The effect of initiating tenofovir on HIV treatment outcomes in adults in southern Africa: a regression discontinuity analysis

By  Dr. Alana Brennan  Dr. Jacob Bor  Mary-Ann Davies, Kathryn Stinson, Robin Wood, Hans Prozesky, Frank Tanser, Geoffrey Fatti, Till Bärnighausen, Gilles Wandeler, Andrew Boulle, Izukanji Sikazwe, Arianna Zanolini  Dr. Matthew Fox  |  | 


Background: Due to toxicities associated with the NRTI stavudine, many countries now recommend initiating HIV patients on tenofovir as the standard NRTI in first-line therapy. Exploiting national guideline changes in South Africa and Zambia, we assessed the causal impact of a policy to initiate tenofovir on ART outcomes using a regression discontinuity design.  Objectives: To establish the causal effect of tenofovir (vs. stavudine) on patient outcomes in South Africa and Zambia’s public sector HIV clinics.  Methods:  STUDY POPULATION: Prospective cohort study of ART-naïve, non-pregnant, HIV patients >16 years old who initiated first-line ART in South Africa or Zambia (IeDEA-SA cohort). Patients were included if they initiated ART +/-12 months around the national guideline changes: 1 April 2010 in South Africa and 1 July 2007 in Zambia. All patients had the potential for 24-months of follow-up.  OUTCOMES: Indicator for death, loss to follow-up (defined as not attending the clinic in at least 6-months), single-drug substitution (changing NRTI within first-line ART), immunological and virologic failure (defined as two consecutive viral loads >400 copies/mL) in the first 24-months on ART.

Conference:  IWHOD 2015, Catania, Sicily, 26-28 March 2015

Publication details