Background:
Suggested dose ranges for new medicines often prove to be different from those that are empirically derived in the field. The release of the first long-acting SGA (risperidone CONSTA) suggests that recommended doses are, in the main, too low and the highest dose does not allow the full spectrum of doses that are seen with the oral equivalent. To investigate this, recommended CONSTA doses (in risperidone oral equivalents) are compared with the distribution of doses found in a large pharmacoepidemiological database.
Methods:
From the OPEN/SEER pharmacoepidemiological data archives. Mean doses of SGAs are examined over time and most likely mean stable doses calculated. These are compared with projected mean doses and are further compared with suggestions based on the Seeman's radioligand-free Kis. CONSTA is examined in particular.
Results:
Projected final mean doses are summarized with respect to estimated near-maximal effective doses (Davis et al. 2003). With respect to CONSTA, the range of doses appears to be equivalent to only half of that available with the oral form.
Conclusions:
This suggests that careful attention needs to be paid to patient selection. The current selection guidelines take in to account the censored dose range and it is suggested that this may influence those correctly assigned to therapy with this new medication form. It also suggests that the empirical finding that a proportion of patients require doses higher than the recommended maximum is consistent with known RISe distributions in clinical practice.
