Using a Modified Lorenz Curve and Gini Coefficient to Enhance Interpretation of Treatment Effect on Patient-reported Outcomes
*Joseph C Cappelleri, Pfizer Inc
Keywords: patient-reported outcomes, interpretation, Gini coefficient, Lorenz curve
To enhance interpretation of treatment effect on patient-reported outcomes (PROs), we extend the traditional Gini coefficient and its Lorenz curve to two independent samples. We do this by plotting the cumulative percentage of individuals in the control group on the horizontal axis and the cumulative percentage of individuals in the treatment group on the vertical axis. We illustrate the method using a large smoking cessation trial on varenicline. For example, the modified Gini coefficient of 0.69 can be interpreted as, across each score on urge to smoke, on average 69% more subjects in the varenicline group had less urge to smoke compared with the number of subjects in the placebo group. The modified Gini coefficient optimizes information by integrating across all percentiles of two treatment group distributions in order to enrich the meaning of treatment effect on PROs.