The purpose of this study was to describe and compare the circadian rhythm of body temperature and cortisol, as well as self-reported clock times of sleep onset and offset on weekdays and weekends in 19 healthy adult "larks" (morning chronotypes) and "owls" (evening chronotypes), defined by the Home and Ostberg questionnaire. Day-active subjects entered the General Clinical Research Center, where blood was sampled every 2 h over 38 h for later analysis for cortisol concentration by enzyme immunoassay. Rectal body temperature was measured continuously. Lights were turned off at 22:30 for sleep and turned on at 06:00, when subjects were awakened. The acrophases (peak times) of the cortisol and temperature rhythms occurred 55 minutes (P < or = .05) and 68 minutes (P < .01), respectively, earlier in the morningness group. The amplitude of the cortisol rhythm was lower in the eveningness than in the morningness group (P = n.s.). Subject groups differed on all indices of habitual and preferred timing of sleep and work weekdays and weekends (P = .05-.001).

This study examined associations of psychosocial factors with cardiovascular reactivity in two groups of men and women--spouse caregivers of individuals with Alzheimer's disease (M age = 69.4 N = 82) and controls (M age = 68.5, N = 78) group-matched for age and gender. Cardiovascular responses to an emotional task (speech sample about one's spouse) yielded higher systolic and diastolic blood pressures (SBP, DBP) than a spoken cognitive task, which in turn yielded higher BPs and heart rate (HR) than the baseline rest period. HR was greater in response to the two tasks than in response to the baseline period, but it did not differ across tasks. Regression models of SBP, DBP, and HR reactivity in response to the two tasks demonstrated that after controlling for hypertension and gender, combinations of hostility, anger expression, avoidance coping, Type A behavior and Expressed Emotion (criticism) explained more reactivity in response to the emotional (8-12%) than the cognitive task (4-7%). Caregivers were more reactive than controls only if they were hypertensive. Psychosocial factors may be as important in explaining reactivity in older adults as in younger adults.

A method that measures 12 analytes in urine and reflects possible exposure to pesticides was developed. The sample preparation involves enzyme hydrolysis and solvent extraction through the use of laboratory robotics, followed by phase-transfer catalysis derivatization and silica cleanup. Samples are analyzed by capillary gas chromatography and tandem mass spectrometry using an isotope dilution technique with 13C-labeled internal standards. The limit of detection is 1 microgram/L (1 part per billion) for most analytes, and most analytes have a linear response up to 100 micrograms/L. The precision of the method is reflected in the variation observed in quality control materials over 33 months; the variation averaged 17% for these analytes. On the basis of the detectable analyte levels of unspiked urine samples collected from unexposed volunteers, this method can be used to measure the low levels necessary for establishing reference range values of the selected pesticides or metabolites.