Interpretive rules for both standard scal es and factor scal es of the MMPI-168 were devel oped to predict long-term outcome (6 months to 3 years postdischarge) of an inpatient pain management program. Rul es for both standard-scale and factor-scale interpretation correctly identified 79% of pain program "successes" and "failures." Resul ts provide support for the MMPI-168 as a brief, useful measure of personality functioning in a physical medicine setting.Several studies have assessed the full Minnesota Multiphasic Personality Inventory (MMPI) as a predictor of pain program outcome. Maruta, Swanson, and Swenson (1979) found no statistically significant differences on MMPI scales between a group of failures and a group of successes from a pain management program, but did find that high scores on Hypochondriasis and Hysteria scales contributed to an index of premorbid factors predictive of poor pain program outcome. Painter, Seres, and Newman (1980) were unable to discriminate successful from unsuccessful patients of a pain center using the full MMPI. Kuperman, Golden, and Blume (1979) reported that, among a group of patients whose pain problems had a substantiated organic origin, a high Hypochondriasis scale score was associated with poor treatment outcomes in a pain clinic.A shorter form of the MMPI, the MMPI-168 (Overall & Gomez-Mont, 1974), appears to have utility in assessing pain patients. Calsyn, Spengler, and Freeman (1977) demonstrated that factor scoring of the MMPI-168 provided sufficient information to discriminate patients with organic pain from patients with a pain problem consisting of a mix of psychological, social, and organic factors with 74.5% accuracy. In a cross-validation sample, these same authors separated patients with primarily behavioral pain problems from patients with primarily organic pain disorders with 83% accuracy. The Somatization factor was most useful in making these discriminations. The authors