Insomnia
Conditions
Keywords
Chatbot, Youth, Insomnia, CBT-I, RCT
Brief summary
Insomnia is prevalent in youth, and it associates with depression and other psychiatric disorders, leading to increased mental health burden. Cognitive Behavioral Therapy for Insomnia (CBT-I) is recommended as first-line treatment for insomnia. Digital tools have been employed to automate mental health interventions, in order to address deterrents such as clinician shortage, limited appointment availability, high cost, and stigma of seeking help. Digital CBT-I is shown to be effective in treating insomnia. Future digital intervention will incorporate patient-centered design, input from key stakeholders, and new understandings of behavior change. Artificial Intelligence (AI)-powered chatbots are utilized in different industries for better customer experience. AI chatbot is also utilized in the mental health industry to extend the boundary of digital interventions from accommodating didactic and informational content to providing interactive, intelligent, and most importantly, patient-centered conversational agents. Some famous AI mental health chatbots in Western societies were developed to give tailored feedback, respond to emotions that a user expresses, and encourage users to complete an intervention. This study will investigate the effect of a CBT-I chatbot on insomnia to provide further evidence on mental health chatbot.
Interventions
BEHAVIORALCognitive Behavioral Therapy for Insomnia in Chatbot
Digital interventions can relieve the worldwide burden of mental disorders. The low set-up costs and barriers of online platforms make digital interventions very cost-effective. By using the Internet as a delivery medium, many people can enjoy unrestricted access to self-help information. Unlike traditional face-to-face intervention, the effects of digital self-help interventions are scalable. The current study attempts to extend the boundary of digital interventions from accommodating didactic and informational content to providing interactive, intelligent, and most importantly, patient-centered conversational agents. AI chatbots can provide suitable recommendations and training materials to users according to their behavioral, mental, and motivational readiness. Since existing AI chatbots are developed for Western societies, a culture-specific Chinese chatbot will fill the research and service gaps.
Sponsors
Chinese University of Hong Kong
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 35 Years
Healthy volunteers
Yes
Inclusion criteria
* score 8 or higher on the Insomnia Severity Index (ISI) * between 18 and 35 years old * read and type Chinese * own a smartphone * have consistent internet access to their smartphone
Exclusion criteria
* have a self-reported diagnosis of any psychiatric disorder(s) * have been on regular medication(s) in the past month * do not have instant messaging app(s) will be excluded.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| The Insomnia Severity Index (ISI) | 10-20 minutes | The Insomnia Severity Index (ISI) will be used to measure insomnia symptoms in the previous month. ISI includes 7 items with a five-point Likert format (0 = not at all to 4 = very much). The total score ranges from 0 to 28. A score ≥ 8 indicates clinically significant insomnia in Chinese adolescents (sensitivity = 87%, specificity = 75%). The Cronbach's alpha and the test-retest reliability of ISI were 0.83 and 0.79, respectively. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| The reduced Horne and Östberg Morningness and Eveningness Questionnaire (rMEQ) | 10-20 minutes | The reduced Horne and Östberg Morningness and Eveningness Questionnaire (rMEQ) was used to measure chronotype preference. rMEQ consists of 5 items where the first 4 items were scored from 1 to 5 while the last item was scored from 0 to 6. The total score ranged from 4 to 25. Three classified types of chronotype were eveningness (score \< 12), intermediate-type (score 12-17), and morningness (score \> 17). The Cronbach's alpha and the test-retest reliability of rMEQ were 0.70 and 0.77, respectively. |
| The Patient Health Questionnaire-9 (PHQ-9) | 10-20 minutes | The Patient Health Questionnaire-9 (PHQ-9) will be used to measure depression symptoms in the past two weeks. PHQ-9 includes 9 items with a four-point Likert format (0 = not at all to 3 = nearly every day). The total score ranges from 0 to 27. A score ≥ 10 indicates clinically significant depression in Chinese population (sensitivity = 88%, specificity = 88%). The Cronbach's alpha and the test-retest reliability of PHQ-9 were 0.86 and 0.84, respectively. |
| Dysfunctional Beliefs and Attitudes about Sleep scale (DBAS-16) | 10-20 minutes | Dysfunctional Beliefs and Attitudes about Sleep scale (DBAS-16) is a shortened version of the original DBAS, it will evaluate sleep-disruptive cognitions. DBAS-16 includes 16 items with a ten-point Likert format (0 = strongly disagree to 10 = strongly agree). The total score is based on the average score of all items. A higher score reflects greater dysfunctional beliefs about sleep. A Pearson correlation coefficient computed between the total scores showed a significant correlation, r(72) = 0.83, P \<0.0001, suggesting adequate temporal stability. A paired t-test revealed that the total score of the DBAS-16 decreased significantly from the first (mean = 4.95, SD = 1.35) to the second (mean = 4.57, SD = 1.48) administration. Cronbach alpha values of 0.77 (clinical) and 0.79 (research) indicate adequate internal consistency. |
| Sleep Hygiene Index (SHI) | 10-20 minutes | Sleep Hygiene Index (SHI) is used to assess sleep hygiene. Each item is rated on a five-point scale ranging from 0 (never) to 4 (always). Total scores range from 0 to 52, with a higher score representing poorer sleep hygiene. Cronbach's alpha = 0.66 and test-retest reliability (r = 0.71). Chinese version of SHI shows internal consistency (α = 0.62, ω = 0.63) and stability (test-retest reliability = 0.90). |
Outcome results
None listed