This study examined factors that help to identify low back pain patients who do not benefit from a lumbar epidural steroid injection (LESI). Two-hundred and forty-nine chronic low back pain patients assessed their pain intensity before, 1 day after, and 2 weeks after receiving a LESI. All patients completed a comprehensive pain questionnaire and a Brief Symptom Inventory (BSI) prior to treatment. Diagnosis and extent of pathology were independently assessed by two physicians. One-hundred and thirty-one patients (52.6%) were followed 1 year after treatment. Results showed that average pain intensity ratings decreased in 62.3% of patients 2 weeks after receiving a LESI. One year after treatment, 62.6% felt that LESI was helpful. Nine patients (7%) felt that the treatment was harmful. Four factors were identified that best predicted poor outcome 2 weeks after LESI: (a) greater number of previous treatments for pain; (b) more medications taken; (c) pain not necessarily increased by activities, and (d) pain increased by coughing. Factors that predicted no benefit 1 year after treatment included (a) pain does not interfere with activities; (b) unemployment due to pain; (c) normal straight-leg raise test prior to treatment; and (d) pain not decreased by medication.