A series of hypotheses concerning relationships between age, sex, social class and the prevalence of handicapping conditions were developed and tested against the data from two large British cross-sectional surveys. The data showed that, as predicted, the prevalence of handicapping conditions was most strongly associated with indicators of social class among the middle-aged and that class related differences on physiological and lifestyle health indicators occurred at an earlier age than differences in the prevalence of handicaps. The relationship between social class and handicap prevalence was found for a variety of types of handicap. One interpretation of the pattern of relationships between indicators of social class and the prevalence of handicap in different age groups is that the risk of experiencing a handicapping condition is affected by long-term environmental effects associated with class. The methodological limitations of cross-sectional surveys in relation to propositions about causality are discussed.