Help! I'm not a chemist, just a statistician turned QA Manager turned QA/QC Manager, so please explain very simply. We are trying to use a new USP method for Sodium Sulfadimethoxine assay. Our old method didn't have enough precision to meet new USP specifications (99.0-100.5). The USP method calls for a 300:100 mixture of monobasic sodium phosphate buffer to methanol for the mobile phase. Both the standard (50 mg) and sample (60 mg) are diluted in 250 ml mobile phase. The standard is Sulfadimethoxine, not Sodium Sulfadimethoxine. Aqueous portion of mobile phase is adjusted to ph 7 prior to adding methanol. The system is an HP 1050 with Spherisorb ODS-2 column. Flow rate 1 ml/min. We premix the mobile phase. We ran a qualification using a sample we knew to be good using 2 analysts on 2 days. Everything looked good. Today we ran some more samples, one of which was in spec using the old method, and the rest of which had injections in-injections out, preps in-preps out using the old method. Now they are all 1-2% lower than when we used the old method. I've also noticed that the retention times shifted today after about the 15th to 20th injection for both standards and samples, and the peak area of the standards increased very slightly. We inject 2 standards every four samples. I thought standards were supposed to compensate for equipment drifts. Should I suspect the standard prep? The check was 99.5%. Should the chemists be degassing the mobile phase more? Should we decrease the flow rate? Clean the column more often? What is the cause? Any input is appreciated.

If I understand you correctly, you are measuring the peak area of samples and standards with two different methods (however, you describe only one method). The amount of sulfadimethoxine is decreasing with time, but it is based on the calculation using internal standards whose area increases with time.
If this is the problem, what is the surprise? your internal standards are evaporating slowly, while a new sample does not. Check the seals of the vials with the internal standards or use more vials of interanl standards.
If this is not the problem, please clarify, and we will try again.

I don´t think she's using any IS. But I agree, the description is a bit confuse.
Are you using a column oven? Peak shift could be due to changes in room temperature during the day.
And, as said by the Anonymous, standars sol. in vials using red rubber seal can suffer evaporation through the day. You can use new vials for each group of standard injections to solve that problem.
If I understand you are comparing values obtained by using the old method with those using the new one, aren't you?
If so, ¿Have you performed any validation work regarding the method transference?
Methods seem to be biased, you can use your statistical knowledge !!

Hope this helps. Good luck.