RESUMO
This study clusters empirically into subgroups the preoperative MMPI profiles of morbidly obese females seeking a gastric bypass, assesses the replicability of clusters across different subject populations and tests the hypotheses that cluster inclusion bears a significant relationship to weight loss and post-operative follow-up behavior. One hundred seventy women participated: 85 were patients at The University of Texas Health Science Center at San Antonio, Texas (UTHSCSA), and 85 were from the Cherryhill Medical Surgical Center (CMSC), Albion, Michigan. A hierarchical cluster analytic technique was applied using the cosine measure and average linkage within groups method. Purification of the two samples resulted in seven groups with 59 cases in the UTHSCSA sample and six groups with 62 cases in the CMSC sample. Five groups containing 94 subjects (55 percent) replicated across the two samples (UTHSCSA = 44; CMSC = 50). Discriminant analysis yielded four statistically significant functions. The chi 2 test was employed to discern significant differences based on group inclusion. Demographic differences between the two geographic samples are discussed. One year follow-up studies found that personality type bears a significant relationship to indexed weight loss, follow-up appointment frequency, and follow-up duration for the San Antonio sample.
