary HIV counseling and testing (VCT) and PMTCT at an antenatal clinic (ANC) between January 2000 and December 2004.


This study took place in St. Francis Hospital Nsambya, a private non-governmental referral hospital in Kampala, Uganda. This hospital has a catchment area of 250,000 people, and was one of the first sites to implement the PMTCT program sponsored by the National Ministry of Health.


From January 2000 to December 2004, 26,556 women attended the antenatal clinic. The mean age of mothers was 25.3 years and mean gestational age at first ANC visit was 26.7 weeks.


This was a retrospective program analysis. All women attending the ANC were offered VCT, and women who tested positive were asked to return to receive either short-course zidovudine (ZDV) or nevirapine (NVP) at 36 weeks. Women planning to breastfeed were counseled to do so exclusively. Free replacement feeding was provided until 2002. All women were encouraged to deliver in the hospital and to attend scheduled postnatal follow-up appointments with their babies (week 1, week 6, week 10, week 14, month 6 and every three months thereafter) until 18 months, when an HIV antibody test was performed. Because of poor postnatal follow-up at the six-week visit during the first year (59%), an "active-monitoring" program was put into place from 2001-2002, which provided transportation reimbursement and free PCR testing of infants at six weeks and six months. In 2003, the hospital became part of the MTCT-Plus initiative, providing family HIV care and access to antiretrovirals. To support the PMTCT program, significant numbers of new staff and financial resources were required.

Primary Outcomes

Service performance outcomes included: the proportion of women counseled, tested, identified as HIV-positive, enrolled in the program, and delivered in the hospital; the proportion of children tested, identified as HIV-positive, and receiving formula; the proportion of male partners counseled, tested, and identified as HIV-positive. Relationship of demographic factors to uptake was also evaluated.


Of the 26,556 women attending the antenatal clinic, 91% received pre-test counseling and 76% of those women agreed to HIV testing. Of those tested, 2011 (11%) were HIV-positive, among whom 1341 (67%) enrolled in the program and received ZDV or NVP; 1212 (90% of those enrolled) delivered in the hospital. The proportion of counseled women who accepted testing remained the same in the first three years (73%), but increased significantly with the MTCT-Plus program (80%). The proportion of HIV-positive women who enrolled increased from 49% in 2000 to an average of 76% in the following years. When 'active monitoring' was implemented to encourage clinic follow-up, the proportion of women coming to the six-week postnatal visit increased from 59% to 71%; this resulted in an increase in the number of infants who were HIV-tested from 46% in 2000 to 77% during 2001-2002. However, when active monitoring was discontinued (and despite MTCT-Plus), only 37% of women returned for follow-up and only 37% of infants were HIV-tested from 2003-2004. Overall, 54% of babies had at least one HIV test, and 12% were identified as HIV-positive. About one-third of mothers adopted replacement feeding, regardless of whether free formula was available. The proportion of male partners counseled remained low (only 3-5%), although, if counseled, most accepted HIV testing (83-96%). HIV prevalence among tested men was high (18-26% HIV-positive).

Women who were <20 years of age, single, had less education, and were not Muslim or Protestant were more likely to accept HIV testing. HIV prevalence was highest among women >30 years old (16.4% HIV-positive vs. 5.3% HIV-positive in teenagers); those with no education (14.9% HIV-positive vs. 9% HIV-positive among those with tertiary education); widows (82% HIV-positive); and those who were cohabiting but unmarried (15% HIV-positive). Women who were >30 years old (65% enrolled) and who had at least secondary education (61-70% enrolled) were most likely to enroll in the program. Women from the local Kampala (Baganda) tribe were the least likely to accept testing and to enroll.


The successful implementation of a PMTCT program for large numbers of women is feasible if the health system has adequate resources and personnel. The authors point out that the involvement of male partners and the availability of ART through the MTCT-Plus program likely played a role in increasing acceptance of HIV testing among women. On the other hand, the high loss to follow-up and the reliance on HIV antibody testing of infants at 18 months in government-sponsored programs may undermine the effectiveness of PMTCT. The lower rate of acceptance and testing among local women may be due to stigma and fear of discrimination if recognized. Improved enrollment over time was most likely due to high-quality counseling at the clinic, the integration of PMTCT into the regular hospital activities, and the experience acquired by the staff.

Quality Rating

Based on the Newcastle-Ottawa quality rating for observational cohort studies, this study was of adequate quality. The study was limited by the following: there is no description of the methodology for obtaining data for the retrospective review, and socio-demographic information was collected for fewer than half of the women in the cohort. Limited information was provided on exclusive breastfeeding or exclusive formula-feeding, and there were no data on infant mortality.

In Context

Enrollment rates of 75% or more in the PMTCT program are higher than reported in other programs,(1,2,3) and according to the authors, may be due to quality counseling services and experienced staff. Issues of stigma, discrimination, and male involvement have been repeatedly shown to be important to successful enrollment in PMTCT programs.(1) The overall HIV transmission rate to infants was 12% and was comparable to other programs. This study took place in a referral hospital funded by external donors, and provided service to women who were largely urban and educated (over two-thirds had secondary or university education). In addition, the vast majority of women delivered in the hospital. Even so, the issues of transportation, problems of postnatal follow-up and stigma were important, and are even more pronounced in rural clinics and populations.

Programmatic Implications

PMTCT programs are being implemented throughout resource-limited countries. The success of these programs will depend on the availability of adequate financial resources to support infrastructure and training of personnel. Staff need to be sensitive to issues of stigma that may prevent women from participating. In addition, comprehensive and integrated services, including general healthcare and HIV care for families, involvement of male partners, and the availability of ART all impact the success of PMTCT. Thus, integrated and comprehensive services are likely to result in more successful programs. Even in an urban hospital, problems of postnatal follow-up persist, so the burden of transportation costs and other logistical issues need to be addressed.


  1. Perez F, Orne-Gliemann J, Mukotekwa T, Miller A, Glenshaw M, Mahomva A, et al. Prevention of mother to child transmission of HIV: evaluation of a pilot programme in a district hospital in rural Zimbabwe. BMJ 2004 Nov 13;329(7475):1147-50.
  2. van't Hoog AH, Mbori-Ngacha DA, Marum LH, Otieno JA, Misore AO, Nganga LW, et al. Preventing mother-to-child transmission of HIV in Western Kenya: operational issues. J Acquir Immune Defic Syndr 2005 Nov 1;40(3):344-9.
  3. Sweat MD, O'Reilly KR, Schmid GP, Denison J, de Zoysa I. Cost-effectiveness of nevirapine to prevent mother-to-child HIV transmission in eight African countries. AIDS 2004 Aug 20;18(12):1661-71.