A provisional model to predict blood pressure response to biofeedback-assisted relaxation

Blood pressure (BP) response to biofeedback-assisted relaxation is not uniform among hypertensive individuals. The purpose of this exploratory study was to determine if selected psychophysiological variables could be used to identify individuals able to lower blood pressure using biofeedback-assisted relaxation. Responders were defined using a preset criterion of 5 mm Hg or greater decrease in mean arterial pressure. A logistic regression model derived from five variables (heart rate, finger temperature, forehead muscle tension, plasma renin response to furosemide, and mean arterial pressure response to furosemide) provided significant predictive power for BP response, exhibiting a sensitivity of 84.6% and a specificity of 80.0%. With future validation, the proposed model may provide useful information to identify patients likely to benefit from biofeedback-assisted relaxation.