Quieting response training: predictors of long-term outcome.

Extensive personality, demographic, and symptom-related information from a heterogeneous group of more than 300 patients was examined for patterns that would discriminate between successful and unsuccessful outcomes in patients undergoing Quieting Response (QR) training, an 8-week program that integrated EMG and thermal biofeedback with a variety of relaxation exercises. Follow-up evaluations were at 3, 6, 12 and 24 months. Agreement between expected (from stepwise discriminant analyses) and observed outcomes was 65% (whole sample), 62% (headache only), and 70% (Raynaud's only). Unimproved patient MMPI scales D, Sc, F, Hy, and Pt were significantly elevated. A variety of additional measures used indicated elevated levels of psychological stress among unsuccessful patients. While stereotypical descriptions of successful and unsuccessful patients are proposed, patterns from outcome groups were more similar than dissimilar, and no combination of factors was identified that had sufficient predictive power for use with future patients. No statistical grounds for exclusion were found among patterns of responses to the measures employed. Factors contributing to the inaccuracy of outcome prediction are discussed.