Skip to main content
. Author manuscript; available in PMC: 2017 Sep 1.
Published in final edited form as: J Am Geriatr Soc. 2016 Aug 22;64(9):1823–1829. doi: 10.1111/jgs.14283

Figure 1.

Figure 1

Each marker’s association to hippocampal volume was subjected to an adjusted linear model in the entire sample as shown. The solid black lines show the linear relationships, dashed lines show lowess diagnostic smoothers supporting linearity assumptions. The β value is interpreted as the difference in hippocampal volume associated with a standard deviation increase in the marker level. Plotted points and associations are adjusted for age, intracranial volume, education, sex and race. Inflammatory marker levels were Winsorized to avoid extreme outlier effects; results were robust to exclusion of these points.