Analysing morbidity determination with flawed data: the case of dysentery in regional India

A method is suggested for removing measurement error from morbidity data collected in Third World countries. The method assumes the existence of diseases which are independent of socio-economic factors and exploits observed correlations between these disease and these factors to construct an index of measurement error. This index may be used as an explanatory variable in the analysis of other diseases which are hypothesised to have socio-economic causes which are the object of research. The method is applied to statewide data for India where the index of measurement error is based on influenza data which is then used to isolate the socio-economic factors that influenced dysentery.