Online Program Home
My Program

Abstract Details

Activity Number: 257 - SPEED: Longitudinal/Correlated Data
Type: Contributed
Date/Time: Monday, July 30, 2018 : 2:00 PM to 2:45 PM
Sponsor: Biometrics Section
Abstract #332942
Title: Interrupted Time Series Analysis to Evaluate the Effect of a Multicenter Collaborative Effort to Improve Care for Adult Intensive Care Patients
Author(s): Alai Tan* and Michele C. Balas
Companies: Ohio State University College of Nursing and Ohio State University College of Nursing
Keywords: Interrupted time series; joinpoint regression; segmented regression

Interrupted time series (ITS) design is a widely used quasi-experimental approach for implementation studies in health care. ITS analysis aims to identify changes in disease outcomes or care utilization by analyzing time series health care data before and after implementation. Segmented regression modeling is an often used method for ITS analysis, which requires specifying the time point of interest and hypothesized impact model a priori. Alternatively, joinpoint regression method can also identify time point(s) at which significant trend change occurred without specification of time points of interest and impact model. In the present study, we applied the segmented regression modeling and joinpoint regression analysis to evaluate the effect of a multicenter collaborative effort to improve care for adult intensive care patients. Both methods found significant increase in ABCDEF bundle adoption after the implementation of the collaborative effort, although the two methods identified somewhat different post-implementation trend. We propose reporting and interpreting results with scrutiny by taking account of the strength and limitations of each method.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program