Open Access Research article

Early infant diagnosis of HIV in three regions in Tanzania; successes and challenges

Mercy G Chiduo1*, Bruno P Mmbando1, Zahra P Theilgaard2, Ib C Bygbjerg3, Jan Gerstoft2, Martha Lemnge1 and Terese L Katzenstein2

Author Affiliations

1 National Institute for Medical Research Tanga Centre, Bombo Road, Box 5004, Tanga, Tanzania

2 Department of Infectious Diseases, University Hospital of Copenhagen, Blegdamsvej 9, Copenhagen DK-2100, Denmark

3 Department of International Health, Immunology and Microbiology, Faculty of Health Sciences, University of Copenhagen, Øster Farimagsgade 5, Copenhagen DK-1014, Denmark

For all author emails, please log on.

BMC Public Health 2013, 13:910  doi:10.1186/1471-2458-13-910

Published: 2 October 2013



By the end of 2009 an estimated 2.5 million children worldwide were living with HIV-1, mostly as a consequence of vertical transmission, and more than 90% of these children live in sub-Saharan Africa. In 2008 the World Health Organization (WHO), recommended early initiation of Highly Active Antiretroviral Therapy (HAART) to all HIV infected infants diagnosed within the first year of life, and since 2010, within the first two years of life, irrespective of CD4 count or WHO clinical stage. The study aims were to describe implementation of EID programs in three Tanzanian regions with differences in HIV prevalences and logistical set-up with regard to HIV DNA testing.


Data were obtained by review of the prevention from mother to child transmission of HIV (PMTCT) registers from 2009–2011 at the Reproductive and Child Health Clinics (RCH) and from the databases from the Care and Treatment Clinics (CTC) in all the three regions; Kilimanjaro, Mbeya and Tanga. Statistical tests used were Poisson regression model and rank sum test.


During the period of 2009 – 2011 a total of 4,860 exposed infants were registered from the reviewed sites, of whom 4,292 (88.3%) were screened for HIV infection. Overall proportion of tested infants in the three regions increased from 77.2% in 2009 to 97.8% in 2011. A total of 452 (10.5%) were found to be HIV infected (judged by the result of the first test). The prevalence of HIV infection among infants was higher in Mbeya when compared to Kilimanjaro region RR = 1.872 (95%CI = 1.408 – 2.543) p < 0.001. However sample turnaround time was significantly shorter in both Mbeya (2.7 weeks) and Tanga (5.0 weeks) as compared to Kilimanjaro (7.0 weeks), p=<0.001. A substantial of loss to follow-up (LTFU) was evident at all stages of EID services in the period of 2009 to 2011. Among the infants who were receiving treatment, 61% were found to be LFTU during the review period.


The study showed an increase in testing of HIV exposed infants within the three years, there is large variations of HIV prevalence among the regions. Challenges like; sample turnaround time and LTFU must be overcome before this can translate into the intended goal of early initiation of lifelong lifesaving antiretroviral therapy for the infants.

Early infant diagnosis; ART; Lost to follow-up; Tanzania