In the December Improvement science and research methods seminar, Professor Graham Law discussed ‘Interrupted time series, COVID lockdown and ambulance data’. Graham Law is a Professor in Medical Statistics in the College of Social Science and a senior member of CaHRU. He has published extensively in epidemiology and medical statistics, focussing on sleep and the consequences of good and poor sleep on metabolic and cardiovascular health.
Interrupted Time Series Analysis is a quasi-experimental design that can be used to evaluate interventions using longitudinal data collected over a period of time period. It is arguably the “next best” approach for dealing with interventions when randomisation is not possible, for example in the case of ‘natural experiments’. The COVID-19 lockdown provided a natural experiment involving a clear ‘intervention’ that allowed this approach.
Graham explained the methods, how these can be interpreted, the drawbacks and future areas for research. He started by explaining the concept of a conterfactual. He then went on to describe time series and the approach to modelling this mathematically. He them discussed approaches to timing of the interruption, the shape of the distribution, the importance of stationary data and the use of moving averages using ambulance data during the period of COVID-19 lockdown to illustrate this.
To view the presentation please click below.