Mood, location and physical position as predictors of ambulatory blood pressure and heart rate: Application of a multi-level random effects model

Despite the popularity of ambulatory blood pressure monitoring (ABPM) in behavioral research and the interest in the role of negative affect in cardiovascular disease, few studies have examined the prevalence and cardiovascular effects of mood states on blood pressure (BP) and heart rate during everyday life. Those that have done this have used suboptimal analytic techniques. The present analysis is based on 24-hour ambulatory blood pressure (ABP) readings from 246 normotensive subjects who are part of an ongoing worksite study. Subjects were asked to report their position, location, activity, mood, and social interaction at the time of each BP reading. The data are analyzed by a multilevel random effects model based on a repeated measures ANOVA. The initial analysis evaluates the effects of position (standing, sitting, reclining) and location (home, work, other). The BP changes associated with position are relatively large (8-12 mm Hg) and stable across individuals, while those for location are modest (2-4 mm Hg) and show considerable inter-subject variation. Subjects were asked to record one or more of twelve mood states; 65% of readings were reported as 'neutral,' 12% as 'happy,' 7% as 'anxious,' 6% as 'tired,' and less than 1% as 'angry.' After controlling for position, location, and significant moods, significant increases of both systolic and diastolic BP were observed for 'angry' (4/4 mm Hg systolic/diastolic BP), 'excited' (3/2 mm Hg), and 'happy' (1/1 mm Hg). Significant decreases of both BPs were observed for 'tired' (- 2/-1 mm Hg). Heart rate changes were generally inconsistent, except for an increase associated with feeling 'rushed' (1 bpm). We conclude that while position and location account for a substantial part of the overall BP and heart rate variance, the effects of position are greater than would be expected on purely hemodynamic grounds and may be attributed to unreported differences in the level of arousal. Anger was the least regularly reported mood but is associated with the largest increase of BP. In total, mood does not account for much of the variance in BP or heart rate.