Paper Session 318: Psychosocial Factors in PainD. Treatment Approaches (Medical/Interventional)D15 - Opioids in Non-Cancer Pain(318) Regular use of prescribed opioids: Association with com-

mon psychiatric disorders in a population-based sampleM. Sullivan, M. Edlund, D. Steffick, J. Unutzer; University of Washington, Seat-tle, WAIn response to campaigns to improve treatment of patients with chronicnon-malignant pain, prescription opioid use has doubled between 1980and 2000. Treatment guidelines have suggested that opioid use in pa-tients with current mood, anxiety and substance use disorders may notbe appropriate. We analyzed the association of regular use of prescribedopioids with these psychiatric disorders in a population-based sampleusing cross-sectional data from the 1998 and 2002 waves of the HealthCare for Communities Study (N�14,113). After excluding patients withcancer, 435 (3%) of these respondents reported that they took an opioidmedication “at least several times a week for at least one month ormore.” These respondents were more likely meet DSM-IV criteria for:Major Depression (OR�4.4), Dysthymia (OR�4.2), Generalized AnxietyDisorder (OR�3.4), and Panic Disorder (OR�4.9). They were also morelikely to report problem use of prescription(OR�4.1)or illicit drugs(OR�4.4), but not problem drinking (OR�0.9). Those receiving opioidswere more likely to perceive a need for mental health treatment(OR�2.3), but this became non-significant (OR�1.0) after adjusting forthese mental disorders. Other significant univariate predictors of regu-lar opioid use were: age, education, income, work disability, self-ratedhealth, physical-component score from the SF-12, and chronic pain con-ditions including back pain and headaches. After adjusting for these,those on opioids were still more likely to meet criteria for at least one ofthese common mood or anxiety disorders (OR�1.8). The effect of thesedisorders was more significant in patients reporting low levels of paininterference with daily activities. These data suggest that some opioidprescribing may not be meeting current standards for appropriateness.Many patients receiving chronic opioids have unmet needs for mentalhealth and substance abuse care.

F. Treatment Approaches (Psychosocial and Cognitive)F01 - Cognitive/Behavioral Approaches(318) Pain-related acceptance and physical impairment in in-

dividuals with chronic low back painK. Vowles, D. McNeil, R. Gross, M. McDaniel, A. Mouse, M. Bates, P. Gallimore,C. McCall; West Virginia University, Morgantown, WVTreatments using acceptance-based approaches have been utilized for avariety of health care concerns with data regarding their application toa chronic pain population recently emerging. Acceptance of chronicpain is related to psychosocial constructs generally (e.g., depression,pain-related anxiety); however, its relation to the overt behavioral vari-ables associated with such states remains unevaluated. The present in-vestigation assessed the specific effects of an acceptance-based versuscontrol-based instructional set on demonstrated physical impairment in74 individuals with chronic low back pain. Participants completed abaseline evaluation of physical impairment and then listened to audiotaped instructions detailing one of three approaches (pain acceptance,pain control, and no change). They then repeated the physical impair-ment assessment and were asked to utilize the approach detailed in theaudio taped instructions to guide behavior during the tasks. Resultsindicated that, after controlling for baseline levels of physical impair-ment, individuals receiving the acceptance instructions were less physi-cally impaired at the second assessment compared to the other twogroups, which did not differ from one another. Further, individuals inthe acceptance group exhibited a 16.3% improvement in impairmentlevel, while the pain control group worsened by 8.3% and the no changegroup worsened by 2.5%; these changes were also statistically differentfrom one another. Group membership was generally unrelated to painreported during the physical impairment assessment, a finding theoret-ically consistent with acceptance models. These results lend further sup-port to the value of acceptance in actual patient behavior, as well as theability of a relatively simple acceptance-based intervention to improvefunctioning over the short term in a sample of individuals with chronicpain.

(318) Predicting the outcome of pain management amongchronic back pain patients by using motivational stagesof chronic pain management

M. Petraschka, S. Zenker, M. Schenk, C. Spies; Department of Anesthesiologyand Intensive Care Medicine, Charite-University Medicine, Berlin, GermanyAccording to Prochaska’s transtheoretical model, the aim of the studywas to evaluate the motivational stages of chronic back pain patientsand their association with pain intensity, chronicity and healthcare sys-tem expenses. After ethical committee approval and informed consent,Kern’s pain stages of change questionnaire (PSOCQ, German versionFQ-STAPM)1 was completed by 163 chronic back pain patients. Painintensity was measured by the numeric rating scale (NRS), pain chronic-ity was measured by the Mainz Pain Staging System. As psychometrictests, the lower pain disability index (PDI), the Hospital Anxiety andDepression Scale (HADS) and a quality of life score (SF36) were used.Health care system expenses were considered as number of consultedphysicians, number of stays in hospital and number of rehabilitationprograms. Statistical analysis: All data were analyzed by chi-square-test.Significant results demonstrated by chi-square-test were analyzed fur-ther with Whitney-Mann U test. Patients were in the following stages:precontemplation in 30%, preparation in 19%, action in 30%, and main-tenance in 21%. A lower level of pain chronicity was related to a signif-icantly higher level of motivation. The intensity of pain in the precon-templation stage patients was significantly higher compared to patientsin the maintenance stage. Moreover, there was a significant increase inhealthcare system expenses by lesser motivated patients. Patients in themaintenance stage used significantly less opioids than patients in theprecontemplation stage. The higher motivated patients had a signifi-cantly lower PDI, a significantly lower HADS, and a significantly higherquality of life compared to less motivated patients. The results indicatethat the FQ-STAPM might be a useful tool to classify chronic back painpatients and to work out a non-overtaxed strategy together with thepatient relevant to the outcome of pain management among chronicback pain patients. (1. Maurischat C, Eur J Pain, 2002)

(318) Separate and combined effects of CBT and drug ther-apy: Psychosocial outcomes in the treatment of chronictension-type headache

K. Holroyd, M. Stensland; Ohio University, Athens, OHThis study reports data from the Treatment of Tension-type Headacheclinical trial evaluating tricyclic antidepressant medication (AM), cogni-tive-behavioral stress-management therapy (SMT) and their combina-tion (AM � SMT) versus pill placebo (PL) in the treatment of chronictension-type headache (CTTH). 203 patients (mean age � 37; 76% fe-male; mean headache days/mo. � 26) with an International HeadacheSociety diagnosis of CTTH were randomized to the 4 conditions. Follow-ing a 2-month dose adjustment/ SMT administration period participantswere followed for 12 months. Participants completed the HeadacheSelf-Efficacy Scale (HSE) and Headache-Specific Locus of Control (HSLC)at 8 assessment points and the Headache Disability Inventory (HDI), BeckDepression Inventory (BDI), Trait Anxiety Scale (TAS), and Medical Out-comes Study SF20 (MOSSF20) at 4 assessment points.. On the HSE, andthree HSLOC subscales significant SMT X Time interactions (p � .001)revealed that participants in the two SMT groups showed significantchange on all 4 measures by the end of the first 4 weeks of treatmentwith changes maintained through the 12 month follow-up. In contrast,the AM & PL groups showed no change on these measures. Similarly, onthe HDI a significant SMT X Time interaction revealed significant reduc-tions in headache-related disability in the SMT groups, but not the AM &PL groups. On the BDI, TAS and MOSSF20 only Time effects were signif-icant indicating no treatment differences. Participants with a mood dis-order diagnosis had higher BDI scores and showed greater reductions inBDI scores than participants with no diagnosis. Although the three ac-tive treatments were equally effective in reducing tension-type head-ache activity relative to placebo, the three treatments differed in theirimpact on psychosocial measures. SMT, but not drug therapy alteredparticipants’ beliefs about the variables controlling their headaches andtheir ability to manage headache episodes.

S4 Abstracts