Decentralization of CD4 testing in resource-limited settings: 7 years of experience in six African countries

Improving access to CD4 testing in resource-limited settings can be achieved through both centralized and decentralized testing networks. Decentralized testing models are more suitable for countries where the HIV epidemic affects a large portion of rural populations. Timely access to accurate CD4 results is crucial at the primary level of the health system. For the past 7 years, the Institute of Human Virology of the University of Maryland School of Medicine has implemented a flexible and sustainable three-phase model: (1) site assessment and improvement, (2) appropriate technology selection with capacity building through practical training and laboratory mentoring, and (3) quality management system strengthening and monitoring, to support accessibility to reliable CD4 counting at the point of service. CD4 testing capacity was established in 122 of 229 (53%) laboratories supported in Nigeria, Uganda, Kenya, Zambia, Tanzania, and Rwanda. Among those in rural settings, 46% (69/151) had CD4 testing available at site level, with a functioning flow cytometer installed at 28% (8/29) and 50% (61/122) of level 1 and level 2 sites, respectively. To strengthen local capacity, a total of 1,152 laboratory technicians were trained through 188 training sessions provided both on-site and at central locations. The overall quality of CD4 total testing procedure was assessed at 76% (92/121) of the laboratories, with 25% (23/92), 34% (31/92), and 33% (30/92) of them reporting excellent, good, and satisfactory performance. Balancing country-specific factors with the location of the clinic, number of patients, and the expected workload, was crucial in adapting this flexible model for decentralizing CD4 testing. The close collaboration with local governments and private vendors was key to successfully expanding access to CD4 testing within the framework of HIV care and treatment programs and for the sustainability of medical laboratories in resource-limited settings.