Objective: To describe the proportion of the households that experienced catastrophic health expenditure and tocompare the magnitude of catastrophic health expenditure that is estimated from three different data sets in Vietnam.Methods: The study we are comparing with is based on data from the Vietnam Living Standard Survey (VLSS) 1997/98.We have used data from the 2001 re-survey. The FilaBavi sample consists of 11,089 households. We have also conducteda special survey from July 2001 to June 2002. The sample is smaller, 629 households, but they have been followed for anentire year with monthly interviews. For VLSS and FilaBavi, re-census survey households were interviewed once with a recall period of one month.Findings: In the VLSS data and in the FilaBavi re-census survey it was found that 9%-10% of the households experienced catastrophic healthcare spending. But, only 5% (average per month) and 1% (for the whole year) of the households in thespecial survey report catastrophic spending.Conclusions: We suggest that the major reason for the difference in the estimates is the different data collection methods. When doing a cross sectional study with a relatively short recall period there is a risk that households will tend to overestimate non-recurrent large expenditures as that for health.