THAE0402 | SHIFTING HIV PROVIDER WORKFLOWS WITH DSD SCALE-UP IN ZAMBIA: RESULTS FROM TIME AND MOTION STUDIES

Anushka Reddy Marri¹, Linda Sande², Taurai Makwalu³, Aniset Kamanga³, Prudence Haimbe³, Priscilla Lumano-Mulenga⁴, Hilda Shakwelele³, Amy Huber², Sophie Pascoe², Sydeny Rosen^1|2 | ¹Department of Global Health, Boston University School Of Public Health, Boston, United States; ²Health Economics and Epidemiology Research Office, University of the Witwatersrand Faculty of Health Sciences, Johannesburg, South Africa; ³Clinton Health Access Initiative, Zambia; ⁴Ministry of Health, Zambia

We observed providers’ time use at two time points (T1: 2021, T2: 2024) in 12 ART facilities across Lusaka and Central Provinces. Providers were observed for up to two full days each. Each provider activity was timed and categorized by use: direct patient care (DPC), indirect patient support (IPS) (patient-related records, referrals, and facility-related administrative tasks), or non-patient tasks (NPT) (meetings, breaks). Patient type based on ART status and DSD model was documented for DPC interactions. We report median proportions of time spent across categories. 119 providers were observed for 237 days in total, for a median of 7.6 hours/person/day. DPC time increased from a quarter of the workday (25%) at T1 to more than a third (34%) at T2. IPS time increased from 22% in T1 to 37% in T2. NPT time declined slightly from 30% in T1 to 29% in T2. Three-quarters of DPC time was spent with established clients (on ART > 6 months) in T1 and T2; DPC time with early treatment (on ART < 6 months) clients rose from 3.5% in T1 to 5% in T2, and DPC time with newly initiating clients doubled from 2.5% in T1 to 5% in T2. Across T1–T2, 40–70% of IPS was spent on patient-related tasks and 40–60% on facility-related tasks. Time spent on clients in conventional care (CC) declined from 20% in T1 to 13% in T2. Time spent on 6-month-dispensing (6MMD) clients increased from 45% in T1 to 67% in T2. Nurses and clinicians spent a median of 7-8 minutes per CC client and 5-6 minutes per 6MMD client throughout the study period. As more clients received 6MMD in Zambia, total provider time spent on this model increased, while provider time per client remained stable. Increased indirect and non-patient activities suggest broadening of provider responsibilities. These findings can inform strategies to optimize efficiency as service delivery models evolve.