Segmented Linear Regression Models for Assessing Change in Retrospective Studies in Healthcare

<div>Length of stay versus time <i>t</i> in days from the start of the study. The vertical dashed line marks the onset of the HFU. The black line shows the line-line model as the best-fitting linear model for entire time period: first line <i>y</i> = −0.0012<i>t</i> + 20.2556 and second line <i>y</i> = −0.0150 (<i>t</i> − 1698) + 18.2532 with delimiting time at 1698. The dashed blue lines are the best-fitting linear models for the pre-HFU and post-HFU periods. The best model for the pre-HFU period is the single line with equation <i>y</i> = –0.0021<i>t</i> + 20.7696, and best model for post-HFU is the single line with equation <i>y</i> = –0.01769 (<i>t</i> − 1550) + 48.8532.</div>

Computational and Mathematical Methods in Medicine

fig3

Figure 3

Figure 3: Segmented Linear Regression Models for Assessing Change in Retrospective Studies in Healthcare 