• Roger Chou and Paul Shekelle.
• Department of Medicine, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, MC BICC, Portland, OR 97239, USA. chour@ohsu.edu
• JAMA. 2010 Apr 7; 303 (13): 1295-302.

ContextLow back pain is extremely common. Early identification of patients more likely to develop persistent disabling symptoms could help guide decisions regarding follow-up and management.ObjectiveTo systematically review the usefulness of individual risk factors or risk prediction instruments for identifying patients more likely to develop persistent disabling low back pain.Data SourcesElectronic searches of MEDLINE (1966-January 2010) and EMBASE (1974-February 2010) and review of the bibliographies of retrieved articles.Study SelectionProspective studies of patients with fewer than 8 weeks of low back pain from which likelihood ratios (LRs) were calculated for prediction of persistent disabling low back pain for findings attainable during the clinical evaluation.Data ExtractionTwo authors independently assessed studies and extracted data to estimate LRs.Data SynthesisA total of 20 studies evaluating 10,842 patients were identified. Presence of nonorganic signs (median [range] LR, 3.0 [1.7-4.6]), high levels of maladaptive pain coping behaviors (median [range] LR, 2.5 [2.2-2.8]), high baseline functional impairment (median [range] LR, 2.1 [1.2-2.7]), presence of psychiatric comorbidities (median [range] LR, 2.2 [1.9-2.3]), and low general health status (median [range] LR, 1.8 [1.1-2.0]) were the most useful predictors of worse outcomes at 1 year. Low levels of fear avoidance (median [range] LR, 0.39 [0.38-0.40]) and low baseline functional impairment (median [range] LR, 0.40 [0.10-0.52]) were the most useful items for predicting recovery at 1 year. Results were similar for outcomes at 3 to 6 months. Variables related to the work environment, baseline pain, and presence of radiculopathy were less useful for predicting worse outcomes (median LRs approximately 1.5), and a history of prior low back pain episodes and demographic variables were not useful (median LRs approximately 1.0). Several risk prediction instruments were useful for predicting outcomes, but none were extensively validated, and some validation studies showed LRs similar to estimates for individual risk factors.ConclusionThe most helpful components for predicting persistent disabling low back pain were maladaptive pain coping behaviors, nonorganic signs, functional impairment, general health status, and presence of psychiatric comorbidities.

19 keywords...

hide…