A treatment-related adverse event was any untoward medical occurrence attributed to gabapentin in a participant who received gabapentin. Participants with treatment-related adverse events were counted by each candidate risk factor (including gender, age, and disease eligible for the survey) to assess whether these were risk factors for the treatment-related adverse events. No inferential analyses of risk factors were performed because of a small number of the events (5 events).

Participants who responded to the treatment with gabapentin were counted by the baseline severity of epileptic seizure (mild, moderate and severe) to assess whether the baseline severity of epileptic seizure was a factor affecting the treatment efficacy.

Participants who responded to the treatment with gabapentin were counted by the baseline frequency of epileptic seizure (<=8 versus >8 episodes/per 4 weeks) to assess whether the baseline frequency of epileptic seizure was a factor affecting the treatment efficacy.

Participants who responded to the treatment with gabapentin were counted by the number of concomitant epileptic drugs at baseline across 5 categories (no drug, 1 drug, 2 drugs, 3 drugs, and 4 or more drugs) to assess whether the number of concomitant epileptic drugs at baseline was a factor affecting the treatment efficacy.

Participants who responded to the treatment with gabapentin were counted by the treatment period (non-long term [less than 1 year] or long term [1 year or more]) to assess whether the treatment period with gabapentin was a factor affecting the treatment efficacy.

Aggressive behaviors including affect lability and hostility were determined as key survey items by the sponsor (Pfizer Japan Inc.). These events were defined as the 101 preferred terms listed by pharmaceuticals and medical devices agency and classified according to MedDRA/J version 17.1. A treatment-related adverse event was any untoward medical occurrence attributed to gabapentin in a participant who received gabapentin. Relatedness to gabapentin was assessed by the investigator and sponsor.

R Ratio was calculated by the following formula, where B represented the baseline frequency of epileptic seizures during the previous 4 weeks from the treatment start date, whereas T represented the frequency of epileptic seizures during the previous 4 weeks from the end of assessment period: R Ratio = (T - B) / (T + B). R Ratio is within the range of -1 to +1, and a negative value represents a reduction in the frequency of seizure.

Responder rate, which was defined as the percentage of participants whose R ratio was - 0.333 or less, was presented along with the corresponding exact 2-sided 95% CI. R ratio of - 0.333 or less corresponded to the decrease of epileptic seizure frequency by 50% or more.

Reduction from baseline in epileptic seizure frequency was defined by the following formula, where B represented the baseline frequency of epileptic seizures during the previous 4 weeks from the treatment start date, whereas T represented the frequency of epileptic seizures during the previous 4 weeks from the end of assessment period: Reduction from baseline in epileptic seizure frequency (%) = [(T-B) / B] X 100. The median percentages were presented along with the corresponding minimum and maximum percentages.

Participants taking gabapentin tablets/syrup according to the Japanese package insert.

Clinical efficacy rate, which was defined as the percentage of participants who achieved clinical efficacy over the total number of efficacy analysis population, was presented along with the corresponding exact 2-sided 95% CI. For the basis of efficacy evaluation, frequencies of epileptic seizure were recorded during the previous 4 weeks from the treatment start date, and that from the end date of assessment period. Clinical efficacy was assessed according to the following categories: (1) effective, (2) not effective, or (3) not assessable.

The analysis population comprised of the participants in the efficacy analysis population from which those with data not assessable were excluded. n=number of participants with assessable data at each post-baseline time point.

A treatment-related adverse event was any untoward medical occurrence attributed to gabapentin in a participant who received gabapentin. Expectedness of the adverse event was determined according to the Japanese package insert. Relatedness to gabapentin was assessed by the investigator and sponsor (Pfizer Japan Inc.).

The safety analysis population comprised of participants who had met the inclusion criteria and had taken gabapentin at least once.

A treatment-related adverse event was any untoward medical occurrence attributed to gabapentin in a participant who received gabapentin. Participants with treatment-related adverse events were counted by each candidate risk factor (including gender, age, and disease eligible for the survey) to assess whether these were risk factors for the treatment-related adverse events. No inferential analyses of risk factors were performed because of a small number of the events (5 events).

The safety analysis population comprised of participants who had met the inclusion criteria and had taken gabapentin at least once. No data displayed because outcome measure has zero total participants analyzed.

Participants who responded to the treatment with gabapentin were counted by the baseline severity of epileptic seizure (mild, moderate and severe) to assess whether the baseline severity of epileptic seizure was a factor affecting the treatment efficacy.

The efficacy analysis population comprised of the participants who had at least one post-baseline efficacy evaluation for the clinical efficacy and seizure frequency in the safety analysis population. Participants with diseases not eligible for the survey were excluded from the efficacy analysis population.

Participants who responded to the treatment with gabapentin were counted by the baseline frequency of epileptic seizure (<=8 versus >8 episodes/per 4 weeks) to assess whether the baseline frequency of epileptic seizure was a factor affecting the treatment efficacy.

The efficacy analysis population comprised of the participants who had at least one post-baseline efficacy evaluation for the clinical efficacy and seizure frequency in the safety analysis population. Participants with diseases not eligible for the survey were excluded from the efficacy analysis population.

Participants who responded to the treatment with gabapentin were counted by the number of concomitant epileptic drugs at baseline across 5 categories (no drug, 1 drug, 2 drugs, 3 drugs, and 4 or more drugs) to assess whether the number of concomitant epileptic drugs at baseline was a factor affecting the treatment efficacy.

The efficacy analysis population comprised of the participants who had at least one post-baseline efficacy evaluation for the clinical efficacy and seizure frequency in the safety analysis population. Participants with diseases not eligible for the survey were excluded from the efficacy analysis population.

Participants who responded to the treatment with gabapentin were counted by the treatment period (non-long term [less than 1 year] or long term [1 year or more]) to assess whether the treatment period with gabapentin was a factor affecting the treatment efficacy.

The efficacy analysis population comprised of the participants who had at least one post-baseline efficacy evaluation for the clinical efficacy and seizure frequency in the safety analysis population. Participants with diseases not eligible for the survey were excluded from the efficacy analysis population.

Central nervous system depressant actions including somnolence and ataxia were determined as key survey items by the sponsor (Pfizer Japan Inc.). These events were defined according to MedDRA/J version 17.1 as the events classified in "psychiatric disorders" or "nervous system disorders" of the system organ classes, or those classified in "asthenia" or "gait disturbance" of the preferred terms. A treatment-related adverse event was any untoward medical occurrence attributed to gabapentin in a participant who received gabapentin. Relatedness to gabapentin was assessed by the investigator and sponsor.

The safety analysis population comprised of participants who had met the inclusion criteria and had taken gabapentin at least once.

Aggressive behaviors including affect lability and hostility were determined as key survey items by the sponsor (Pfizer Japan Inc.). These events were defined as the 101 preferred terms listed by pharmaceuticals and medical devices agency and classified according to MedDRA/J version 17.1. A treatment-related adverse event was any untoward medical occurrence attributed to gabapentin in a participant who received gabapentin. Relatedness to gabapentin was assessed by the investigator and sponsor.

The safety analysis population comprised of participants who had met the inclusion criteria and had taken gabapentin at least once.

R Ratio was calculated by the following formula, where B represented the baseline frequency of epileptic seizures during the previous 4 weeks from the treatment start date, whereas T represented the frequency of epileptic seizures during the previous 4 weeks from the end of assessment period: R Ratio = (T - B) / (T + B). R Ratio is within the range of -1 to +1, and a negative value represents a reduction in the frequency of seizure.

The analysis population comprised of the participants in the efficacy analysis population who had assessable data of the frequency of epileptic seizures at the start of gabapentin treatment and at the end of assessment period. n=number of participants with assessable data at each post-baseline time point.

Responder rate, which was defined as the percentage of participants whose R ratio was - 0.333 or less, was presented along with the corresponding exact 2-sided 95% CI. R ratio of - 0.333 or less corresponded to the decrease of epileptic seizure frequency by 50% or more.

The analysis population comprised of the participants in the efficacy analysis population who had assessable data of the frequency of epileptic seizures at the start of gabapentin treatment and at the end of assessment period. n=number of participants with assessable data at each post-baseline time point.

Reduction from baseline in epileptic seizure frequency was defined by the following formula, where B represented the baseline frequency of epileptic seizures during the previous 4 weeks from the treatment start date, whereas T represented the frequency of epileptic seizures during the previous 4 weeks from the end of assessment period: Reduction from baseline in epileptic seizure frequency (%) = [(T-B) / B] X 100. The median percentages were presented along with the corresponding minimum and maximum percentages.

The analysis population comprised of the participants in the efficacy analysis population who had assessable data of the frequency of epileptic seizures at the start of gabapentin treatment and at the end of assessment period. n=number of participants with assessable data at each post-baseline time point.