Skip to main content

Table 1 Variables in the Interrupted Time Series equationa

From: Interrupted time series design to evaluate ICD-9-CM to ICD-10-CM coding changes on trends in Colorado emergency department visits related to traumatic brain injury

Variable Description
Yt The rate of TBI related ED visits per 100,000 population in a given month (t).
timet Represents month values from 1 (January 2012) to 72 (December 2017).
β0 The intercept or the outcome at the start of the ICD-9-CM period (baseline level at time 0, t = 0, of December 2011)
β1 The coefficient for the independent variable of time, that starts at 1 for the first month and 72 for the last month for the 72 total months in the six-year period from January 2012 to December 2017
β2 The coefficient indicating the level of change in the rate of TBI-related ED visits immediately after the transition
ICD10CM A dummy variable indicating the start of ICD-10-CM (the interruption in the time series) is 0 for each month prior to October 2015, and 1 for each month beginning October 2015 and after.
β3 The coefficient indicates the change in the slope of the monthly TBI rate after the transition to ICD-10-CM, compared to slope β1
time_after_ICD10CM The variable indicates in rank order or sequential order the months at and after the beginning of ICD-10-CM. This variable is 0 prior to ICD-10-CM, 1 for October 2015, 2 for November 2015, 3 for December 2015, etc.
vt The error term represents the random variability not explained by the variables in model
  1. aSlavova et al. (2018) provides further explanation of the variables