Interrupted Time Series Analysis

Abstract

Interrupted Time Series Analysis develops a comprehensive set of models and methods for drawing causal inferences from time series. Example analyses of social, behavioural, and biomedical time series illustrate a general strategy for building AutoRegressive Integrated Moving Average (ARIMA) impact models. The classic Box-Jenkins-Tiao model-building strategy is supplemented with recent auxiliary tests for transformation, differencing and model selection. New developments, including Bayesian hypothesis testing and synthetic control group designs are described and their prospects for widespread adoption are discussed. Example analyses make optimal use of graphical illustrations. Mathematical methods used in the example analyses are explicated assuming only exposure to an introductory statistics course. Design and Analysis of Time Series Experiments (DATSE) and other appropriate authorities are cited for formal proofs. Forty completed example analyses are used to demonstrate the implications of model properties. The example analyses are suitable for use as problem sets for classrooms, workshops, and short-courses.

The effects of a leaflet‐based intervention, ‘Hypos can strike twice’, on recurrent hypoglycaemic attendances by ambulance services: A non‐randomised stepped wedge study

“…This approach seemed to be more robust because it accounted for the secular trends prior to the intervention period and it also allowed the alignment of all clusters at the time the intervention started, directly comparing all calls prior to the intervention and post intervention. 25,26 Thus, the results of ITSA indicated a reduced number of unsuccessful calls in the post-intervention period when aligning all clusters which was maintained for clusters two and three (details are presented in the Supporting Information, Section B). Moreover, the time to a repeat attendance was longer in the intervention period when compared to the non-intervention period (see the Supporting Information, Section E, Figure S.E1).…”

COVID-19 Leads to Accelerated Increases in Children's BMI z-Score Gain: An Interrupted Time-Series Study

“…Time series data consist of a continuous sequence of repeated observations collected at consistent intervals in a sample of participants. 46 In an interrupted time series design, the sequence of repeated observations are "interrupted" by a naturally occurring intervention at a known timepoint. 47,48 In the current study, the repeated observations consisted of collecting the heights and weights of all children at 3 participating schools each fall from 2017 to 2020.…”

Assessing The Impact of A Restrictive Opioid Prescribing Law in West Virginia

“…Statistical analysis was done utilizing an autoregressive integrated moving average (ARIMA) interrupted time series analysis (ITS) by a trained statistician. ITS analysis is particularly well equipped to evaluate interventions [18,19], and the ARIMA model is one of the most common interrupted time series methods [20] and widely used in health care research [17,[21][22][23]. ARIMA was first introduced by Box and Jenkins in 1976 [24] that combined Auto Regressive (AR) model and Moving Average (MA) model to forecast stationary and nonstationary time series.…”