UV-Assay Cost Structure

As long as the health authorities accept 90-110% specification limits on the drug assay, the normalization method presented above will barely suffice for batch release purposes. Since there is a general trend toward tightening the specification limits to 95-105% (this has to do with the availability of improved instrumentation and a world-wide acceptance of GMP-standards), a move toward options 1 (HPLC) and 2 (DA-UV) above is inevitable. 

For an example of curve fitting involving classical propagation of errors in a potentiometric titration setting, see Ref. 142. 

Problem The drug substance in a pharmaceutical product is to be assayed in an economical fashion. The following constraints are imposed  

Options The analyst elects to first study photometry and place the three reference concentrations symmetrically about the nominal value (= 100%). The initial test procedure consists of using references at 80, 100, and 120% at the beginning of the series, and then a 100% reference after every fifth determination of an unknown sample. 

First Results The confidence interval from linear regression C1(X) was found to be too large in effect, the plant would have to produce with a 0% tolerance for drug substance content a tolerance of about 1% is considered necessary, cf. Fig. 2.12 and Section 4.24 (Fig. 4.35). 

The relative standard deviation of the determination was found to be 0,5% (photometer) resp, 0.7% (HPLC ) for samples and references. The relative content varies by nearly 1% due to inhomogeneities in the tablet material and machine tolerances. 

---