Predicting mortality with biomarkers: a population-based prospective cohort study for elderly Costa Ricans

Table 2 Crude death rate ratios of normalized biomarkers estimated with an independent hazard regression model for each biomarker (effect of 1 SD on the death RR)

	Main	Square	Age 80+	Male
Metabolic hormones
Glycosylated hemoglobin (HbA1c)	1.42**	0.99	0.89	0.86+
Fasting glucose	1.37**	1.00	0.84+	0.81*
CV biomarkers
Diastolic blood pressure	0.95	1.06**	0.92	1.06
Systolic blood pressure	0.88+	1.07**	0.89	1.11+
Metabolic – lipids
Triglycerides	0.90	1.05**	0.88	1.04
Total cholesterol	0.79**	1.07**	1.09	1.02
HDL cholesterol	0.84*	1.05*	1.15	1.00
Total/HDL cholesterol ratio	0.99	1.02**	0.89	1.01
LDL cholesterol	0.80*	1.04+	1.08	1.02
Stress hormones
Urinary cortisol	1.00	0.99	1.15	1.12+
DHEAS	0.90	1.08**	0.93	0.87
Epinephrine	1.16	1.00	0.88	0.85
Norepinephrine	0.98	1.01	1.06	1.09
Inflammation, immune system
CRP	1.42**	0.97*	0.93	0.99
Organ-specific functional reserve
Creatinine clearance	0.72**	1.10**	1.21+	0.97
Handgrip strength	0.41**	0.98	0.98	1.69**
Distance in 10 seconds	0.56**	1.01	1.16+	1.08
Pulmonary peak flow	0.47**	0.97	0.97	1.39*
Nutrition, body size
Knee height	0.92	0.97	0.98	1.14
Waist circumference	1.08	1.06**	0.85+	1.02
BMI	1.04	1.02**	0.77**	0.98
Waist/hip ratio	1.16+	0.99	0.85+	0.99

Significance assessed with robust estimates of standard errors: + means significant at *p < 0.10; at **p < 0.05; and at p < 0.01.
All regression models also include control variables for sex, age (continuous), and Nicoya.

ISSN: 1478-7954