The daily pain rating scale (DPRS) consists of an 11-point numeric scale ranging from 0 (no pain) to 10 (worst possible pain). Participants described their pain during the past 24 hours by choosing the appropriate number between 0 and 10.

All participants in the Full Analysis Set (FAS) population, consisting of all participants randomized to treatment that received at least 1 dose of study medication.

The daily pain rating scale (DPRS) consists of an 11-point numeric scale ranging from 0 (no pain) to 10 (worst possible pain). Participants described their pain during the past 24 hours by choosing the appropriate number between 0 and 10. The mean endpoint pain score was obtained from the last 7 available DPRS scores of the daily pain diary while the participant was on study medication, up to and including the day after the last Week 8 (Day 57) dose.

All participants in the Full Analysis Set (FAS) population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The Last Observation Carried Forward (LOCF) method was used.

The DPRS consists of an 11-point numeric scale ranging from 0 (no pain) to 10 (worst possible pain). Participants described their pain during the past 24 hours by choosing the appropriate number between 0 and 10. The weekly mean pain score was the sum of the daily scores divided by the number of diary entries during that week. The overall change is the average change from Weeks 1 to 9.

The FAS population consisted of all participants randomized to treatment that received at least 1 dose of study medication. N=number of evaluable participants at the specified time point.

Pain-related sleep interference was assessed on an 11-point numerical rating scale ranging from 0 (did not interfere with sleep) to 10 (completely interfered [unable to sleep due to pain]). Participants were to describe how their pain had interfered with their sleep during the past 24 hours by choosing the appropriate number between 0 and 10.

All participants in the Full Analysis Set (FAS) population, consisting of all participants randomized to treatment that received at least 1 dose of study medication.

Pain-related sleep interference was assessed on an 11-point numerical rating scale ranging from 0 (did not interfere with sleep) to 10 (completely interfered [unable to sleep due to pain]). Participants were to describe how their pain had interfered with their sleep during the past 24 hours by choosing the appropriate number between 0 and 10. The mean endpoint score was obtained from the last 7 available scores of the daily diary while the participant was on study medication, up to and including the day after the last Week 9 (Day 63) dose.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

Pain-related sleep interference was assessed on an 11-point numerical rating scale ranging from 0 (did not interfere with sleep) to 10 (completely interfered [unable to sleep due to pain]). Participants were to describe how their pain had interfered with their sleep during the past 24 hours by choosing the appropriate number between 0 and 10. The weekly mean score was the sum of the daily scores divided by the number of diary entries during that week. The overall change is the average change from Weeks 1 to 9.

The FAS population consisted of all participants randomized to treatment that received at least 1 dose of study medication. N=number of evaluable participants at the specified time point.

The DPRS consists of an 11-point numeric scale ranging from 0 (no pain) to 10 (worst possible pain). Participants described their pain during the past 24 hours by choosing the appropriate number between 0 and 10. A 30% responder was a participant who had 30% reduction or more in mean pain score at the end of the fixed dose phase (Day 63/Week 9) (Study Endpoint) compared to baseline.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

SF-MPQ was assessed according to the participant's answer to the SF-MPQ questionnaire. The score for each composite scale (sensory, affective, and total) was derived by summing the reported intensity value for each item within a particular scale where None=0, Mild=1, Moderate=2, and Severe=3. The sensory score was the sum of the scores of the first 11 pain descriptors (throbbing, shooting, stabbing, sharp, cramping, gnawing, hot-burning, aching, heavy, tender, and splitting) and could range from 0-33. The affective score was the sum of the scores of the last 4 pain descriptors (tiring-exhausting, sickening, fearful, and punishing-cruel) and could range from 0-12. The total score was the sum of the scores of all 15 pain descriptors and could range from 0 to 45. Higher scores indicated greater pain.

The FAS population consisted of all participants randomized to treatment that received at least 1 dose of study medication. N=number of evaluable participants at the specified time point. No inferential analyses were performed.

The VAS was part of the Short Form McGill Pain Questionnaire (SF-MPQ) scale and reflected the overall pain intensity score, The pain VAS was a horizontal line; 100 millimeters (mm) in length, was self-administered by the participant in order to rate pain from 0 (no pain) to 100 (worst possible pain). The PPI was part of the SF-MPQ scale and measured the participant's present pain intensity on a 6-point scale ranging from 0 (no pain) to 5 (excruciating).

All participants in the FAS population, consisting of all participants randomized to treatment that received at least 1 dose of study medication.

The VAS was part of the SF-MPQ scale and reflected the overall pain intensity score. The pain VAS was a horizontal line; 100 mm in length, was self-administered by the participant in order to rate pain from 0 (no pain) to 100 (worst possible pain).

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The PPI was part of the SF-MPQ scale and measured the participant's present pain intensity on a 6-point scale ranging from 0 (no pain) to 5 (excruciating).

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated instrument which assesses sleep quantity and quality with 12 items (7 subscale scores: sleep disturbance, snoring, awakening short of breath/with headache, sleep adequacy, somnolence, sleep quantity, optimal sleep; and a 9-item overall sleep problems index). Subscale scores total range: 0-100 (except sleep quantity [range 0-24 hours], optimal sleep [yes:1, no:0]). Higher scores=poorer sleep outcomes (except sleep quantity, adequacy, and optimal sleep).

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. N=number of evaluable participants for each category

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. For sleep disturbance, the subscale score also ranged from 0 to 100, with higher scores representing greater sleep disturbance.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The snoring subscale score also ranged from 0 to 100, with lower scores indicating less snoring.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The awaken short of breath subscale also ranged from 0 to 100, with lower scores indicating less difficulty in breathing.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The MOS Sleep Quantity sub-scale scores ranged from 0 to 24 (number of hours slept).

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The MOS optimal sleep subscale was a binary outcome derived from the sleep quantity responses: the response was YES if sleep quantity was 7 or 8 hours per night.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The sleep adequacy subscale also ranged from 0 to 100, with higher scores indicating greater sleep adequacy.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The somnolence subscale score also ranged from 0 to 100, with lower scores indicating less somnolence.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The MOS-Sleep Scale was a participant-rated questionnaire consisting of 12 items that assessed key constructs of sleep. Instrument scoring yielded 7 subscales (sleep disturbance, snoring, awaken short of breath or with a headache, quantity of sleep, optimal sleep, sleep adequacy, and somnolence) as well as a 9-item overall sleep problems index. The total score ranged from 0 to 100. The sleep problems index subscale score also ranged from 0 to 100, with lower scores indicating fewer sleep problems.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The CGIC was a clinician-rated global measure that provided a clinically relevant and easy to interpret account of a clinician's perception of the clinical importance of the participant's improvement or worsening during their involvement in a clinical study. Clinicians rated the participant's overall improvement on a 7-point scale where scores ranged from 1 (very much improved) to 7 (very much worse).

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The PGIC was a participant-rated global measure that provided a clinically relevant and easy to interpret account of a participant's perception of the clinical importance of their own improvement or worsening during their involvement in a clinical study. Participants rated their overall improvement on a 7-point scale where scores ranged from 1 (very much improved) to 7 (very much worse).

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The HADS was a self-administered questionnaire that consisted of 2 subscales, 1 measuring anxiety (HADS-A Scale) and the other measuring depression (HADS-D Scale). Each subscale comprised of 7 items; participants assessed how each item applied to them on a scale of 0 (no presence of anxiety or depression) to 3 (severe feeling of anxiety or depression). Subscores from HADS-A (Anxiety) and HADS-D (Depression) were not to be combined. The interpretation of each HADS subscales was as follows: 0-7 normal, 8-10 mild, 11-14 moderate and 15-21 severe.

All participants in the FAS population, consisting of all participants randomized to treatment that received at least 1 dose of study medication.

The HADS was a self-administered questionnaire that consisted of 2 subscales, 1 measuring anxiety (HADS-A Scale) and the other measuring depression (HADS-D Scale). Each subscale was comprised of 7 items; participants assessed how each item applied to them on a scale of 0 (no presence of anxiety or depression) to 3 (severe feeling of anxiety or depression). Subscores from HADS-A (Anxiety) and HADS-D (Depression) were not to be combined. The interpretation of each HADS subscales was as follows: 0-7 normal, 8-10 mild, 11-14 moderate and 15-21 severe.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.

The HADS was a self-administered questionnaire that consisted of 2 subscales, 1 measuring anxiety (HADS-A Scale) and the other measuring depression (HADS-D Scale). Each subscale was comprised of 7 items; participants assessed how each item applied to them on a scale of 0 (no presence of anxiety or depression) to 3 (severe feeling of anxiety or depression). Subscores from HADS-A (Anxiety) and HADS-D (Depression) were not to be combined. The interpretation of each HADS subscales was as follows: 0-7 normal, 8-10 mild, 11-14 moderate and 15-21 severe.

All participants in the FAS population (all participants randomized to treatment that received at least 1 dose of study medication) who had available data for this outcome measure. The LOCF method was used.