OBJECTIVES: Little is known about the characteristics of duloxetine patients initiating therapy for chronic lower back pain (CLBP).  This study identified demographic and clinical characteristics associated with initiation of duloxetine versus other pain medications among CLBP patients. METHODS: Adult CLBP patients who initiated medication therapy (duloxetine, anticonvulsants, antidepressants, muscle relaxants, non-steroidal anti-inflammatory drugs [NSAIDs], opioids, or benzodiazepines) between November 1, 2010-March 31, 2011 were selected from medical and pharmacy claims from Medco Health Solutions. Treatment initiation was defined as no pill coverage of the medication over the previous 90 days. The dispense date of the first initiated medication was set as the index date. Comorbidities and prior medication use was assessed during the 6 months before the index date. Multiple logistic regression models were performed to identify predictors of initiating duloxetine versus other pain medication. Separate models were estimated with Charlson comorbidity index (CCI) + use of selected medications, and individual comorbidities + number of unique medications, respectively. RESULTS: We identified 38,943 CLBP patients (mean age 58.0 years; 62.8 % female) who initiated opioids (36.0%), NSAIDs (21.1%), muscle relaxants (13.6%), benzodiazepines (8.9%), antidepressants (8.7%), anticonvulsants (7.5%), or duloxetine (4.1%). Patients who initiated duloxetine had significantly higher CCI and more inpatient stays than patients who initiated other pain medications except for anticonvulsants. The regression results showed older patients were less likely to initiate duloxetine than anticonvulsants or opioids. Patients with higher CCI were more likely to initiate on duloxetine except for anticonvulsants and antidepressants. Prior use of pregabalin, gabapentin, antidepressants, and benzodiazepines were associated with a higher likelihood of initiating duloxetine. Patients who had more unique medications, peripheral vascular disease, or depression were more likely to initiate duloxetine. CONCLUSIONS: Presence of selected demographic characteristics, comorbidities and prior use of medications were significant predictors of duloxetine initiation among CLBP patients.