| Unique ID issued by UMIN | UMIN000061189 |
|---|---|
| Receipt number | R000070024 |
| Scientific Title | Effects of Personalized Feedback Generated by Generative Artificial Intelligence Based on Objective Sleep Metrics on Sleep, Mood States, and Sports Injury Severity in High School Female Soccer Players: Pilot Randomized Controlled Trial |
| Date of disclosure of the study information | 2026/04/07 |
| Last modified on | 2026/04/07 21:55:10 |
Effects of AI-Based Personalized Sleep Feedback in High School Female Soccer Players
AI Sleep Feedback Trial
Effects of Personalized Feedback Generated by Generative Artificial Intelligence Based on Objective Sleep Metrics on Sleep, Mood States, and Sports Injury Severity in High School Female Soccer Players: Pilot Randomized Controlled Trial
AI Sleep Intervention Pilot RCT
| Japan |
Sleep problems in high school female soccer players
| Child |
Others
NO
The objective of this study is to investigate the effects of personalized feedback generated by generative artificial intelligence based on objective sleep metrics on sleep, mood states, and sports injury severity in high school female soccer players.
Efficacy
Exploratory
Pragmatic
Not applicable
Sleep efficiency measured by Fitbit Inspire 3. Change from the mean value during the baseline observation period (weeks 1-4) to the mean value during the intervention period (weeks 6-9).
Total sleep time, wake after sleep onset, deep sleep duration, light sleep duration, and rapid eye movement sleep duration, all measured by Fitbit Inspire 3 and compared between the mean values during weeks 1-4 and weeks 6-9; total score of the Pittsburgh Sleep Quality Index Japanese version (PSQI-J); Sleep Difficulty Score; total score of the Japanese version of the Epworth Sleepiness Scale (ESS-J); Total Mood Disturbance score of the Japanese version of the Profile of Mood States 2nd Edition; and sports injury severity score assessed using a modified Japanese version of the Oslo Sports Trauma Research Center questionnaire on health problems. Questionnaire-based outcomes were assessed at week 5 and week 10.
Interventional
Parallel
Randomized
Individual
Open -but assessor(s) are blinded
Active
YES
NO
Institution is considered as a block.
NO
No need to know
2
Educational,Counseling,Training
| Behavior,custom |
Intervention group: During the 4-week intervention period (weeks 6 to 9), personalized sleep feedback generated by ChatGPT-4o was delivered once daily via the LINE Messaging API based on the previous night's Fitbit Inspire 3 sleep metrics (eg, total sleep time, bedtime, wake time, sleep efficiency, and sleep stages) and each participant's mean values during the baseline observation period (weeks 1 to 4). Delivery was automatically triggered after morning Fitbit data synchronization and was generally provided once each morning before the participant started daily activities.
Control group: During the 4-week intervention period (weeks 6 to 9), general and nonpersonalized sleep-related information was delivered once daily via LINE. The contents were limited to basic sleep hygiene, the role of sleep, and general knowledge for athletes (eg, environmental adjustment and relaxation strategies), and did not include advice based on individual sleep data. Messages were delivered every morning at 8:00 AM.
| 15 | years-old | <= |
| 18 | years-old | >= |
Female
Female soccer players belonging to a high school sports club who were expected to continue regular training and matches during the study period. Participants were required to be able to wear Fitbit Inspire 3 and receive messages via LINE, and to provide assent with written parental consent after receiving an explanation of the study.
Participants were excluded if they were expected to be absent for a prolonged period during the study, were unable to wear Fitbit Inspire 3 continuously, were unable to receive messages continuously via LINE, or were considered unsuitable for participation by the principal investigator.
25
| 1st name | Hayato |
| Middle name | |
| Last name | Kedoin |
Nippon Sport Science University Medical Vocational School
Department of Judo Therapy and Health Science
2270034
New Sophia 307, 1-11-19 Katsuradai, Aoba-ku, Yokohama, Kanagawa, Japan
08064188981
kedoin@nittai-iryo.ac.jp
| 1st name | Hayato |
| Middle name | |
| Last name | Kedoin |
Nippon Sport Science University Medical Vocational School
Department of Judo Therapy and Health Science
2270034
2-2-7 Yoga, Setagaya-ku, Tokyo 158-0097, Japan
08064188981
kedoin@nittai-iryo.ac.jp
Nippon Sport Science University Medical Vocational School
Nippon Sport Science University
Other
Research Ethics Committee of Nippon Sport Science University
7-1-1 Fukasawa, Setagaya-ku, Tokyo, Japan
08064188981
kedohayato1004@gmail.com
NO
| 2026 | Year | 04 | Month | 07 | Day |
Unpublished
25
All 25 participants were included in the final analysis. Compared with the control group, the intervention group showed significant improvements in changes in sleep efficiency, rapid eye movement sleep duration, total PSQI-J score, Sleep Difficulty Score, and Total Mood Disturbance. In contrast, no significant between-group differences were observed in total sleep time, wake after sleep onset, deep sleep duration, light sleep duration, daytime sleepiness, or sports injury severity score.
| 2026 | Year | 04 | Month | 07 | Day |
Participants were 25 female high school soccer players with a mean age of 16.5 years (SD 0.7). After randomization, 12 participants were allocated to the intervention group and 13 to the control group. At baseline, there were no significant between-group differences in demographic characteristics or any outcome measures.
A total of 25 participants were enrolled before study initiation. After randomization, 12 participants were assigned to the intervention group and 13 to the control group. There were no dropouts during the study period, and all 25 participants were included in the final analysis.
No serious adverse events related to the intervention were reported in this study.
The primary outcome was sleep efficiency. Secondary outcomes were total sleep time, wake after sleep onset, deep sleep duration, light sleep duration, rapid eye movement sleep duration, total PSQI-J score, Sleep Difficulty Score, total ESS-J score, Total Mood Disturbance score, and sports injury severity score assessed using a modified OSTRC-H-J. Objective sleep metrics were continuously measured using Fitbit Inspire 3, and questionnaire-based assessments were conducted at week 5 and week 10.
Completed
| 2025 | Year | 08 | Month | 01 | Day |
| 2025 | Year | 08 | Month | 01 | Day |
| 2025 | Year | 08 | Month | 20 | Day |
| 2025 | Year | 10 | Month | 01 | Day |
| 2026 | Year | 04 | Month | 07 | Day |
| 2026 | Year | 04 | Month | 07 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000070024