Psychosocial and health status variables independently predict health care seeking in fibromyalgia.

OBJECTIVE: To determine whether variables derived from the self-regulatory model of health and illness behavior accurately predict status as a patient or nonpatient with fibromyalgia (FM). METHODS: Subjects were 79 patients who met American College of Rheumatology (ACR) criteria for FM and 39 community residents who met ACR criteria for FM but had not sought medical care for their symptoms (nonpatients). Subjects were administered 14 measures that produced 6 domains of variables: background demographics and pain duration; psychiatric morbidity; and personality, environmental, cognitive, and health status factors. These domains were entered in 4 different hierarchical logistic regression analyses to predict status as patient or nonpatient. RESULTS: The full regression model was statistically significant (P < 0.0001) and correctly identified 90.7% of the subjects with a sensitivity of 92.4% and a specificity of 87.2%. The best individual predictors of group status were self-reports of self-efficacy, negative affect, recent stressful events, and perceived pain. Relative to nonpatients, patients reported higher levels of negative affect and perceived pain and a greater number of recent stressful experiences, as well as lower levels of self-efficacy. CONCLUSION: Consistent with the self-regulatory model of health and illness behavior, psychosocial and health status variables predict health care-seeking behavior in persons with FM independently of background demographics and psychiatric morbidity. These variables may influence the severity of symptoms experienced by persons with this disorder as well as their health care-seeking behavior, but they are not necessary to produce abnormal pain sensitivity in FM.

Psychometric assessment of motives for using cocaine in men with substance use disorders.

A 15-item self-report questionnaire from a previous study of social, coping, and enhancement motives for drinking was reworded to assess motives for using cocaine. Data from 121 cocaine users indicated satisfactory reliabilities for the 3 new 5-item scales, and the fit of a 3-factor structure in confirmatory factor analysis was comparable to that of the earlier study of drinking motives. Removal of 4 items substantially improved the fit to the 3-factor model without reducing reliabilities. Some evidence of validity was inferred from a correlation between reported use of cocaine for coping and a measure of general maladjustment.