Chronic pain and the measurement of personality: do states influence traits?

Abstract

Study design: This is a structured evidence-based review of all available studies on the effect of pain, (a state phenomenon) on the measurement of personality characteristics (a trait phenomenon).

Objectives: To determine whether pain treatment changes trait scores.

Summary of background data: Recent evidence from the psychiatric literature indicates that the measurement of personality characteristics (traits) can be affected or changed by the presence of state psychiatric disorders, for example, depression. At issue then is whether the measurement of chronic pain patients' (CPPs') trait characteristics is affected by the presence of pain, a state problem.

Methods: Computer and manual literature searches for pain studies that reported a prepain treatment and postpain treatment (test-retest) personality test or inventory score produced 35 such reports. These references were reviewed in detail and information relating to the above problem was abstracted and placed into tabular form. Each report was also categorized as to the type of study it represented according to the guidelines developed by the Agency of Health Care Policy and Research (AHCPR). In addition, a list of 15 quality criteria was utilized to measure the quality of each study. Each study was independently categorized for each criterion as positive (criterion filled), negative (criterion not filled), or not applicable, by two of the authors. Only studies having a quality score of 65% or greater were utilized to formulate the conclusions of this review. The strength and consistency of the evidence represented by the remaining studies were then categorized according to the AHCPR guidelines. Conclusions of this review were based on these results.

Results: Of the 35 reports, 32 had quality scores of 65% or greater. According to the AHCPR guidelines, there was a consistent finding that the Minnesota Multiphasic Personality Inventory (MMPI) scores changed (improved) with treatment. In reference to the Millon Behavioral Health Inventory, Locus of Control, the Symptom Checklist-90-Revised (SCL-90-R), trait anxiety, and personality disorders, there were not enough studies to draw conclusions about consistency. In reference to coping/self-efficacy inventories, somatization/illness behavior inventories, and personality questionnaire studies, there was a generally consistent finding that these tests changed (improved) with pain treatment. Overall, of the 32 reports, 92.3% demonstrated a change in trait scores (improvement) with pain treatment. This evidence was categorized as consistent. Finally, 100% of a subgroup of reports (N = 12) that had controlled for pain indicated that there was a relationship between a change in pain scores and a change in trait scores.

Conclusions: Based on the above results, it was concluded that some trait tests and inventories may not be pain state independent. Therefore, caution is warranted in interpreting postpain development personality profiles as being indicative of the true prepain personality structure, if measured by these tests. Why trait scores may change with treatment, confounding test-retest issues, and whether trait tests actually measure what they allegedly measure are discussed.