Development and validation of patient-based exponentially weighted moving average quality control models for three antipsychotic drugs and their metabolites by LC-MS/MS.

BACKGROUND

Therapeutic drug monitoring of antipsychotics is essential for personalized dosing. Traditional internal quality control (IQC) lacks real-time capability, potentially delaying error detection. Patient-Based Real-Time Quality Control (PBRTQC) can overcome this by continuously analyzing patient data. However, its application to antipsychotic drug assays remains underexplored.

OBJECTIVE

To develop and validate novel PBRTQC models based on the exponentially weighted moving average (EWMA) for the real-time monitoring of three first-line antipsychotics (aripiprazole, clozapine, quetiapine) and their major metabolites-addressing a critical gap in laboratory quality assurance for psychiatric pharmacotherapy.

METHODS

Chronologically ordered concentration data were split into training and validation sets. Truncation limits were optimized via trimming and winsorization. Multiple EWMA models with different weighting factors (λ) were evaluated. Model selection was based on the false positive alarm rate (FAR) and the average number of patients before error detection (ANPed), validated using simulated constant and proportional errors.

RESULTS

Optimal models for most analytes employed trimming at 1.5 standard deviation (SD) with λ = 0.03. The introduced PBRTQC system demonstrated high sensitivity: for a 15% analytical bias, the ANPed was <20 for quetiapine, <40 for desalkylquetiapine and aripiprazole, <50 for dehydroaripiprazole, and < 30 for clozapine/norclozapine. All ANPeds fell below 20 at a 30% bias, confirming effective error detection.

CONCLUSION

EWMA-based PBRTQC models specifically tailored for antipsychotic drug monitoring were successfully established. The models provide a practical, real-time complement to traditional IQC, enabling the early detection of analytical shifts and enhancing the reliability of laboratory data for clinical decision-making in psychiatry.