Purpose Dissipation of pharmaceutical compounds entered into the natural environment is an important process minimizing the adverse effects on the living organisms. The aim of the present study is to determine the effects of varying initial concentrations on the stability and degradation of three different pharmaceuticals in a chosen soil. Methods The behavior of three widely used pharmaceuticals (carbamazepine, metoprolol, and sertraline) in soil (Haplic Chernozem on loess) has been investigated using a tandem mass spectrometry (LC–MS/MS)-based analytical strategy. Results The stability of pharmaceuticals follows carbamazepine  >  sertraline  >  metoprolol except for the higher tested concentration, where sertraline exhibits an exceptionally faster dissipation. While the intermediate concentrations (~ 100 ng g −1 ) were generally shown to be the most stable, the fastest dissipation was observed with the highest tested concentration (~ 1000 ng g −1 ) for carbamazepine and sertraline, and for the lowest concentration (~ 10 ng g −1 ) for metoprolol. The fastest dissipation trends observed for carbamazepine and sertraline were likely explained by the higher nutritional values offered by the parent compounds to trigger the microbial growth, whereas the lowest load likely explains the fastest dissipation of metoprolol. The dissipation pattern was more or less similar irrespective of the mode of application (i.e., individually or as a mixture of all) of pharmaceuticals. Three carbamazepine metabolites are also identified, in which the concentrations of carbamazepine-10,11-epoxide mostly increased upon time, whereas the other two (rac trans-10,11-dihydro-10,11-dihydroxy carbamazepine, and 10,11-dihydro carbamazepine) did not increase much during the experiment timescale. Conclusions The present study clearly showed the influence of initial concentrations on the dissipations; however, it did not show any precise (i.e., either increasing or decreasing) trends. More experimental efforts involving much broader sample sizes are, therefore, necessary for a better understanding of this transformation mechanism and a possible description of the nonlinear trend.