Analyzing the Consumption of Various Pharmaceuticals in Turkey: A Panel Approach

Objective: The aim of this study is to analyze the pharmaceutical consumption in Turkey and investigate different types of models explaining this consumption by using the panel approach. Material and Methods: Different panel data models were constructed by using the panel data regression in order to explain the relationship between the sales and the total production costs of various pharmaceutical forms supplied by one of the leading pharmaceutical manufacturing companies in Turkey. In this study, 5 different forms of pharmaceuticals concerning general or beta-lactam pharmaceuticals were examined over 72 periods. Results: It can be seen that pharmaceutical consumption including general and beta-lactam pharmaceuticals was generally affected by time. Unit root analysis was performed for the variables used in this study and it was determined that the variables were stationary. Moreover, log-linear forms of the models were used in order to stabilize the variations in the variables. The findings implied that the dynamic panel two-way fixed effect model explained the relationship between the sales and the total production costs better than the others because of R2 = 0,952, RMSE = 0,242, J-stat = 0,175 and the Hausman test result. Conclusion: The sales of various pharmaceutical forms over the long period were modeled by using the panel data regression. This analysis allows modeling of different approaches. It can be concluded that dynamic panel data regression models can also give good results in the studies related to health sciences depending on time.