Ask Mallard

Interrupted time series

An interrupted time series (ITS) evaluates whether the level or trend of an outcome changes after a clearly timed intervention or event.
NHMRC EVIDENCE LEVEL · 2009Depends on how the study is designed · InterventionProvisional design classification. Study quality and certainty of findings require separate appraisal.
Why this level?Specify whether this interrupted time series has a parallel control group. NHMRC 2009 source

NHMRC names an interrupted time series twice, once on each side of whether there is a control group: III-2 with a parallel control, III-3 without one. Which one applies is a fact about your study, not about the design.

Where the NHMRC 2009 hierarchy names this design, it places it here:

Research questionLevelWhen
InterventionIII-2The interrupted time series includes a parallel control group.
InterventionIII-3The interrupted time series has no parallel control group.

What question does this help answer?

“Did an outcome trajectory change when a policy, program, clinical intervention, or system change was introduced?”

When to use it

ITS is useful when an intervention occurs at a clearly defined time and repeated outcome measurements are available for sufficient periods before and after the intervention.

When not to use it

Avoid a simple ITS when too few time points exist, the intervention date is unclear, major co-interventions occur at the same time, or the outcome has strong seasonal or autocorrelated patterns that cannot be modeled adequately.

How it works in plain language

The analysis estimates the pre-intervention trend, any immediate level change after the intervention, and any change in the post-intervention slope. A controlled ITS can strengthen the design by comparing with an unaffected series.

What a level change and a slope change look like

An interrupted time series. Points before the intervention follow a rising trend. After it, the series jumps up and rises more steeply. A dashed line continues the earlier trend as the counterfactual, and the gaps between it and the later line are the level change and the slope change.Interrupted time series: the trend before, and what it would have doneTimeOutcomeIntervention startsCounterfactual: the pre-period trend continuedLevel changeSlope changeThe counterfactual is an assumption, not an observation. Several points before the interruption are neededto estimate the trend it extends.
Conceptual diagram. It shows the structure of the idea, not data from any study.

Why Mallard may recommend it

Mallard may recommend ITS for health-system policy changes, formulary changes, clinical decision support launches, staffing interventions, or public-health policies that affect a population at a known time.

Key assumptions and design requirements

  • The intervention timing is well defined.
  • The pre-intervention trend provides a credible counterfactual trajectory.
  • Autocorrelation and seasonality are evaluated.
  • Other events occurring at the intervention time are considered.
  • The hypothesized impact shape is preferably specified before analysis.

What to watch out for

A visible before/after difference is not enough. Ignoring pre-existing trends, autocorrelation, seasonality, or concurrent changes can produce misleading estimates.

Clinical example

Evaluate monthly inappropriate imaging rates for 24 months before and 24 months after implementation of an electronic decision support tool.

How to report it

No reporting guideline dedicated to interrupted time series was found in a PubMed search on 2026-09-16. STROBE is the guideline to use, with RECORD where the series comes from routinely collected data. Report the number and spacing of points before and after, the exact date of the interruption and why it is that date, and how autocorrelation and seasonality were handled. A reporting guideline is not a design manual: use it to make the study transparent, not as a substitute for careful design. Check the target journal's own requirements as well, which this page cannot know.

Code for this method
Loading…
Recent studies that used this
Searching PubMed…
Part of the learning path Evaluate a change on your own ward (step 4 of 6).
Related Learn pages
Sources & further reading
  1. Interrupted time series regression for the evaluation of public health interventions: a tutorial (opens in a new tab) · Int J Epidemiol · doi 10.1093/ije/dyw098
  2. Segmented Regression in an Interrupted Time Series Study Design (opens in a new tab) · Anesth Analg · doi 10.1213/ANE.0000000000005269
  3. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies (opens in a new tab) · BMJ · doi 10.1136/bmj.39335.541782.AD
  4. The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement (opens in a new tab) · PLoS Med · doi 10.1371/journal.pmed.1001885
Last source review: 2026-09-05
AI-generated educational content
Mallard uses reputable methodological and reporting sources, but this material may contain errors or omit important context. Verify important methodological decisions with the cited sources and, when appropriate, a qualified research methodologist or statistician.