Prediction of Promotional Effect Using Neural Network Modeling

The effect promotional spending has on the prescribing activities of physicians has been a subject of ongoing discussion in the pharmaceutical industry. Although heavy promotion may establish access to physicians, thereby increasing market share, these expenditures must be judged in the context of the entire earnings of a company to ensure a healthy return on investment (ROI). Nevertheless, little detailed material is published about the methodological aspects of measuring and predicting the effects of promotion in the pharmaceutical industry. We address this oversight by investigating the quantification and prediction utility of a neural network in establishing the cause-and-effect relationship between promotional spending and variation in (the volume of) prescriptions. Our work presents some evidence of the nonlinear relationship between promotional spending and prescription responsiveness. The validity of our approach was further evaluated using unknown sample point, and the predictive power is superior to the benchmark obtained by multiple linear regression analysis.