REGRESSION AND CALIBRATION FOR ANALYTICAL SEPARATION TECHNIQUES - PART-I - DESIGN CONSIDERATIONS

Proper use of calibration methods in separation science is an indispen sable prerequisite for accurate and precise quantitative analytical me thods. This paper outlines an economic and straightforward strategy to calibration method development. Special emphasis is put on situations related to separation techniques in drug quality control. Taking prec autions against potential problems while validating the analytical met hod, economic calibration method development includes a number of cons iderations and experiments prior to the first calibration. These prior steps include the specification of the calibration range, estimation of the error due to sample preparation, paying regard to matrix interf erences, deciding upon the use of an internal standard and examination of the robustness and precision of the method under development. Thor oughly considering all these issues accelerates method development, su pports evaluation of the gathered data, and makes method validation ea sier. Experimental designs for calibrations are suggested for method d evelopment (24 measurements), routine and re-calibrations (6 measureme nts). (C) 1997 Elsevier Science B.V.