The paper examines the use of the new measure of social class in the UK, the National Statistics Socio-Economic Classification (NS-SEC) and other socio-economic variables in explaining differences in health between British South Asians and the majority White population. There are a number of hypotheses which try to explain ethnic differences in health and yet there have been relatively few empirical studies which test the explanatory value of these hypotheses. Cross sectional data from the fourth National Survey of Ethnic Minorities (1993-1994) with 2860 white, 1268 Indian and 1771 Pakistani and Bangladeshi adult respondents are analysed. The associations of self-rated health with ethnicity, social class, local area deprivation and standard of living are analysed. Pakistani and Bangladeshi respondents have the poorest self-rated health, followed by Indians. Differences in self-rated health between ethnic groups reduce to non-significance after adjusting for social class, local area deprivation and standard of living. There is some evidence of social class differences in the health of Indians and not much evidence for Pakistanis and Bangladeshis. The NS-SEC is useful in explaining ethnic differences in health. The poorer health of Indians, Pakistanis and Bangladeshis compared to Whites may be largely understood in terms of factors related to occupational social class, material living conditions and local area deprivation.