Multivariate Analysis of the MMPI Profiles of Patients with Multiple Pain Complaints

Abstract

The MMPI profiles of 123 male and 221 female patients with multiple pain complaints referred for psychological evaluation over a three-year period were examined using a multivariate clustering technique. Replicable subgroups were found in the patient cohorts in both the male and female samples. Multiple discriminant analysis and analysis of variance showed that the subgroups differed significantly from one another and within subgroup variation was small relative to between subgroup variation. It is suggested that pain patients are characterized by a variety of personality styles; subgroup membership may be an indication of behavior and personality dynamics which may have implications for treatment selection. Inferences as to possible treatment responses based on subgroup membership are made.

Psychogenic Pain and Psychological Variables: a Psychometric Study

“…These studies have described psychogenic pain patients as emotional, overconcerned, active, tending to deny psychological conflicts and prone to suppress the expression of their emotions and feelings [Oken, 1974], Other authors have hypothesized for psychogenic pain pa tients a perceptual and cognitive distortion psychometrically measurable through the as sessment of alexithymic traits [Sifneos, 1982], Psychometric studies using the MMPI have not been conclusive because they have informed about a psychological profile defined by hypochondriac, hysterical and depressive features [Sternbach et al, 1973;Corominas et al, 1978;Armentrout et al, 1982;Main and Wadell, 1982], common to chronic patients in general. McCreary [1979] also found high scores in the social introversion subscale and although other au thors reported several MMPI profiles in chronic pain patients [Prokop et al, 1980;McGill et al, 1983;Bradley and Van der Heide, 1984], there are some doubts about the validity of studies using MMPI.…”

MMPI profiles of chronic pain patients: Some methodological considerations that concern clusters and descriptors

“…Prokop et al (1980) had found three replicated subtypes among female patients with multiple pain complaints (excluding low back pain), but did not report the mean scores of the MMPI variables, so these could not be used.…”

“…Of the nine studies cited in this paper that directly involve the clustering of MMPI profiles of chronic pain patients, it is interesting to note that while the clustering program always is identified, the nature of the basic MMPI data is often difficult to determine. Eight of the nine studies (Armentrout, Moore, Parker, Hewett, & Feltz, 1982;Bernstein & Garbin, 1983;Bradley, Prokop, Margolis, & Gentry, 1978;Hart, 1984;McCreary, 1985;McGill, Lawlis, Selby, Mooney, & McCoy, 1983;Prokop, Bradley, Margolis, & Gentry, 1980;Snyder & Power, 1981), however, appear to use K-corrected T scores, while one study (Kudrow & Sutkus, 1979) used K-corrected raw scores. There are also occasional instances (e.g., McCreary, 1985;McGill et al, 1983) in which male and female cases are mixed or in which a pattern previously derived from a female sample is utilized with male patients (e.g., Hart, 1984).…”

Subtypes of Psychosocial Adjustment to breast cancer

“…Sternbach, 1974). Over the course of a number of investigations, using a variety of criteria for determining the optimal number of clusters, a certain degree of replicability emerged for a core set of patient subgroups (Armentrout et al, 1982;Bradley et al, 1978;Hart, 1984;McCreary, 1985;Prokop et al, 1980;Rosen et al, 1987). Novel or less common subgroups also occasionally emerged, which did not replicate consistently (cf.…”