II. Quantitative Analytical Procedure

Abstract

A statistical methodology has been established to describe and compare, quantitatively, the ototoxic effects of aminoglycoside antibiotics. Sampling techniques for obtaining reliable estimates of cell damage in the organ of Corti and the vestibular system are described in this paper. The relationship of drug dosage to the percentage of hair cell damage in each inner ear sensory region was expressed as the slope of the least-squares linear regression line and was tested to reject the null hypothesis (slope u = 0) at the 0.05 level of significance. It was determined that, in each of the cases discussed, dosage and damage correlated positively and that by repeated computations, each time controlling for one observed or measured variable, guinea pigs contributing skewed results could be eliminated on a rational basis. By this method of careful selection, statistically significant relationships have been established from the data.