OBJECTIVES: To assess the cost-effectiveness of maternal triple-drug ART or infant nevirapine to reduce HIV transmission during breastfeeding from HIV-infected Malawian mothers, not meeting treatment guidelines for their own health. Strategies were evaluated at two entry points into the health system: 1) time of delivery, for mothers who have not accessed ART antenatally, and 2) antenatally. METHODS: Cost-effectiveness was evaluated in terms of healthcare costs per HIV transmission averted, life-years gained, quality adjusted life-years (QALYs) gained, and disability adjusted life-years (DALYs) averted. The risks of HIV transmission associated with the interventions were taken from the Kesho Bora and BAN studies. Antenatal regimens were 1) zidovudine (ZDV); 2) ZDV, lamivudine  (3TC), lopinavir/ritonavir; and (3) no antenatal regimen.  The postnatal regimens were 1) maternal antiretrovirals (ZDV, 3TC, lopinavir/ritonavir); 2) infant nevirapine (NVP); and 3) no extended post-natal prophylaxis. Scenario analyses included: alternate drug combinations and prices, and earlier antenatal initiation.  RESULTS: For mothers presenting at delivery, infant NVP during breastfeeding is likely to be considered cost-effective.  Incremental cost-effectiveness ratios (ICERs) compared to standard of care are $264 per HIV transmission averted, $15 per life-year saved, $16 per QALY and $15 per DALY averted.  For mothers presenting antenatally, antenatal ZDV plus infant NVP during breastfeeding appears cost-effective. ICERs compared to ZDV plus standard of care are $667 per HIV transmission averted, $37 per life-year saved, $39 per QALY and $37 per DALY averted.  CONCLUSIONS: For pregnant, HIV-1 infected women presenting at delivery and not requiring antiretrovirals for their own health, infant NVP throughout breastfeeding is likely to be considered cost-effective.  For those presenting antenatally, antenatal ZDV followed by infant NVP throughout breastfeeding (WHO Option A) is likely to be considered cost-effective.