OBJECTIVES: Examine the impact of adjusting for price differences (conversion to international dollars), as advocated by the World Health Organization (WHO), when conducting cost-effectiveness analyses in developing nations. METHODS: We created a hypothetical cost-effectiveness analysis, set in India, with 2 strategies: standard care (STD-CARE) and intervention. Two costing methods were used: unadjusted exchange rate (XCHG), and international dollars (I$) where non-traded goods (e.g. services or locally traded goods) and per-capita gross domestic product (GDP) were adjusted for international price differences using purchasing power parity (PPP) estimates from WHO. Three different interventions were considered: 1) Costs are all internationally traded goods (TRADED); 2) Costs are all non-traded goods (NON-TRADED); and 3) Costs evenly split between traded and non-traded goods (MIX). STD-CARE costs freely varied from 0-100% traded goods. GDP was about $600 using the exchange rate, I$1865 using PPP. The outcome measure was the incremental cost-effectiveness ratio stated as multiples of GDPs/DALY (Disability-adjusted Life Year). We considered the commonly used threshold of 1GDP, as well as WHO recommendations: =1GDP “very cost-effective”, 1-3GDP “cost-effective”. Parameters were chosen to achieve an ICER of 4GDP for XCHG. RESULTS: By definition, the ICER is constant (4GDP) for XCHG. For NON-TRADED using $I, the ICER was =4GDP in all cases. For TRADED using I$, the ICER was <1 when <79% of STD-CARE costs were traded goods. For MIX using I$, ICERs were never below 1GDP, but were below 3GDP when <76% of STD-CARE costs were traded goods. CONCLUSION: Adjusting prices using purchasing power parity (I$'s) as opposed to the unadjusted official exchange rate can dramatically alter conclusions when comparing interventions involving different proportions of traded goods. In particular, an intervention with costs primarily from traded goods (e.g. vaccination) may be clearly not cost-effective using the exchange rate, yet appear very cost-effective after adjusting for purchasing power parity.