Although the prevalence of sleep problems in patients with chronic pain (CP) is reported to be as high as 70%, the relationships between pain and sleep have not been thoroughly quantified. Data from clinical trials of duloxetine in patients with diabetic peripheral neuropathic pain (DPNP) are used to investigate associations between pain and sleep. Data were pooled from 3 double-blind, randomized, placebo-controlled, 12-week trials of patients with DPNP in which major mood disorders were excluded. Studies 1-3 (N=342, N=334, and N=348) compared duloxetine 60mg QD and 60mg BID with placebo. Efficacy measures included average daily pain severity and average night pain severity, both collected via patient diaries, and the Brief Pain Inventory (BPI) interference items. The weekly mean scores of the daily pain severity and night pain severity were calculated for Weeks 1-12. BPI was administered at Weeks 4, 8, and 12. At baseline, both daily and night pain severity were positively and significantly correlated with sleep interference (r=0.37 and 0.57, respectively; p<.001). Both doses of duloxetine were significantly superior to placebo in reducing both daily and night pain at Week 1 and at all assessment times thereafter. Placebo response was greater on sleep interference than for either pain measure. However, one of the duloxetine doses was superior to placebo in reducing sleep interference at Weeks 4 and 8, and both doses were significant at Week 12. Correlations between changes to endpoint in daily pain and night pain with sleep interference changes were 0.48 and 0.53, respectively (p < .001). Results suggest moderate to strong associations at baseline between daily pain, night pain and sleep interference, as well as between changes in pain (daily and night) and changes in sleep interference. Although causality was not established, these findings suggest improvements in pain will be associated with less interference in sleep.