The relation of age or HDL cholesterol to IHD and CVD was evaluated in quartile categories.
The subjects were divided into quartiles using levels of osteoprotegerin, osteocalcin, FGF23, and 25-hydroxyvitamin D3, and the presence or absence of cervical arterial calcification was evaluated in each quartile (Table 1).
For biomarkers with no interactions with diabetes, associations were further evaluated in separate models, first as quartiles, adjusted for age and race, and then as continuous variables (per SD or log-unit increase), adjusted for age, race, BMI and WC (Table 4).
The association between cataract prevalence for each quartile of intake or blood levels was evaluated in terms of OR.
In addition, when change in MOS SPI was evaluated across quartiles of change in body weight, rMSSD or VO2peak, no significant between-group differences were noted (data not shown).
Change in HbA1c was evaluated across quartiles of change in body composition and fitness.
Additionally, among completed participants, changes in MOS SPI score were evaluated across quartiles of change in body weight, rMSSD and VO2peak following adjustment for age, treatment, BMI, sleep medication use, HRT use and baseline MOS SPI score.
We evaluated fasting insulin concentration (in quartiles), impaired fasting glucose (100 125 mg/dL), and impaired glucose tolerance (2-h glucose concentration 140 199 mg/dL) as parallel secondary exposures (25).
We also evaluated calibration within quartiles of predicted mortality.
To obtain a single (summary) effect estimate for urinary cadmium with each outcome, we evaluated this cadmium quartile trend variable in the fully adjusted models.
