Multilevel non-linear interrupted time series analysis

TL;DR

This paper introduces a hierarchical Bayesian multilevel non-linear interrupted time series model that combines generalized additive models with latent time series modeling to analyze complex intervention effects across subpopulations.

Contribution

It proposes a novel combination of non-linear interrupted time series analysis with multilevel Bayesian modeling and hierarchical model selection to improve causal inference across subgroups.

Findings

Applied to prostate cancer diagnosis rates, COVID-19 hospitalization rates, and Medicaid expansion effects.

Demonstrated ability to model variability in intervention effects across subpopulations.

Showed improved parsimony and partial pooling in complex time series data.

Abstract

Recent advances in interrupted time series analysis permit characterization of a typical non-linear interruption effect through use of generalized additive models. Concurrently, advances in latent time series modeling allow efficient Bayesian multilevel time series models. We propose to combine these concepts with a hierarchical model selection prior to characterize interruption effects with a multilevel structure, encouraging parsimony and partial pooling while incorporating meaningful variability in causal effects across subpopulations of interest, while allowing poststratification. These models are demonstrated with three applications: 1) the effect of the introduction of the prostate specific antigen test on prostate cancer diagnosis rates by race and age group, 2) the change in stroke or trans-ischemic attack hospitalization rates across Medicare beneficiaries by rurality in the months after the start of the COVID-19 pandemic, and 3) the effect of Medicaid expansion in Missouri on the proportion of inpatient hospitalizations discharged with Medicaid as a primary payer by key age groupings and sex.