Skip to main content
. 2018 Feb 9;13(2):e0192298. doi: 10.1371/journal.pone.0192298

Fig 1. Relative bias of the exposure-outcome relation when the exposure and confounder contain random measurement error.

Fig 1

The relative bias is expressed as a % of the adjusted exposure-outcome relation when there is no ME (reference standard; see Table 2). The amount of added ME is expressed as a percentage of the total variance of the variable. In (a) and (b) ME is added to the exposure, SBP, and to a confounder; DBP in (a) or ABI in (b). In (c) ME is added to the exposure, CIMT, and a confounder, SBP. Age and sex were additionally included as confounders for all multivariable analyses. Red colors indicate and an underestimation of the exposure-outcome relation due to ME, whereas blue colors indicate an overestimation. ME = measurement error; SBP = systolic blood pressure; DBP = diastolic blood pressure; ABI = ankle-brachial index; CIMT = carotid intima media thickness.