Cost of using a patient tracer to reduce loss to follow-up and ascertain patient status in a large antiretroviral therapy program in Johannesburg, South Africa

OBJECTIVE: To evaluate a pilot intervention to engage a patient tracer to follow up lost patients at a large public clinic in South Africa. METHODS: A social worker spent 4 months contacting by telephone a random sample of patients who had initiated antiretroviral therapy (ART) at least 6 months earlier and were ≥1 month late for a scheduled visit. The tracer was authorized to assist patients to return to care if needed. Cost was calculated from the perspective of the clinic. RESULTS: The tracer was able to determine the final status of 260 of a sample of 493 lost patients. Of the 260, 55 (21%) had died, 56 (21%) were still on ART at the same site, 79 (30%) reported transferring to another site and 70 (27%) had discontinued treatment. Among those discontinuing, commonly cited reasons were relocation (n = 18, 26%), traditional medicine or religious beliefs (n = 11, 16%), fear of disclosure or other family barriers (n = 9, 13%), and employment obstacles (n = 7, 10%). Twenty patients returned to care at the original site as a result of the intervention, at an average cost of $432 per patient returned. CONCLUSIONS: A patient tracer was an effective way to determine the final status of lost patients and succeeded in returning some to care, but the cost per patient returned was high. Better information systems allowing sites to track deaths and transfers would greatly improve the efficiency of loss to follow-up interventions.