Skip to main content
. 2012 Mar 22;97(6):1970–1976. doi: 10.1210/jc.2011-2967

Table 3.

Multivariable analyses of noncancer death and cancer death in relation to levels and longitudinal changes in IGF-I, IGFBP-1, and IGFBP-3 among Cardiovascular Health Study All Stars Study participants

Noncancer death (228 events)
Overall cancer death (39 events)
Adjusted HR 95% CI P Adjusted HR 95% CI P
IGF-I, change in level 0.91 0.79, 1.06 0.23 1.03 0.72, 1.47 0.87
IGF-I, level at baseline visit 0.95 0.83, 1.10 0.51 0.71 0.49, 1.03 0.07
IGFBP-1, change in level 1.14 1.02, 1.27 0.02 1.40 1.10, 1.77 0.01
IGFBP-1, level at baseline visit 1.16 1.04, 1.30 0.01 1.23 0.92, 1.67 0.17
IGFBP-3, change in level 0.89 0.77, 1.03 0.13 1.09 0.76, 1.55 0.65
IGFBP-3, level at baseline visit 0.91 0.79, 1.05 0.19 0.72 0.49, 1.05 0.09

Change in IGF-I, IGFBP-1, and IGFBP-3 levels defined as percentage change between 1996–1997 visit and 2005–2006 visit, rescaled as sd of change. Baseline level refers to measurement obtained at the 2005–2006 visit, rescaled as sd of level. Hazard ratios represent relative hazard of death per sd increase in change or baseline level of IGF-I, IGFBP-1, or IGFBP-3. Models were adjusted for age, sex, race, diabetes, body mass index, creatinine, albumin, and C-reactive protein. The sd values were: 48.4 μg/liter for change in IGF-I level, 36.2 μg/liter for change in IGFBP-1 level, and 618.6 μg/liter for change in IGFBP-3 level. Table 1 displays sd values for IGF-I, IGFBP-1, and IGFBP-3 levels at baseline. HR, Hazard ratio.