On lies and health statistics: some Latin American examples

New methods of demographic analysis are producing estimates of fertility and mortality which are sometimes at great variance with "official" figures generated by the statistics organizations of the different countries and which are reproduced in international reference books. This discrepancy is greatest with regard to infant mortality. Using Latin American examples, the magnitude of this discrepancy is explored, biases in estimating causes of mortality are identified, and a consideration is made of morbidity figures, which, as they are generated by health care systems with very low coverages of population, tend to seriously underrepresent the prevalent levels of disease. A structural interpretaton is made of the Latin American situation, linking this crisis of health statistics with a more general crisis of the "developmentist" model under which these systems flourished, and with an upsurge in political repression in the Continent which will tend in future to increase the inaccuracy of "official" health statistics data. Finally, alternative health statistics procedures are proposed.