The Harmonic Standardisation of Comparisons between Success Rates in Two Heterogeneous Groups of Patients

The comparison of numbers of successes and failures in two treatment groups by the usual chi-squared criterion with one degree of freedom becomies invalid if the groups are heterogeneous, falling into categories with different success rates in propoitions which differ systematically for the treatments. If these nuimbers are adjlusted beforehand by a simple procedure of the type used in stanclarlising mortality rates the validity of the criterion is restored, subject to multiplication by aii inflating factor which exceeds unity by an amouint which is usually so small that it may be neglected.