ILLNESS BEHAVIOR AND THE MEASUREMENT OF CLASS DIFFERENTIALS IN MORBIDITY

Summ. measures of morbidity are necessary for research and health poli cy purposes but those used at present may provide unreliable estimates of the magnitude of social class differences in health. Class differe nces in illness behaviour will affect self-reported health measures in dependently of any differences in health. The aspects of illness behav iour which are relevant will vary with the morbidity measure. It may b e possible to test for illness behaviour effects during data analysis; if not, the effects need to be considered when results are interprete d. Future studies might develop a battery of standardized instruments to estimate the proportion of a population which is free of serious di sease.