Interrupted Time Series Analysis
A quasi-experimental design method that estimates the level change and slope change brought about by an intervention by analyzing data from multiple time points before and after the intervention. It is used for causal evaluation.
Timeline (last 90 days)
Interrupted time series analysis (ITS) uses observations at multiple time points before and after the implementation of an intervention to estimate the level change and slope change brought about by the intervention, and can partially control for trends that evolve naturally over time.
The limitation of ITS is that it is difficult to rule out other events occurring simultaneously with the intervention (such as policy adjustments or seasonal epidemics), so it usually needs to be used in conjunction with a control group or sensitivity analysis.
Randomized trials, controlled implementation studies, and interrupted time series are not new in academia, but in the actual evaluation of medical AI, their frequency of use is far lower than simple predictive performance reporting and before-and-after comparisons.
All Facts (3)
Interrupted time series analysis (ITS) uses observations at multiple time points before and after the implementation of an intervention to estimate the level change and slope change brought about by the intervention, and can partially control for trends that evolve naturally over time.
50%UnverifiedThe limitation of ITS is that it is difficult to rule out other events occurring simultaneously with the intervention (such as policy adjustments or seasonal epidemics), so it usually needs to be used in conjunction with a control group or sensitivity analysis.
50%UnverifiedRandomized trials, controlled implementation studies, and interrupted time series are not new in academia, but in the actual evaluation of medical AI, their frequency of use is far lower than simple predictive performance reporting and before-and-after comparisons.
50%