OBJECTIVES: To compare the impact of 12 weeks’ BTDS 20 and BTDS 5 treatment on sleep among patients with moderate-to-severe chronic low back pain (CLBP), to examine changes in the burden of CLBP on sleep with 12 weeks of BTDS use, and to describe sleep outcomes over 12 months of continued BTDS treatment.   METHODS: This post-hoc analysis used data from a double-blind (DB) randomized trial evaluating BTDS 20 against BTDS 5 for treatment of opioid-experienced patients with moderate-to-severe CLBP and its 12 month open-label treatment extension. During the course of BTDS treatment, patients completed the 12-item Medical Outcomes Study Sleep Scale, which measures several sleep domains, including Disturbance and overall Quality. ANCOVA models compared scores between treatment arms during the 12-week DB phase, and repeated measures mixed-models analysis compared scores across visits during the 12-month treatment. Burden was examined by comparing trial patients with an age- and gender-matched U.S. general population sample (GPS) at trial baseline and DB endpoint.   RESULTS: BTDS 20 patients showed significantly less Disturbance and better overall Quality than BTDS 5 patients at week 12 DB (P<0.05), an advantage that emerged by week 4. Analyses of Disturbance and Quality scores across DB indicated statistically significant effects for treatment (P<0.05), but not for visit or their interaction (P>0.05). No reduction from the end of DB Disturbance and Quality scores were observed following 12 months of BTDS 20 treatment. At baseline, patients’ Disturbance and Quality scores were significantly worse than those of the GPS; by week 12 of the DB phase, BTDS 20 patients’ average scores improved over the GPS, while BTDS 5 patients’ did not.   CONCLUSIONS: Moderate-to-severe CLBP patients receiving BTDS 20 exhibited larger improvements in sleep Disturbance and Quality than those receiving BTDS 5. Improvements in sleep were sustained during 12 months of continued BTDS treatment.