The effectiveness and cost implications of task-shifting in the delivery of antiretroviral therapy to HIV-infected patients: a systematic review.

INTRODUCTION Human resource shortages are a challenge to the rollout of antiretroviral therapy (ART) for HIV-infected patients, particularly in sub-Saharan Africa. Task-shifting has been recommended as an approach to reduce the impact of human resource shortages. We conducted a systematic review of randomized controlled trials and quasi-experimental studies to assess the effectiveness of task-shifting, and its impact on costs of ART provision. METHODS We searched MEDLINE, EMBASE, PSYCINFO, Cochrane Library, Web of Knowledge and the Current Controlled Trials Register for articles published up to January 2011. We included studies evaluating any task-shifting model against any other intervention using any of the following outcomes: mortality (all causes); occurrence of new AIDS-defining illness; virological outcomes; CD4 cell count; adherence to ART medicines (e.g. self-report and pill counts); hospital admissions; clinic visits; toxicity or adverse events; quality of life indicators; costs and cost-effectiveness. We did not pool the results because of high levels of clinical heterogeneity. RESULTS We identified six effectiveness studies including a total of 19 767 patients. Non-inferior patient outcomes were achieved with task-shifting from doctors to nurses, or from health care professionals to mid-level workers or lay health workers. However, most of the identified studies were underpowered to detect any difference. Three studies were identified on the cost implications of task-shifting. Task-shifting resulted in substantial cost and physician time savings. CONCLUSIONS The reviewed evidence suggests that task-shifting from doctors to nurses, or from health care professionals to lay health workers can potentially reduce costs of ART provision without compromising health outcomes for patients. Task-shifting is therefore a potentially effective and cost-effective approach to addressing the human resource limitations to ART rollout. However, most of the studies conducted were relatively small and more evidence is needed for each task-shifting model as it is currently limited.