Background: The adoption of ''Test and Start'' to accelerate the achievement of the 90-90-90 goals has increased the number of ART clients. In many countries, this requires the adoption of streamlined service delivery approaches to combat potential health systems constraints - notably the lack of adequate human resources for health (HRH). Differentiated service delivery (DSD) models are one such innovative approach. Under these models, drug refills are separated from clinical consultations, and stable ART clients have less frequent clinical consultations; receive their refills faster and less frequently; and can be managed outside health facilities by health workers or ART clients themselves.
Implementing DSD models in high HIV burden, resource limited settings often requires optimization of the limited HRH resources to include task-shifting and sharing. In 2017, the HRH2030 program, funded by USAID, through PEPFAR, explored such practices at 20 health facilities offering DSD models in Uganda to better understand staffing requirements.
Methods: 1,720 client flow observations were undertaken across eight ''critical tasks'' along the ART service delivery continuum for DSD and the standard models of care. For each observation, service providers performing specific tasks were identified and their frequency of task performance tallied. For analysis, these service providers were grouped into professional and lay health workers as shown in the table.
Results: It was found that the proportion of critical tasks carried out by both professional and lay health service providers (51% vs 49%, p=0.2408) was similar. Nurses performed 66% of observed tasks in the professional group while facility-based auxiliaries - a type of trained, facility-based but non-professionally certified lay provider performed 48% of all tasks in the second group. Additionally, lay providers were found to be more involved in enrollment and triage (p< 0.0001) while the professional group were more involved in clinical consultations, laboratory/blood work and drug dispensing (p< 0.0001).
Conclusions: Appropriate task-shifting and sharing can reduce overburdening and optimize client flows hence improving health worker utilization during the implementation of DSD models. Our data suggest that if countries allowed greater use of task shifting/sharing and lay workers through DSD, limited clinical staff could be prioritized for ART initiation or complex cases.



[Frequency of service provision by various service providers]

Download the e-Poster