Psychological dimensions of distress in chronic pain patients: A factor analytic study of Symptom Checklist-90 responses.

Abstract

The Symptom Checklist-90 (SCL-90) is a potentially useful measure of psychopathology in chronic pain patients. To date, the factor structure of the SCL-90 has not been examined for pain populations. A sample of 600 outpatients treated for chronic pain was randomly split and their SCL-90 responses analyzed via maximum likelihood (with oblique rotation) factoring. A second-order splithalf factor analysis was performed and several restricted factor models were compared within the Linear Structural Relations format (LBREL;Joreskog & Sorbom, 1978, 1979. A 10-fector model was judged most meaningful and statistically appropriate in the first-order analysis. The secondorder analyses produced three factors tentatively named Somatic Distress, Cognitive Distress, and Distrust. All factors replicated across the data halves, thus providing evidence for factorial stability. The complementary use of exploratory and confirmatory factoring methods is illustrated and discussed.It is widely acknowledged that the clinical management of chronic pain requires a multidisciplinary regimen that includes some optimal combination of social, psychological, and medical interventions (Crue, 1985;Schneider & Karoly, 1983;Turk, Meichenbaum, & Genest, 1983). This perspective has given impetus to a burgeoning literature concerned with the development of psychological assessment protocols suitable for medical patients (Leavitt, 1982;Nigl, 1984;Southwick& White, 1983).Consequently, there is a need to examine the responses of medical patients to standardized psychiatric measures in order to adapt such instruments for clinical use within the hospital setting.Several investigators (e.g., Hendler, 1981;D. P. Schwartz & DeGood, 1983) have advocated using the Symptom Checklist-90 (SCL-90;Derogatis, 1977) as a measure of psychological distress in chronic pain patients. The SCL-90 is a multidimensional self-report inventory composed of 90 items, each describing a physical or psychiatric symptom. The instructions require the patient to indicate on a 5-point scale (ranging from not at all [0] to extremely [4]) how much a given symptom has caused discomfort during the past 2 months. The instrument consists of nine subscales: Somatization (SOM), Obsessive-

Work Retention and Nonspecific Low Back Pain

“…Surprisingly, in this study, only somatization 1 (i.e., the bodily expression of psychologic distress) 22 was associated with work retention (HR 2.49, 95% confidence interval [CI] 1.25-4.93; P ϭ 0.01). Individuals who presented low levels of somatization 1 were more likely to retain work compared to individuals with high levels of somatization 1.…”

“…The scoring procedure used in this study was modified from the original version. The modified version of SCL-90R used was the one proposed by Shutty et al, 22 which was tested and validated in a large chronic pain population, with primarily musculoskeletal ailments. It used 10 different constructs, such as somatization 1 and 2, obsessive compulsive, interpersonal sensitivity, depression-anxiety, depression-cognition, hostility, phobic anxiety.…”

Dimensional characteristics of the SCL-90-R: Evaluation of gender differences in dually diagnosed inpatients

“…Furthermore, for males, partial replication was evidenced for the original hostility and phobic anxiety scales (Derogatis & Cleary, 1977a). Visual inspection of the items on the hostility and phobic anxiety components also revealed considerable overlap with Holcomb et al's (1983) hostility and phobic reaction factors and Shutty et al's (1986) first-order phobic factor. When considered in conjunction with previous research, these data support the use of the hostility and phobic anxiety scales and their associated norms with dually diagnosed males.…”

“…Independent confirmatory factor analyses were conducted to assess the tenability of Derogatis and Cleary's (1977a) original 9-factor structure, Schwarzwald et al's (1991) 5-factor model, Holcomb et al's (1983) 9-factor model, and Shutty et al's (1986) 10-first-order factor model within a dually diagnosed sample. All analyses were based on a simple structure common-factor model and were assessed independently for males and females.…”

Psychiatric symptom dimensions among asian americans and native hawaiians: An analysis of the symptom checklist

“…Researchers have evaluated the SCL-90 symptom dimension structure on different subpopulations including groups differentiated by gender (Derogatis & Cleary, 1977b) groups differentiated by social class (Derogatis, Lipman, Covi, & Rickels, 1971), chronic pain patients (Hendler, 1981;Schwartz & DeGood, 1983;Shutty, DeGood, & Schwartz, 1986), hospitalized patients (Holcomb, Adams, & Ponder, 1983), and college students (Walkey & McCormick, 1985). Few studies, however, have explored the performance of the SCL-90 among different ethnic groups, and even fewer studies have looked at differences among the specific ethnic groups in this research.…”