Related Experiment Videos
Interrupted time series analysis in clinical research.
Lloyd K Matowe1, Cathie A Leister, Concetta Crivera
1Faculty of Pharmacy, Department of Pharmacy Practice, Kuwait University, Safat, Kuwait. l.matowe@hsc.kuniv.edu.kw
The Annals of Pharmacotherapy
|July 5, 2003
Summary
Interrupted time series analysis is useful for clinical trials. This method showed significant QTc interval changes in simulated patients receiving experimental medication, demonstrating its value in patient self-controlled studies.
Area of Science:
- Clinical trial design
- Biostatistics
- Pharmacodynamics
Background:
- Clinical trials often require robust statistical methods to analyze serial data.
- Interrupted time series (ITS) analysis is a statistical technique that can be applied to longitudinal data.
- ITS is particularly useful when participants serve as their own controls.
Purpose of the Study:
- To evaluate the utility of interrupted time series analysis in the context of clinical trial design.
- To demonstrate how ITS can be applied to analyze simulated safety data from a Phase I clinical study.
Main Methods:
- Simulated electrocardiographic (ECG) data from 18 healthy volunteers were used.
- Data represented serial ECGs collected before and during treatment with an experimental medication.
- SAS software was employed for visual inspection and statistical testing of trend, seasonality, and autocorrelation.
Main Results:
- No significant trend, seasonality, or autocorrelation was detected in the simulated data.
- Statistically significant changes in QTc intervals were observed in 11 out of 18 simulated individuals post-treatment.
- A significant interaction between time of day and treatment effect was noted in 4 simulated patients.
Conclusions:
- Interrupted time series analysis provides a valuable tool for clinical research.
- ITS is suitable for studies where patients act as their own controls and data are collected at regular intervals.
- This method enhances the analysis of serial data in clinical trial settings.