The convergence and predictive validity of the Multidimensional Pain Inventory and the Personality Assessment Inventory among individuals with chronic pain.

Abstract

Objective: To explore the convergence, redundancy, and validity of the Multidimensional Pain Inventory (MPI) and the Personality Assessment Inventory (PAI) in a chronic pain treatment setting. Participants: Data from intake (N 235) and follow-up (N 187) for individuals with an average of 9 years of chronic pain who participated in a 20-day integrative treatment program were analyzed. Outcome Measures: Oswestry Disability Index, Beck Depression and Anxiety inventories, Rand Short-Form Health Survey, and clinician-rated ability to stand and carry. Results: Conjoint factor analyses suggested that the MPI and PAI combine to tap five orthogonal factors: Negative Affect, Support, Externalizing, Physical Dysfunction, and Impulsivity. MPI and PAI scales significantly related to various aspects of client functioning, although these scales were more limited in predicting clinician-rated markers and change during treatment. Conclusion: Results support the combined use of the MPI and PAI to understand patient heterogeneity and predict treatment outcome in chronic pain samples.