Oral tablet, administered daily before bedtime

Perceptual distortions associated with transitions between wakefulness and sleep were termed as hypnagogic (occurring during the onset of sleep) or hypnopompic (occurring during onset of wakefulness) hallucinations.

The pre-specified terms which were suggestive of abuse potential on this study included depersonalization (feeling of watching oneself act, while having no control over a situation), derealization (alteration in the perception or experience of the external world so that it seems unreal), dissociation (includes a wide array of experiences from mild detachment from immediate surroundings to more severe detachment from physical and emotional experience), euphoric mood (exaggerated feeling of physical and emotional well-being and optimism not consonant with apparent stimuli or events), mania (state of abnormally elevated or irritable mood, arousal, and/or energy levels), hallucination (perception in the absence of a stimulus which has qualities of real perception), and potential study medication misuse.

Rebound insomnia was defined as insomnia that occurred following discontinuation of a sedative substance taken to relieve primary insomnia, and was assessed based on subjective total sleep time (sTST) as recorded in the participant's morning e-diary. A strict categorical analysis method (Yes/No) was used in which a participant was considered to have potentially experienced rebound (Yes) if the Morning Diary participant-reported sTST value (in minutes) on any of the 3 nights of the Run-out Phase (first 3 nights of the Discontinuation Phase) occurring after one year of treatment (Month 13) was less than the last value at baseline one year earlier (Month 1).

Rebound insomnia was defined as insomnia that occurred following discontinuation of a sedative substance taken to relieve primary insomnia, and was assessed based on subjective time to sleep onset (sTSO) as recorded in the participant's morning e-diary. A strict categorical analysis method (Yes/No) was used in which a participant was considered to have potentially experienced rebound (Yes) if the Morning Diary participant-reported sTSO value (in minutes) on any of the first 3 nights of the Run-out Phase occurring after one year of treatment (Month 13) was greater than the last value at baseline one year earlier (Month 1).

The sTSTm was defined as the average over time of daily e-diary values for a participant's report of the total amount of time spent asleep before waking for the day (measured in minutes). Weekly sTSTm values (Week 1, Week 2, etc.) were the average of the daily e-diary values for the week. A summary value of this measure for Month 1 was obtained by taking the average of weekly sTSTm values for Weeks 1 through 4; (Week 1 + Week 2 + Week 3 + Week 4) ÷ 4. LS Mean Change from Baseline in sTSTm was then calculated at Week 1, Week 2, Week 3, Week 4, and Month 1.

The sTSOm was defined as the average over time of daily e-diary values for a participant's report of the time he or she required to fall asleep (measured in minutes). Weekly sTSOm values (Week 1, Week 2, etc.) were the average of the daily e-diary values for the week. A summary value of this measure for Month 1 was obtained by taking the average of weekly sTSTm values for Weeks 1 through 4; (Week 1 + Week 2 + Week 3 + Week 4) ÷ 4. LS Mean Change from Baseline in sTSOm was then calculated at Week 1, Week 2, Week 3, Week 4, and Month 1.

Total of all reporting groups

The baseline sTSTm was defined as the participant's reported total amount of time spent asleep before waking for the day (measured in minutes), calculated as the mean of the last 7 (non-missing) daily e-diary values obtained during the placebo run-in phase.

(n=517, n=254)

The baseline sTSOm was defined as the participant's reported time that he or she required to fall asleep (measured in minutes), calculated as the mean of the last 7 (non-missing) daily e-diary values obtained during the placebo run-in phase.

(n=517, n=254)

Sleep paralysis was defined as the inability to perform voluntary muscle movements during sleep. Sleep paralysis adverse events included sleep-onset paralysis (paralysis as one is falling asleep).

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment

Complex sleep-related behaviors were reported as ECIs and were characterized by patients engaging in specific activities while asleep (e.g., eating, drinking, preparing meals, making phone calls, having sex, driving, and sleep walking).

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment.

Falls were adjudicated (to establish whether a fall event was due to cataplexy).

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment.

Suicidal ideation included suicidal plans, suicidal tendency, death wishes, life weariness, and suicidal intention. Suicidal behaviors included suicide attempts, suicide gesture, and self-injurious behaviour. Suicidal ideation and/or behavior was reported as an AE and considered an ECI.

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment.

Perceptual distortions associated with transitions between wakefulness and sleep were termed as hypnagogic (occurring during the onset of sleep) or hypnopompic (occurring during onset of wakefulness) hallucinations.

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment.

The pre-specified terms which were suggestive of abuse potential on this study included depersonalization (feeling of watching oneself act, while having no control over a situation), derealization (alteration in the perception or experience of the external world so that it seems unreal), dissociation (includes a wide array of experiences from mild detachment from immediate surroundings to more severe detachment from physical and emotional experience), euphoric mood (exaggerated feeling of physical and emotional well-being and optimism not consonant with apparent stimuli or events), mania (state of abnormally elevated or irritable mood, arousal, and/or energy levels), hallucination (perception in the absence of a stimulus which has qualities of real perception), and potential study medication misuse.

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment.

Withdrawal effects assessed using Tyrer WSQ, which evaluated the presence/absence and severity of withdrawal symptoms with 20 items (i.e. sensitivity to noise, light, smell, touch, feeling unreal, etc). The Tyrer WSQ was completed as part of the evening e-diary prior to dosing on the Month 12 visit and on the 3 consecutive evenings of the DB Run-out Phase (first 3 nights of DB Discontinuation Phase). Responses rated 0 (No), 1 (Yes-moderate), or 2 (Yes-severe); range from 0 (no withdrawal) to 40 (severe withdrawal).

A participant was defined to have a withdrawal symptom if an item during any of the 3 DB Run-out days had emerged for the first time, or had worsened compared to the measurement obtained at the end of the Treatment phase (Month 12). For single night analysis, a patient was defined to have withdrawal effects if the number of withdrawal symptoms (emergent or worsening) was ≥3. For across night analysis, withdrawal was defined as a total of ≥3 symptoms across the 3 nights.

Participants in the APaT population who completed the entire DB Treatment Phase, had at least one measurement at the end of the DB Treatment Phase (Month 12), had taken at least one dose of Run-out study medication, and had a measurement on at least one of the nights of the DB Run-out Phase.

Rebound insomnia was defined as insomnia that occurred following discontinuation of a sedative substance taken to relieve primary insomnia, and was assessed based on subjective total sleep time (sTST) as recorded in the participant's morning e-diary. A strict categorical analysis method (Yes/No) was used in which a participant was considered to have potentially experienced rebound (Yes) if the Morning Diary participant-reported sTST value (in minutes) on any of the 3 nights of the Run-out Phase (first 3 nights of the Discontinuation Phase) occurring after one year of treatment (Month 13) was less than the last value at baseline one year earlier (Month 1).

Participants in the APaT population who completed the entire DB Treatment Phase, had a baseline measurement, had taken at least one dose of DB Run-out study medication, and had a measurement on at least one of the nights of the DB Run-out Phase.

Rebound insomnia was defined as insomnia that occurred following discontinuation of a sedative substance taken to relieve primary insomnia, and was assessed based on subjective time to sleep onset (sTSO) as recorded in the participant's morning e-diary. A strict categorical analysis method (Yes/No) was used in which a participant was considered to have potentially experienced rebound (Yes) if the Morning Diary participant-reported sTSO value (in minutes) on any of the first 3 nights of the Run-out Phase occurring after one year of treatment (Month 13) was greater than the last value at baseline one year earlier (Month 1).

Participants in the APaT population who completed the entire DB Treatment Phase, had a baseline measurement, had taken at least one dose of DB Run-out study medication, and had a measurement on at least one of the nights of the DB Run-out Phase.

The sTSTm was defined as the average over time of daily e-diary values for a participant's report of the total amount of time spent asleep before waking for the day (measured in minutes). Weekly sTSTm values (Week 1, Week 2, etc.) were the average of the daily e-diary values for the week. A summary value of this measure for Month 1 was obtained by taking the average of weekly sTSTm values for Weeks 1 through 4; (Week 1 + Week 2 + Week 3 + Week 4) ÷ 4. LS Mean Change from Baseline in sTSTm was then calculated at Week 1, Week 2, Week 3, Week 4, and Month 1.

Full Analysis Set (FAS)-Efficacy; all randomized participants who had ≥1 post-randomization observation for the analysis endpoint subsequent to ≥1 dose of study treatment, and baseline data for those analyses that required baseline data. The number included in the FAS may vary across endpoints due to the degree of missing data for each endpoint.

The sTSOm was defined as the average over time of daily e-diary values for a participant's report of the time he or she required to fall asleep (measured in minutes). Weekly sTSOm values (Week 1, Week 2, etc.) were the average of the daily e-diary values for the week. A summary value of this measure for Month 1 was obtained by taking the average of weekly sTSTm values for Weeks 1 through 4; (Week 1 + Week 2 + Week 3 + Week 4) ÷ 4. LS Mean Change from Baseline in sTSOm was then calculated at Week 1, Week 2, Week 3, Week 4, and Month 1.

Full Analysis Set (FAS)-Efficacy; all randomized participants who had ≥1 post-randomization observation for the analysis endpoint subsequent to ≥1 dose of study treatment, and baseline data for those analyses that required baseline data. The number included in the FAS may vary across endpoints due to the degree of missing data for each endpoint.

Suicidal ideation and/or behavior that occurred on study was also assessed using the C-SSRS, a rater-administered questionnaire used to prospectively assess suicidal ideation and suicidal behavior. C-SSRS assessment was based upon a clinician's interpretation of the participant's responses to the C-SSRS questions, not by a numbered scale.

Suicidal ideation and/or behaviors identified on the C-SSRS may not have been considered an adverse event, based on the investigator's judgment.

All Participants as Treated (APaT) population; all randomized participants who received at least one dose of study treatment.

After a 1-week single-blind placebo run-in, participants received dose-matched placebo to suvorexant (administered according to age) daily before bedtime for 12 months during the DB Treatment Phase.

Following treatment with suvorexant during the 12-Month DB Treatment Phase, participants received their same dose of suvorexant during a 2-month DB Randomized Discontinuation Phase.

Following treatment with suvorexant during the 12-Month DB Treatment Phase, participants received dose-matched placebo to suvorexant during a 2-month DB Randomized Discontinuation Phase.

Following treatment with dose-matched placebo to suvorexant during the 12-Month DB Treatment Phase, participants continued to receive dose-matched placebo to suvorexant during a 2-month DB Randomized Discontinuation Phase.

Following treatment with suvorexant during both the 12-Month DB Treatment Period and the DB Run-out period, participants entered a Follow-up Phase which concluded with a follow-up phone call 14 days after the last dose of study medication (or 14 days after the Discontinuation visit, whichever time point was later) to report AEs.

Following treatment with suvorexant during the 12-Month DB Treatment Period and treatment with dose-matched placebo during the DB the Run-out period, participants entered a Follow-up Phase which concluded with a follow-up phone call 14 days after the last dose of study medication (or 14 days after the Discontinuation visit, whichever time point was later) to report AEs.

Following treatment with dose-matched placebo during both the 12-Month DB Treatment Period and the DB Run-out period, participants entered a Follow-up Phase which concluded with a follow-up phone call 14 days after the last dose of study medication (or 14 days after the Discontinuation visit, whichever time point was later) to report AEs.