Morphine is increasingly used in patients with chronic non-cancer pain, but a major concern associated with chronic use relates to possible cognitive side-effects. The aim of this long-term prospective study was to evaluate the cognitive impact of oral sustained release morphine in patients with non-cancer pain. A battery of neuropsychological tests to explore attention, psychomotor speed and memory was administered. The effects of morphine on pain, quality of life, mood, subjective memory impairment and side-effects were also investigated. Evaluations were performed at baseline in patients free from opioids and then after 3, 6 and 12 months. Twenty-eight patients were included: 18 received oral sustained morphine (range 40-140 mg/day), ten patients stopped morphine prematurely because of side-effects or insufficient pain relief and were followed as a control group. There was no impairment of any neuropsychological variable over time in the morphine treated patients in comparison with the control group. Two measures of information processing speed - the Stroop interference score and the digit symbol test were improved at 6 and 12 months and there were significant correlations with the pain relief and improvement of mood. Self-reported memory impairment improved notably in responders to morphine. Morphine induced persisting effects on pain, and to a lesser extent on quality of life and mood. The visual analog scale score for side-effects increased at 12 months and essentially consisted of gastrointestinal disorders. This study demonstrates that 12 months treatment with oral morphine does not disrupt cognitive functioning in patients with chronic non-cancer pain and instead results in moderate improvement of some aspects of cognitive functioning, as a consequence of the pain relief and concomitant improvement of well-being and mood.