Fact-checked by Grok 2 weeks ago

Interrupted time series

Interrupted time series (ITS) analysis is a quasi-experimental research design used to evaluate the causal impact of an intervention, policy, or event on an outcome variable by examining longitudinal data collected at multiple, equally spaced time points before and after the interruption. This approach models the time series to detect changes in the level (immediate shift in the outcome) or slope (alteration in the trend) attributable to the intervention, while accounting for underlying patterns such as autocorrelation, seasonality, and secular trends. ITS is particularly valuable in settings where randomized controlled trials are impractical, such as population-level health or social policy evaluations, as it provides robust evidence by leveraging pre-intervention data as a counterfactual. The ITS design was introduced in 1963 by and Julian C. Stanley as a powerful quasi-experimental alternative to true experiments, especially for assessing educational and social interventions where ethical or logistical constraints prevent . They emphasized its strength in ruling out many threats to , such as maturation, history, and instrumentation, through repeated observations that establish a baseline trend. Building on this, and George C. Tiao developed formal statistical methods in 1965, introducing intervention analysis techniques for non-stationary to test for level shifts and trend changes using autoregressive integrated moving average (ARIMA) models. These foundational contributions established ITS as a cornerstone of in observational data, with subsequent refinements incorporating for simpler applications. ITS has been widely applied across disciplines, notably in to assess the effects of legislation like smoke-free policies or vaccination campaigns, in for policy evaluations such as minimum wage increases, and in for analyzing the impact of laws on crime rates. Key advantages include its ability to handle time-varying confounders and provide effect estimates over time, making it superior to simple pre-post comparisons. However, effective implementation requires at least 12 data points pre- and post-intervention to ensure statistical power when using monthly data, and assumptions like the absence of concurrent events or stable pre-intervention trends must hold to avoid bias. Modern extensions, such as multiple-group ITS or Bayesian approaches, address these limitations for complex, clustered data.

Overview

Definition

An interrupted time series (ITS) is a quasi-experimental that involves collecting multiple observations of an outcome variable at regular intervals over an extended period, with the time series divided into pre- and post- phases to evaluate the effects of a specific . This design detects potential changes in the level (immediate shift) or trend (slope over time) of the outcome that can be attributed to the , such as a implementation or program rollout, by comparing patterns before and after the interruption point. Unlike true experiments that employ to to establish , ITS relies on temporal sequencing and the natural variation in time-series data rather than , making it particularly suitable for real-world settings where ethical or practical constraints prevent controlled trials. The core components include the time series data itself—typically comprising repeated measures of outcomes like rates, counts, or continuous variables (e.g., monthly hospitalization rates)—the clearly defined acting as the interruption, and the analytical focus on discernible shifts in the series' trajectory post-intervention.

Core Principles

Interrupted time series (ITS) analysis relies on the principle of level change, which detects an immediate shift in the outcome variable's value immediately following the , representing a discontinuous jump from the pre- trajectory. This abrupt alteration contrasts with the ongoing trend observed prior to the interruption and serves as a primary indicator of the 's short-term effect. Complementing the level change, the principle of slope change examines alterations in the trend or rate of change in the outcome after the , signaling a sustained or long-term impact. For instance, if the pre-intervention trend shows a gradual increase, a post-intervention change might accelerate or decelerate this progression, thereby quantifying the intervention's influence on the underlying . Visually, ITS is represented through time-series plots that display the outcome variable over sequential time points, with a vertical line marking the point to separate pre- and post- phases. These plots typically fit linear trends to the pre- data to infer a counterfactual—what the outcome would have been without the —and compare it to the observed post- path, highlighting any deviations in level or slope. In , ITS enhances by leveraging multiple observations before and after the to account for time-dependent confounds such as maturation and regression to the mean, thereby isolating the intervention's effect more robustly than simpler pre-post designs. This repeated measurement strengthens the design's ability to rule out alternative explanations for observed changes. Reliable estimation of trends in ITS requires a minimum of 8-12 data points both pre- and post-intervention to capture underlying patterns and detect meaningful changes with sufficient statistical power. Fewer observations may obscure true effects or inflate Type II errors, while more extensive series improve precision.

History

Origins

The interrupted time series (ITS) design was first systematically introduced as a quasi-experimental method by psychologists Donald T. Campbell and Julian C. Stanley in their 1963 monograph Experimental and Quasi-Experimental Designs for Research. They classified the time-series experiment—characterized by multiple observations before and after an intervention—as one of the stronger alternatives to randomized controlled trials (RCTs) when randomization is infeasible, particularly for its ability to control for maturation, instrumentation, and testing effects through repeated measures. This framework positioned ITS within broader quasi-experimental research, emphasizing its utility for causal inference in naturalistic settings without random assignment. During the and , ITS saw early applications in and to evaluate non-randomized programs. These applications highlighted ITS's strength in detecting intervention-induced shifts amid ongoing temporal processes, fostering its adoption in program . The methodological roots of ITS also drew from , building on pre-existing analysis techniques for policy evaluation, notably the autoregressive integrated moving average () models introduced by and Gwilym M. Jenkins in their 1970 book Time Series Analysis: Forecasting and Control. These models, originally developed for forecasting, were adapted to assess interventions by modeling structural breaks in data, enabling the detection of policy impacts on economic indicators. This econometric influence provided a rigorous statistical foundation for ITS, bridging descriptive with . A key milestone in formalizing ITS for the social sciences came with the 1980 publication of Interrupted Time Series Analysis by David McDowall, Richard McCleary, Errol E. Meidinger, and Richard A. Hay Jr., which integrated ARIMA-based intervention analysis with quasi-experimental principles to offer practical tools for researchers in fields like and . The book emphasized model specification, estimation, and diagnostics tailored to social data, solidifying ITS as a standard method for evaluating interventions in non-experimental contexts.

Key Developments

In the , interrupted time series (ITS) analysis saw significant methodological refinement through the integration of techniques, which allowed for more precise estimation of effects on both level and trend changes in time series . This approach was particularly advanced in research, where Wagner et al. provided comprehensive guidelines in 2002 for applying to evaluate longitudinal effects of interventions, such as medication use policies, emphasizing its quasi-experimental strength over simpler before-after designs. These developments built on earlier foundations by enhancing interpretability for policy impacts without requiring randomization. The marked the rise of Bayesian approaches in ITS, offering flexible handling of uncertainty and complex structures in data. A pivotal contribution was Brodersen et al.'s 2015 introduction of structural models within a Bayesian framework to infer causal impacts, using state-space representations to predict counterfactual outcomes and quantify effects in settings like and . This method, implemented in tools like the CausalImpact, addressed limitations of classical models by incorporating priors and enabling robust inference even with sparse data. Computational advances further propelled ITS adoption through dedicated software in the 2010s. In , the itsadug package, developed around , facilitated the analysis of autocorrelated using generalized additive mixed models (GAMMs), supporting and model diagnostics for evaluations. Similarly, Stata's itsa command, introduced by in , streamlined single- and multiple-group ITS analyses via ordinary least-squares , making the method accessible for program evaluations in and sciences. These tools democratized ITS by reducing analytical barriers and promoting standardized reporting. Post-2010 trends emphasized controlled ITS designs and integrations with advanced techniques to mitigate , enhancing causal validity in observational settings. Linden and Adams's 2011 propensity score-based weighting model, for instance, improved inference by reweighting control series to mimic treated units' pre-intervention trajectories, particularly useful in evaluations with non-equivalent groups. This approach, extended in subsequent works, has intersected with methods for confounder adjustment, such as synthetic controls, allowing hybrid models to handle high-dimensional covariates in real-world applications. Global adoption of ITS surged post-2000, with publication volumes increasing substantially due to its utility in evaluating interventions amid major events. This spike was driven by applications assessing impacts like those from the on healthcare access and outcomes, where ITS provided timely evidence for measures and reform effects in multiple countries. By the 2020s, ITS had become a cornerstone for population-level studies. The further accelerated its use, with numerous studies applying ITS to assess the effects of lockdowns, vaccination programs, and other measures on infection rates, hospitalizations, and mortality.

Design Elements

Basic Design

The basic design of an interrupted time series (ITS) study involves observing a single outcome over multiple time points before and after the introduction of an , without the use of a comparison group. This single-group approach relies on the longitudinal data from one unit—such as a , , or —to establish a trend and detect potential changes attributable to the . Implementation begins with selecting an outcome variable that is measurable and relevant to the intervention, such as admission rates or fatalities. Next, the intervention timing must be precisely defined, typically as a specific when the change takes effect, enabling a clear demarcation between pre- and post-periods; this is crucial for abrupt interventions like implementations, though gradual rollouts (e.g., phased program introductions) can also be accommodated by noting the start of the transition. Data collection then proceeds at equally spaced intervals, such as monthly or quarterly, to ensure consistency and capture the temporal pattern. The structure consists of time-indexed observations on the outcome for the single unit, with a sufficient number of pre- points—at least 8–12—to reliably establish the level and trend, allowing for the assessment of any immediate level shift or change in slope following the intervention. Post- should mirror this in quantity to balance the series and enhance power, often drawn from routine administrative records or archives for reliability. A representative example is the evaluation of England's 2007 smoke-free legislation on childhood hospital admissions, using monthly data from April 2002 to November 2010 (63 pre- and 41 post- observations) to model baseline trends and observe post- changes.

Controlled Designs

Controlled interrupted time series (CITS) designs enhance the basic interrupted time series (ITS) approach by incorporating one or more comparison groups unaffected by the , allowing for a more robust estimation of the counterfactual trajectory and strengthening causal inferences. These designs address limitations of single-series ITS, such as vulnerability to concurrent events, by enabling direct comparison of changes in trends between treated and control units. Multiple-group ITS, also known as comparative ITS, involves adding a control series from units not exposed to the to estimate the counterfactual trend for the treated group. The analysis typically uses a model that includes interaction terms between time, status, and group membership to test for pre- parallelism and estimate effects on level and changes relative to the . For instance, parameters in the model capture baseline differences in intercepts and slopes between groups before the , as well as differential changes post-, facilitating tests of group comparability. This approach has been applied in evaluations of policies, such as assessing the impact of reductions on road injuries by comparing implemented areas with similar untreated regions. A hybrid of difference-in-differences (DiD) and ITS integrates the parallel trends assumption from DiD with the time-series structure of ITS to test and model both immediate level shifts and gradual slope changes post-. In this design, pre-intervention trends between treated and control groups are compared to validate assumptions, and post-intervention deviations are analyzed to isolate the effect, often using fixed-effects models or . This method is particularly useful in educational evaluations, where it has demonstrated greater precision than standard DiD under certain conditions, such as when interventions cause sustained trend alterations. Propensity score-weighted ITS, introduced by Linden and Adams, employs a weighting scheme to balance treated and control units based on pre-intervention covariates, thereby creating a counterfactual that mimics the treated group's characteristics. The process involves estimating propensity scores via on baseline data, assigning weights (e.g., inverse probability weights for the on the treated) to control observations, and then applying weighted to assess impacts. This technique was used to evaluate California's Proposition 99 program, revealing a significant reduction in cigarette sales compared to weighted controls, while accommodating multiple treated units without requiring complex matching. Synthetic control ITS extends the by constructing a weighted composite series from multiple untreated units to approximate the treated unit's pre-intervention , particularly when no single matches closely. Originating from Abadie et al.'s framework for comparative case studies, it optimizes weights via minimization of pre-intervention differences in outcomes and predictors, then extrapolates this synthetic series post-intervention to estimate effects in time-series contexts. Applications include policy evaluations like the program, where the synthetic better captured secular trends than traditional controls, enhancing inference in settings with heterogeneous units. These controlled designs collectively reduce threats from historical events or maturation by focusing on relative rather than absolute changes, improving the plausibility of causal attributions in observational settings.

Statistical Methods

Segmented Regression

represents the foundational method for analyzing interrupted time series (ITS) data, employing to model changes in outcome levels and trends before and after an . This approach decomposes the time series into pre- and post- segments, allowing estimation of immediate level shifts and alterations in the underlying trend slope attributable to the . By fitting a single model across the entire series, provides a straightforward framework for quantifying effects while accommodating the temporal structure of the data. The core model equation for segmented regression in ITS analysis is given by: Y_t = \beta_0 + \beta_1 t + \beta_2 T_t + \beta_3 (t - \tau) T_t + \varepsilon_t where Y_t denotes the outcome at time t, t is the time index, T_t is a indicator (0 for pre- periods and 1 for post-), and \tau marks the time point. Here, \beta_0 captures the level at time zero, \beta_1 estimates the pre- trend , \beta_2 quantifies the immediate level change at the intervention, and \beta_3 measures the post- change in relative to the pre- trend. The error term \varepsilon_t accounts for unexplained variation, typically assumed to follow a . Estimation proceeds via ordinary least squares (OLS) to obtain parameter coefficients, with adjustments for serial autocorrelation in the errors using Newey-West standard errors to ensure robust inference. This heteroskedasticity- and autocorrelation-consistent (HAC) estimator corrects for potential correlation in residuals, which is common in time series data, thereby producing valid confidence intervals and p-values. Interpretation centers on the coefficients \beta_2 and \beta_3, where statistical tests (e.g., t-tests) assess their significance to determine the presence of an immediate effect (level change) or a gradual effect (slope change) due to the intervention. Visualization aids comprehension, typically plotting the observed data points alongside the fitted regression lines for pre- and post-intervention segments to illustrate discontinuities or trend shifts. The model assumes in the pre- and post-intervention trends, absence of in residuals (or appropriate adjustment via robust errors), and homoscedasticity (stable variance) across time periods. Violations of these can bias estimates, though the Newey-West correction mitigates issues. Implementation is accessible in statistical software; in , the lm() function fits the model, with the sandwich package providing NeweyWest() for robust standard errors. In , the itsa command automates for ITS, incorporating options for adjustments.

Advanced Modeling Techniques

Advanced modeling techniques in interrupted time series (ITS) analysis extend beyond basic by incorporating dynamic dependencies, probabilistic frameworks, and flexible functional forms to better capture complexities in time series data, such as and non-linear trends. These methods are particularly useful when pre- and post-intervention data exhibit serial correlation or require on counterfactual outcomes without strong assumptions. Autoregressive integrated moving average () models, developed through the Box-Jenkins approach, provide a foundational extension for ITS by explicitly modeling the underlying structure before incorporating the intervention effect. An model decomposes the series into autoregressive () components that capture dependence on past values, integrated (I) components that account for non-stationarity through differencing, and moving average () components that model the influence of past errors. In ITS applications, the intervention is typically represented as a , such as a for permanent level shifts or a pulse function for transient impacts, allowing estimation of both immediate and gradual effects while adjusting for . This approach, introduced in seminal intervention analysis work, enables robust of counterfactual trends and is widely implemented in software like R's forecast package for evaluating population-level interventions. Bayesian structural time series (BSTS) models offer a probabilistic alternative, using state-space formulations to decompose the into trend, , and components while enabling posterior on impacts. In this framework, the observed outcome is modeled as a of latent states, with priors placed on parameters to incorporate uncertainty; the effect is inferred by comparing the posterior distribution of the observed series to a counterfactual predicted from covariates and historical patterns. The CausalImpact , based on this , automates model fitting and provides credible intervals for the causal effect, making it suitable for and evaluations where multiple predictors influence the outcome. This approach, formalized in key work on , excels in handling sparse data and providing intuitive visualizations of impact uncertainty. Non-parametric methods, such as local regression and (locally estimated scatterplot smoothing), allow flexible trend estimation in ITS without assuming global linearity or specific functional forms, making them ideal for data with irregular or unknown patterns. Local regression fits a low-degree (often linear or ) to weighted subsets of nearby data points using kernel functions, producing a smooth estimate of the underlying trend; in ITS, separate fits pre- and post-intervention enable assessment of changes in level or slope via comparison of the resulting curves. extends this by adaptively selecting polynomial degrees and bandwidths, providing robust smoothing even with outliers. These techniques, applied in generalized additive models for ITS, facilitate visual and statistical evaluation of intervention effects in contexts where parametric assumptions may fail. Seasonality in ITS data, common in monthly or quarterly observations, can confound trend estimates unless explicitly modeled; terms or dummy variables offer effective adjustments. terms decompose seasonal cycles into pairs of functions at frequencies, capturing periodic patterns with a parsimonious set of parameters (e.g., 2K terms for K seasonal periods), which are included as regressors in the ITS model to isolate the intervention effect. Alternatively, dummy variables assign binary indicators to each seasonal period (e.g., months), allowing the model to estimate period-specific intercepts while controlling for cyclic variation. These methods, recommended in ITS tutorials for evaluations, ensure unbiased estimates of intervention impacts by removing seasonal artifacts without overparameterization. For scenarios involving multiple interventions, ITS models can be extended by incorporating additional dummy variables to represent each interruption, enabling estimation of cumulative or interactive effects over time. Each dummy captures a step change at its respective point, with interaction terms between dummies and time allowing assessment of slope modifications; this multivariate approach accounts for sequential policy changes without assuming independence between events. Such modeling, implemented in tools like Stata's itsa command, is valuable for complex interventions like phased rollouts, providing granular insights into relative contributions.

Assumptions and Validity

Core Assumptions

Interrupted time series (ITS) analysis relies on several foundational statistical assumptions to enable about an 's impact. These assumptions ensure that observed changes in level or at the intervention point reflect the intervention's effect rather than influences or modeling artifacts. Primarily drawn from frameworks, they emphasize the need for a reliable counterfactual based on pre- . A stable pre-intervention trend is crucial, assuming no abrupt changes or external events disrupt the baseline trajectory before the , allowing for a reliable of what would have occurred absent the . This stability facilitates the detection of intervention-induced shifts in level or slope. The absence of effects is another key assumption, positing that the 's impact begins precisely at the date, with no behavioral or preparatory changes influencing the outcome in the lead-up period. Violations, such as preemptive adjustments by stakeholders, could bias estimates by mimicking or masking the true . Standard regression assumptions of homoscedasticity and must hold for the error terms in the model. Homoscedasticity requires constant variance across time periods, while assumes no correlation among errors, which can be assessed using the Durbin-Watson test. These properties ensure efficient and unbiased parameter estimates in segmented models. In controlled ITS designs, parallel trends are assumed between the treatment and control series prior to the , meaning both would have followed similar trajectories without the , strengthening causal claims by accounting for common time-varying confounders. Finally, sufficient observations are required for adequate statistical , with at least 8 observations pre- and post-, for a total of at least 16 data points, recommended to reliably detect meaningful changes in level or trend. This typically involves multiple pre- and post- time points to establish trends and estimate effects precisely.

Threats to Validity

One primary threat to the validity of interrupted time series (ITS) analyses is the history effect, where external events or co- occurring around the time of the mimic or confound the observed changes in the outcome. For instance, an economic shift or unrelated policy change coinciding with the could produce trends that are erroneously attributed to the itself. This threat is particularly acute in single-group designs lacking series, as it undermines by introducing unobserved confounders. Mitigation strategies include incorporating control groups or series unaffected by the to isolate its effects, thereby enhancing . Autocorrelation in time series data represents another significant validity concern, as serial dependence among observations can lead to underestimated standard errors and inflated Type I error rates, resulting in false positives for intervention effects. This occurs when residuals from the model are correlated over time, violating the independence assumption in standard regression. To address this, analysts can employ robust standard error estimation, such as Newey-West adjustments, or more advanced models like autoregressive integrated moving average (ARIMA) to explicitly account for the autocorrelation structure. Maturation and threats further compromise ITS validity by introducing systematic changes unrelated to the . Maturation refers to natural developmental or temporal trends in the outcome, such as seasonal variations or aging, that could alter pre- and post- trajectories. involves changes in procedures, observer effects, or methods around the intervention point, potentially creating artificial discontinuities. These can be checked and mitigated by assessing the of pre-intervention trends and ensuring consistent protocols across the series. In multiple-group ITS designs, arises when and units are not comparable, leading to differences in trends or levels that confound estimates. Non-equivalent groups, such as regions with differing demographics or prior exposures, may yield biased attributions of change to the rather than inherent group differences. This can be alleviated through propensity score weighting, matching on covariates, or synthetic methods to create more balanced comparisons. Reporting biases, including p-hacking through selective trend modeling or post-hoc adjustment of timing, pose risks to the and of ITS findings. Such practices can exaggerate or fabricate effects by choosing models that yield significant results after inspection. Guidelines emphasize pre-specifying plans, including model forms and points, to minimize these biases and ensure robust reporting.

Applications

Interventions

Interrupted time series (ITS) analyses have become a cornerstone for evaluating interventions, offering a robust quasi-experimental to assess impacts on population-level outcomes like disease incidence, mortality, and health behaviors. This approach excels in contexts by leveraging from routine systems—such as , vital statistics, or national registries—without requiring individual , which is often impractical or unethical for broad-scale policies. By modeling pre- and post-intervention trends, ITS helps isolate intervention effects from underlying secular trends, , and in time-dependent data. A prominent example is the evaluation of the California Tobacco Control Program, initiated in 1990 after voters approved Proposition 99, which raised cigarette taxes by 25 cents per pack and allocated funds for anti-smoking media campaigns and community programs. Using monthly cigarette sales data from 1988 onward, analysis revealed an immediate level drop of approximately 27 packs in the year following the tax increase. This intervention was also associated with an additional 3.9% reduction in the rate of death from ischemic heart disease attributable to lower smoking prevalence. Similarly, the impact of the UK's 1983 compulsory was examined through modeling of monthly road casualty data, including fatalities and serious injuries requiring hospital admission. The analysis detected an immediate level reduction of about 12% in car occupant deaths and 8% in serious injuries post-legislation, with no significant change in overall trends for non-car users, underscoring the policy's targeted protective effect. , a standard ITS method, was used to quantify these abrupt shifts while accounting for pre-existing downward trends in . In the context of infectious disease control, ITS has been instrumental during the to gauge the effects of non-pharmaceutical interventions like . A nationwide study in , utilizing daily data on positivity from testing programs, applied interrupted time series to estimate that the March 2020 lockdown led to a 10.5% immediate reduction in infection incidence, with a gradual further decline of 1.4% per week in the ensuing period, highlighting the intervention's role in curbing transmission amid high uncertainty. ITS has been used to assess vaccine policy impacts, particularly in low-resource settings where randomized trials are challenging, as it enables the use of routine coverage and data to detect changes in morbidity and mortality trends post-introduction. For instance, an ITS evaluation of routine vaccination in demonstrated significant immediate and sustained reductions in rotavirus hospitalizations among children under 5 years old.

Policy and Economic Evaluations

Interrupted time series (ITS) analyses have proven valuable in evaluating the effects of labor and economic policies, particularly through the examination of outcomes following hikes. Seminal work by and Krueger (1994) assessed the 1992 New Jersey increase from $4.25 to $5.05 per hour using a comparative survey of fast-food restaurants, finding no adverse effects and even relative gains compared to neighboring . This approach has been extended in ITS frameworks with state-level quarterly data to model pre- and post-intervention trends in rates, often confirming null or positive impacts on low-wage sectors without significant disemployment. Such applications leverage to isolate policy-induced level shifts and slope changes, providing robust evidence against traditional disemployment predictions in competitive labor markets. In evaluation, ITS designs have been instrumental in quantifying the Clean Air Act of 1970's influence on urban air quality and industrial activity. (2002) utilized a controlled quasi-experimental setup, akin to multiple ITS comparisons across attainment and nonattainment counties, to analyze Census of Manufactures data from 1972–1987. The study estimated a 4–8% reduction in total suspended in regulated areas, alongside modest declines in manufacturing employment (about 0.6% per plant), highlighting the policy's effectiveness in pollution abatement at limited economic cost. These findings underscore ITS's utility in disentangling regulatory impacts from underlying economic trends in longitudinal environmental data. Education policy reforms represent another domain where ITS illuminates intervention outcomes, notably through assessments of accountability measures on academic performance. Dee (2011) applied a comparative ITS design to state panel data on standardized test scores from 1999–2007, targeting the No Child Left Behind Act's 2002 implementation. The analysis revealed a 4–7 percentile gain in fourth- and eighth-grade math achievement in high-stakes states, attributed to a post-intervention slope increase, though reading scores showed negligible or negative shifts, suggesting subject-specific reallocations of instructional resources. Economic applications of ITS extend to high-frequency financial data, such as daily stock indices, to probe interruptions from recessions or market shocks. For instance, event-study variants of ITS have modeled the abrupt volatility spikes and return disruptions during the 2007–2009 recession, estimating persistent negative trend alterations in major indices like the following the collapse. These analyses typically detect level changes exceeding 20% in immediate post-event periods, with gradual slope recoveries over quarters, informing understandings of crisis propagation in equity markets. ITS has seen growing adoption in for testing policing strategies, including those inspired by , which posits that addressing minor disorders prevents major crimes. Harcourt and Ludwig (2006) employed time-series methods on crime data from 1989–2002 to evaluate order-maintenance policing's role in the 1990s crime drop, finding weak evidence for broken windows effects amid confounding factors like demographic shifts. More recent applications, such as in (2010–2018), use ITS on monthly arrest and offense records to assess gentrification-linked enforcement changes, revealing temporary dips in low-level crimes but no sustained impact on violent incidents. This cross-disciplinary use highlights ITS's flexibility in policy contexts requiring from single-unit longitudinal observations. For a more recent example as of 2023, ITS analysis has been applied to evaluate the impact of vaccination campaigns on trends in multiple countries, showing significant reductions in all-cause mortality slopes post-rollout.

Limitations and Extensions

Primary Limitations

Interrupted time series (ITS) analysis demands extensive pre- and post-intervention , typically requiring at least 8-12 observations per period and ideally 50 or more total time points to ensure sufficient statistical power for detecting meaningful changes in level or trend. High-frequency observations, such as monthly or weekly measurements, are preferable to capture underlying patterns like or , while sparse or low-frequency —such as annual aggregates—often result in reduced power, wider confidence intervals, and inability to adjust for time-varying confounders. Routine administrative or are commonly used to meet these requirements, but their must be rigorously assessed for and . Ethical considerations limit ITS to observational designs, as randomizing or withholding interventions—such as policies or clinical guidelines—is often infeasible or prohibited, precluding the establishment of true experimental controls. This reliance on naturally occurring interruptions introduces inherent challenges in attributing solely to the , without the benefits of to balance unobserved confounders. Practical feasibility is hindered by the difficulties in obtaining reliable retrospective data for historical interventions, particularly when administrative records exhibit inconsistencies, missing values, or unaccounted heterogeneity across populations or sites, which can inflate costs and prolong data preparation efforts. While aggregated outcomes from existing sources reduce primary collection burdens, ensuring their validity for long-term series remains resource-intensive. Interpretation of ITS results is complicated by the challenge of isolating the intervention's effect from or spillover, where the policy or exposure inadvertently influences untreated groups or adjacent regions through behavioral, economic, or informational channels. This can confound trend estimates and lead to over- or underestimation of impacts if not explicitly modeled. Inconsistent reporting practices further undermine the reliability of ITS studies, with many failing to adhere to established guidelines such as those from the Cochrane Effective Practice and Organisation of Care (EPOC) group; for instance, only 55% of reviewed healthcare ITS studies accounted for , and just 24% addressed . Such omissions can effect estimates and hinder or synthesis in meta-analyses.

Extensions and Variants

Stepped-wedge designs represent a extension of (ITS) analysis, particularly suited for cluster-randomized trials where an is rolled out sequentially across multiple over time. In this approach, all clusters begin in the and progressively receive the in a randomized order, allowing for the of effects while controlling for time-varying confounders through mixed-effects models that incorporate fixed effects for time periods and random effects for . This design leverages the strengths of ITS by analyzing trends before and after each cluster's point, providing greater statistical power than traditional parallel cluster designs in scenarios where withholding the ethically or logistically challenging. The foundational for analyzing stepped-wedge designs was introduced by Hussey and Hughes, who demonstrated that the power depends on the number of steps, per step, and the magnitude of effects, often using generalized linear mixed models to account for within-cluster correlations. To handle multiple interruptions, such as policy reversals or phased implementations, ITS models can be extended through with additional parameters for each interruption point, allowing of level and changes at multiple time points. This involves including variables for each post-intervention segment and interaction terms to capture shifts in trends, which enables the of cumulative or offsetting effects from successive events. For instance, in evaluating repeated policies, the model can be specified as Y_t = \beta_0 + \beta_1 T + \sum_{i=1}^{k} \beta_{2i} X_i + \sum_{i=1}^{k} \beta_{3i} (T \times X_i) + \epsilon_t, where X_i are indicators for the i-th interruption and k is the number of interruptions, facilitating the isolation of each event's impact while adjusting for . Such extensions are particularly useful in complex policy environments, as they maintain the quasi-experimental rigor of ITS without assuming a single disruption. Integration of techniques enhances ITS by addressing non-linear trends and heterogeneous effects that traditional parametric models may overlook. Random forests, for example, can model complex, non-monotonic pre-intervention trends by aggregating decision trees to predict counterfactual outcomes, improving the accuracy of post-intervention effect estimates in noisy or high-dimensional data. Similarly, causal forests extend this capability to estimate heterogeneous effects by splitting the sample based on covariates that moderate the intervention's , such as demographic factors, which is valuable in personalized policy evaluations. Athey and colleagues formalized causal forests as an adaptation of random forests for treatment effect estimation, showing their asymptotic consistency and ability to handle in observational settings akin to ITS. Recent applications combine these with two-stage ITS frameworks, where machine learning first identifies structural breaks or non-linearities before applying autoregressive models for inference, yielding more robust results in environmental or health . Geospatial ITS extends standard models to incorporate spatial in area-level data, crucial for epidemiological studies where outcomes in neighboring regions influence each other. This involves augmenting with spatial lag or error terms, such as using a spatial Y_t = \rho W Y_t + X_t \beta + \epsilon_t, where W is a spatial weights matrix capturing adjacency, to adjust for effects like spread across borders. In practice, Bayesian hierarchical frameworks integrate this with ITS to evaluate impacts on spatially correlated outcomes, such as metrics, by including random effects for spatial units and time, thereby reducing bias from omitted spatial dependencies. This variant has gained traction in post-pandemic analyses of regional interventions, enabling more precise causal inferences in geographically heterogeneous settings. Future directions in ITS include the development of nonlinear adaptive regression models, as applied to evaluate long-term health outcomes such as birth defects.

References

  1. [1]
    Regression based quasi-experimental approach when ... - NIH
    Interrupted time series analysis is a quasi-experimental design that can evaluate an intervention effect, using longitudinal data.
  2. [2]
    Interrupted time series regression for the evaluation of public health ...
    Jun 8, 2016 · Interrupted time series (ITS) analysis is a valuable study design for evaluating the effectiveness of population-level health interventions.
  3. [3]
    Methods, Applications and Challenges in the Analysis of Interrupted ...
    May 13, 2020 · Interrupted time series (ITS) designs are robust quasi-experimental designs commonly used to evaluate the impact of interventions and programs ...
  4. [4]
    Campbell DT, Stanley JC (1963) - The James Lind Library
    Whole article. Download the full article as a PDF. Campbell DT, Stanley JC (1963). Experimental and quasi-experimental designs for research. Chicago: Rand ...
  5. [5]
    [PDF] EXPERIMENTAL AND QUASI-EXPERIMENT Al DESIGNS FOR ...
    In this chapter we shall examine the validity of 16 experimental designs against 12 com mon threats to valid inference. By experi.Missing: interrupted | Show results with:interrupted
  6. [6]
    A change in level of a non-stationary time series - Oxford Academic
    G. E. P. Box, George C. Tiao, A change in level of a non-stationary time series, Biometrika, Volume 52, Issue 1-2, June 1965, Pages 181–192, https://doi.org ...
  7. [7]
    Interrupted Time Series Design and Analyses in Health Policy ... - NIH
    Aug 2, 2024 · Interrupted time series (ITS) analysis assesses whether an intervention is associated with a shift in the trend of the outcome of interest. ITS ...
  8. [8]
    Design characteristics and statistical methods used in interrupted ...
    Jan 28, 2019 · An interrupted time series (ITS) design is an important observational design used to examine the effects of an intervention or exposure.Background · Search Methods For The... · Data Extraction And...
  9. [9]
    A methodological framework for model selection in interrupted time ...
    Jun 6, 2018 · Interrupted time series (ITS) is a powerful and increasingly popular design for evaluating public health and health service interventions.
  10. [10]
    MultiCenter Interrupted Time Series Analysis: Incorporating Within ...
    Jun 17, 2020 · Interrupted time series (ITS) design involves repeated measurements of an outcome at several time points before and after implementation of ...Abstract · Introduction · Weighted Its Analysis
  11. [11]
    [PDF] Quasi-Experiments: Interrupted Time-Series Designs
    The interrupted time series is a particularly strong quasi- experimental alternative to randomized designs when the latter are not feasible and when a time ...
  12. [12]
    [PDF] The British "Breathalyser" - ASU Center for Problem-Oriented Policing
    The Interrupted Time-Series is a quasi-experimental design (Campbell and Stanley, 1966; Campbell, 1969) for studying the effect of a given. "treatment" on a ...
  13. [13]
    Methodological Progress Note: Interrupted Time Series - Mahant
    May 19, 2021 · In this methodologic progress note, we discuss QEDs for evaluating the impact of healthcare interventions or events and focus on ITS, one of the strongest QEDs.What Is A Quasi-Experimental... · What Is An Interrupted Time... · How Are Segmented Regression...
  14. [14]
    Evolution of the Time Series Experiment by Richard McCleary :: SSRN
    Jul 6, 2017 · A memoir of my days in Evanston. Keywords: time series experiment, evolutionary epistemology. undefined. Suggested Citation: Suggested Citation.
  15. [15]
    Segmented regression analysis of interrupted time series studies in ...
    Interrupted time series design is the strongest, quasi-experimental approach for evaluating longitudinal effects of interventions.Missing: 2001 | Show results with:2001
  16. [16]
    Segmented regression analysis of interrupted time series studies in ...
    Aug 13, 2002 · Interrupted time series design is the strongest, quasi-experimental approach for evaluating longitudinal effects of interventions.
  17. [17]
    Inferring causal impact using Bayesian structural time-series models
    This paper proposes to infer causal impact on the basis of a diffusion-regression state-space model that predicts the counterfactual market response.
  18. [18]
    [PDF] Inferring causal impact using Bayesian structural time-series models
    An important problem in econometrics and marketing is to infer the causal impact that a designed market intervention has exerted on an out-.
  19. [19]
    Conducting Interrupted Time-series Analysis for Single
    In this article, I introduce the new itsa command, which performs interrupted time-series analysis using two ordinary least-squares (OLS) regression-based ...
  20. [20]
    Applying a propensity score-based weighting model to interrupted ...
    The propensity score-based weighting model weights control groups to represent the average outcome the treatment group would have had without the intervention.
  21. [21]
    [PDF] Applying a propensity score-based weighting model to interrupted ...
    The study design is generally referred to as an interrupted time series because the intervention is expected to 'interrupt' the level and/or trend subsequent to ...
  22. [22]
    Interrupted time series regression for the evaluation of public health ...
    Jun 8, 2016 · Interrupted time series (ITS) analysis is a valuable study design for evaluating the effectiveness of population-level health interventions.
  23. [23]
    Estimating effects of health policy interventions using interrupted ...
    Aug 31, 2022 · A classic methodology used in evaluating the impact of health policy interventions is interrupted time-series (ITS) analysis, ...
  24. [24]
    Evaluation and Study Designs for Implementation and Quality ...
    The goal of this basic design is to understand whether a new ... Interrupted time series evaluations are a type of quasi-experimental design ...
  25. [25]
    The use of controls in interrupted time series studies of public health ...
    Interrupted time series analysis differs from most other intervention study designs in that it involves a before-after comparison within a single population, ...
  26. [26]
    [PDF] Conducting interrupted time-series analysis for single- and multiple ...
    In an interrupted time-series analysis, an outcome variable is observed over multiple, equally spaced time periods before and after the introduction of an ...Missing: 1965 | Show results with:1965
  27. [27]
    [PDF] The Validity and Precision of the Comparative Interrupted Time ...
    The distinction between nonexperimental designs and quasi-experimental designs was popularized by Shadish, Cook, and Campell (2002) as a way of emphasizing that ...
  28. [28]
    [PDF] Synthetic Control Methods For Comparative Case Studies
    Building on an idea in. Abadie and Gardeazabal (2003), this paper proposes the use of synthetic control methods to overcome these shortcomings. Our method ...
  29. [29]
    Interrupted time series analysis using autoregressive integrated ...
    Mar 22, 2021 · Interrupted time series analysis is increasingly used to evaluate the impact of large-scale health interventions.Missing: seminal | Show results with:seminal
  30. [30]
    Non-Parametric Generalized Additive Models as a Tool for ... - MDPI
    Feb 3, 2021 · The interrupted time series analysis is a quasi-experimental design used to evaluate the effectiveness of an intervention.
  31. [31]
    Local Polynomial Regression - Sage Research Methods
    An implementation of local polynomial regression called lowess (or loess) is the most commonly available [Page 20]method of nonparametric regression. For these ...Missing: interrupted | Show results with:interrupted
  32. [32]
  33. [33]
    28.2 Interrupted Time Series | A Guide on Data Analysis - Bookdown
    Interrupted Time Series (ITS) is a powerful quasi-experimental method used to assess how an intervention affects the level and/or trend of an outcome over time.
  34. [34]
    Comparison of six statistical methods for interrupted time series ...
    Jun 26, 2021 · The Interrupted Time Series (ITS) is a quasi-experimental design commonly used in public health to evaluate the impact of interventions or ...Missing: hybrid | Show results with:hybrid<|separator|>
  35. [35]
    [PDF] Introduction to Interrupted Time Series Part I. Concepts
    Jan 19, 2016 · Interrupted Time Series (ITS) is a design with a change in level or slope at intervention onset, compared to quasi-experimental designs without ...Missing: core principles
  36. [36]
    [PDF] Testing the Validity of the Single Interrupted Time Series Design
    This paper focuses on short, single interrupted time series. The “short” qualifier means that my analysis will not rely on autoregressive integrated moving ...Missing: seminal | Show results with:seminal
  37. [37]
    The Effects of Seat Belt Legislation on British Road Casualties - jstor
    SUMMARY. Monthly data on road casualties in Great Britain are analysed in order to assess the effects on casualty rates of the seat belt law introduced on ...Missing: interrupted | Show results with:interrupted
  38. [38]
    Physical distancing interventions and incidence of coronavirus ...
    Jul 15, 2020 · In this natural experimental study across 149 countries we used interrupted time-series analyses to compare the change in incidence of covid-19 ...
  39. [39]
    [PDF] Minimum Wages and Employment: A Case Study of the Fast-Food ...
    On April 1, 1992, New Jersey's minimum wage rose from $4.25 to $5.05 per hour. To evaluate the impact of the law we surveyed 410 fast-food restaurants in.
  40. [40]
    [PDF] Time-Series Minimum-Wage Studies: A Meta-analysis David Card
    Jul 2, 2007 · 38-54. Card, David and Krueger, Alan B. "Minimum. Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and. Pennsylvania ...
  41. [41]
    The Impacts of Environmental Regulations on Industrial Activity
    Sep 1, 2001 · This paper estimates the effects of environmental regulations on industrial activity. The analysis is conducted with the most comprehensive data available.
  42. [42]
    Evidence from the 1970 and 1977 Clean Air Act Amendments and ...
    This paper estimates the impacts of the Clean Air Act's division of counties into pollutant‐specific nonattainment and attainment categories on measures of ...
  43. [43]
    [PDF] The impact of no Child Left Behind on student achievement
    This study identifies the impact of. NCLB using a comparative interrupted time series (CITS) design that relies on com- parisons of test-score changes across ...
  44. [44]
    [PDF] Disentangling the Channels of the 2007–09 Recession
    ABSTRACT This paper examines the macroeconomic dynamics of the. 2007–09 recession in the United States and the subsequent slow recovery.
  45. [45]
    [PDF] ARTICLE Broken Windows: New Evidence from New York City and ...
    Jan 31, 2006 · He said he wanted policing based on the so-called broken-windows theory.” Charlie LeDuff, Los Angeles. Police Chief Faces a Huge Challenge, NY ...
  46. [46]
  47. [47]
    Methodology and reporting characteristics of studies using ...
    Jul 4, 2019 · Interrupted time series (ITS) is considered one of the strongest quasi-experimental designs [4]. In an ITS design, data are collected at ...
  48. [48]
    Three Statistical Approaches for Assessment of Intervention Effects
    Feb 22, 2021 · Segmented regression of interrupted time series (ITS) analysis is another quasi-experimental approach for evaluating the impact of an ...
  49. [49]
    Use of interrupted time series methods in the evaluation of health ...
    Oct 14, 2020 · Interrupted time series (ITS) design has increasingly been advocated as a more robust design to evaluating health system quality improvement (QI) interventions.
  50. [50]
    Design and analysis of stepped wedge cluster randomized trials
    Design and analysis of stepped wedge ... 2007 Feb;28(2):182-91. doi: 10.1016/j.cct.2006.05.007. Epub 2006 Jul 7. Authors. Michael A Hussey , James P Hughes ...
  51. [51]
    Estimation and Inference of Heterogeneous Treatment Effects using ...
    Oct 14, 2015 · In this paper, we develop a non-parametric causal forest for estimating heterogeneous treatment effects that extends Breiman's widely used random forest ...
  52. [52]
  53. [53]
    Two-Stage Interrupted Time Series Analysis with Machine Learning
    Jul 11, 2025 · This paper explores the application of a two-stage ITS framework that compares traditional autoregressive integrated moving average (ARIMA) ...
  54. [54]
    A Bayesian Interrupted Time Series framework for evaluating policy ...
    ... Penfold and Zhang, 2013). ITS designs have been used to evaluate policies on ... After the selection process, we had 380 378 observations from 47 555 ...
  55. [55]
    Adaptive Regression for Nonlinear Interrupted Time Series Analyses ...
    The purpose of this article is to provide an overview of adaptive regression modeling and demonstrate its use in conducting nonlinear analyses of interrupted ...<|control11|><|separator|>