Article type
Year
Abstract
Background:
When it is not possible to conduct the gold standard of randomised control trials (RCTs), interrupted time series (ITS) studies are an alternative design. ITS studies are considered one of the strongest quasi-experimental designs to evaluate healthcare interventions, and are increasingly being included in systematic reviews, especially in reviews from the Cochrane Effective Practice and Organisation of Care Group.
In an ITS design, data are collected at multiple time points before and after the intervention. The design compares whether the post-intervention data are different from the patterns or trends observed in the pre-intervention data. Any differences are then attributed to the intervention. The two most common ways to report results from ITS studies are the change in level (difference in predicted means at specific points in time) and change in slope (difference between pre- and post-intervention trend lines). Both results are required to interpret the short and longer term effects of the intervention. These two estimates are often presented in table format which can limit the interpretation of the overall effect of an intervention across multiple ITS studies.
Methods:
We developed a novel graphical tool to present the results from multiple ITS studies in a systematic review. The change in level estimates was standardised and plotted against the change in slope for each study, weighted by standard error of the effects. We will present examples of the plot from existing reviews.
Conclusions:
The ITS plot aids reviewers in summarising ITS results by displaying the heterogeneity of results in a single plot.
The graphical presentation of the results helps policy makers, patients, and healthcare consumers to understand how the intervention affects the outcome of interest.
When it is not possible to conduct the gold standard of randomised control trials (RCTs), interrupted time series (ITS) studies are an alternative design. ITS studies are considered one of the strongest quasi-experimental designs to evaluate healthcare interventions, and are increasingly being included in systematic reviews, especially in reviews from the Cochrane Effective Practice and Organisation of Care Group.
In an ITS design, data are collected at multiple time points before and after the intervention. The design compares whether the post-intervention data are different from the patterns or trends observed in the pre-intervention data. Any differences are then attributed to the intervention. The two most common ways to report results from ITS studies are the change in level (difference in predicted means at specific points in time) and change in slope (difference between pre- and post-intervention trend lines). Both results are required to interpret the short and longer term effects of the intervention. These two estimates are often presented in table format which can limit the interpretation of the overall effect of an intervention across multiple ITS studies.
Methods:
We developed a novel graphical tool to present the results from multiple ITS studies in a systematic review. The change in level estimates was standardised and plotted against the change in slope for each study, weighted by standard error of the effects. We will present examples of the plot from existing reviews.
Conclusions:
The ITS plot aids reviewers in summarising ITS results by displaying the heterogeneity of results in a single plot.
The graphical presentation of the results helps policy makers, patients, and healthcare consumers to understand how the intervention affects the outcome of interest.