Evaluation of a Chatbot-Delivered Structured Psychological Intervention for Teenagers With Adolescent Idiopathic Scoliosis

Evaluation of a Chatbot-Delivered Structured Psychological Intervention (SPI-Bot) for Teenagers With Adolescent Idiopathic Scoliosis: A Pilot Randomized Controlled Trial

Adolescent idiopathic scoliosis is the most common spinal deformity in children and can significantly impact the psychological health of affected teenagers. Despite this, there are few effective psychological treatments available for this population. Cognitive and behavioral strategies show promise in addressing the challenges associated with the condition and its treatment. Chatbots could serve as an accessible and effective tool to support the psychological health of these teenagers.

Using the Centre for eHealth Research and Disease Management Roadmap as a guide, a chatbot-based structured psychological intervention has been developed. This intervention incorporates cognitive and behavioral strategies and aims to improve the psychological health of teenagers with adolescent idiopathic scoliosis. The current phase involves implementing the intervention and conducting a pilot test to assess its feasibility, acceptability, and preliminary effectiveness.

The study will employ a pilot randomized controlled trial design, nested with qualitative interviews. Participants will be recruited from community health organizations in Hong Kong and will be randomly assigned to either a control group or an intervention group using block randomization. The intervention group will engage with the chatbot-based structured psychological intervention for eight consecutive weeks. The control group will interact with a different generative chatbot, participating in casual conversations at a similar frequency. Quantitative assessments will be conducted at three points: baseline, immediately after the intervention, and one month after the intervention. Additionally, qualitative interviews will be held with selected participants from the intervention group at post-intervention to explore their experiences and perceptions of the intervention process.

Study Overview

• Status: Not yet recruiting
• Conditions: Adolescent Idiopathic Scoliosis (AIS); Psychological Health
• Intervention / Treatment: Other: Intervention Group; Other: control group

Detailed Description

Introduction: Adolescent idiopathic scoliosis (AIS) is the most common form of scoliosis, affecting 0.9% to 12.0% of the global population. In Hong Kong, AIS affects about 4.7% of adolescents, with an annual increase of 0.61%. AIS may cause physical deformity and back pain, and severe cases may lead to long-term health issues like impaired lung function and functional limitations in adulthood. Adolescence is crucial for psychological development, with many mental disorders beginning during this period. Teenagers with AIS face additional challenges, such as body image issues, restricted physical function, and pain, increasing their risk for psychological problems. They often report stress, anxiety, and depression, which can lead to more severe mental health issues if untreated. Social isolation and avoidance can worsen these psychological challenges.

Despite the importance of psychological health, it is often overlooked in AIS treatment. While the new medical model emphasizes integrating psychological care with physical treatment, few clinical trials focus on AIS, mainly addressing perioperative anxiety with limited psychological benefits. Cognitive behavioral therapy (CBT) has proven effective for children with long-term conditions and holds promise for AIS. Cognitive techniques can help modify biased thoughts, while behavioral approaches like relaxation training and activity scheduling can alleviate distress and encourage positive behaviors. However, research on CBT for AIS is limited and methodologically flawed, leaving its effectiveness unproven.

Access to psychological interventions like CBT is challenging due to resource shortages and personal barriers such as stigma and cost. Chatbots offer a potential solution by providing accessible, anonymous, and flexible interventions. They are well-received and engaging for youth but have not yet been explored for the AIS population.

Objectives: To address the gap in psychological health among the AIS population, the research team has developed a chatbot-delivered structured psychological intervention (SPI-Bot) that incorporates cognitive and behavioral strategies. The objectives of this pilot randomized controlled trial (RCT) are as follows: 1) The SPI-Bot will be feasible and acceptable, as measured by recruitment, adherence, and attrition rates, utilization and engagement, working alliance, usability, user experience, and adverse events record (primary outcome). 2) The SPI-Bot (intervention) group will be more significantly effective in enhancing psychological distress, psychological well-being, pain, perceived body image, physical function, perceived social support, and quality of life compared to the control group, with the potential for this effect to persist for one-month post-intervention.

Design: The trial is an open-label, double-arm, and single-center pilot RCT with a 1:1 allocation ratio and includes a follow-up evaluation. The intervention group will use the SPI-Bot for eight consecutive weeks. Meanwhile, the control group will interact with a different generative chatbot, engaging in casual conversations at a similar dosage to ensure balanced attention and interaction effects. This clinical trial will be reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT) statement for pilot trials.

Participants and recruitment: Participants will be recruited from community health organisations in Hong Kong. As part of the registration, the participants will have to fill out an initial screening questionnaire that assesses inclusion criteria.

Assessments will take place starting with psychological symptoms screening in hospital outpatient clinic followed by three assessment points: baseline (T0), eight weeks after randomization (T1, post-intervention), and twelve weeks after randomization (T2, follow-up assessment).

• Study Type: Interventional
• Enrollment (Estimated): 52
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Jiaying Li; Phone Number: 852+34002576; Email: jia98ying.li@connect.polyu.hk
• Study Contact Backup: Name: Yan Li; Phone Number: 852+27666388; Email: yan-nursing.li@polyu.edu.hk

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Have a Cobb angle (a standard method for measuring spinal curvature) greater than 10 degrees;
• Exhibit mild to moderate levels of depression, anxiety, or stress (scoring between 7-13 on the depression subscale, 6-12 on the anxiety subscale, or 12-16 on the stress subscale of the Depression Anxiety Stress Scales for Youth);
• Be aged between 10 and 18;
• Be proficient in Cantonese and traditional Chinese character reading;
• Own a smartphone and being adept at using it.

Exclusion Criteria:

• Have a diagnosis of psychiatric disorders or serious physical diseases;
• Are currently undergoing other psychological interventions;
• Are on a surgery waiting list or in the perioperative period (from seven days before to seven days after surgery);
• Are unable to provide consent for study participation;
• Exhibit self-harm or suicidal tendencies (scoring > 0 on the ninth item of the Patient Health Questionnaire-9: "Thoughts that you would be better off dead, or of hurting yourself in some way").

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Other
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Intervention group; The chatbot-delivered structured psychological intervention (SPI-Bot)	Other: Intervention Group; The intervention group will engage with SPI-Bot, a conversational agent designed to deliver cognitive and behavioral strategies for managing psychological symptoms through fully automated interactions. SPI-Bot operates on a template-based platform with predefined dialogue options, utilizing natural language processing technology. The intervention spans eight weeks and includes key components such as psychoeducation, cognitive restructuring, relaxation training, activity scheduling, problem-solving skills training, and social skills training. Participants will complete one session per week over the eight-week period. Participants will receive weekly contact from the research team to report AIS symptoms, receive homework reminders, and address any intervention-related questions.
Active Comparator: Control group; Attention control group	Other: control group; Participants in the control group will interact with a different generative chatbot, engaging in casual conversations at a similar dosage to ensure balanced attention and interaction effects.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Recruitment Rate	The recruitment rate will be calculated as the percentage of eligible participants who enroll in the study out of the total number of eligible participants, along with documenting the reasons for declining participation.	Pre-intervention
Attrition Rate	The attrition rate is the percentage of participants lost to follow-up, and the reasons for attrition will also be recorded.	Immediately after intervention and at 1-month follow-up
Engagement	De-identified aggregated chatbot usage data will be obtained from the developers to evaluate engagement. Metrics collected will include aggregated descriptive summaries: total check-ins, days in the chatbot, and messages sent during the eight-week intervention period.	Immediately after intervention
Working Alliance	The Working Alliance Inventory-Short Revised will be used. It measures the three aspects of the therapeutic alliance: agreement on the tasks of therapy; agreement on the goals of therapy; and development of an affective bond between patient and therapist. The scores of all 12 items range from 1 (never) to 5 (always).	Immediately after intervention
Usability	Usability of the intervention will be evaluated using the System Usability Scale, a tool capable of assessing the usability of diverse technology-based applications. This scale comprises 10 questions to be rated on a five-point Likert scale ranging from 'Strongly Disagree' to 'Strongly Agree'.	Immediately after intervention
User Experience	Participants from the intervention group will be chosen for individual semi-structured interviews based on their adherence status (completed vs. incomplete). Participants will be informed about the interview's objectives and process, and will be encouraged to provide feedback on the experience and perception of the SPI-Bot, the difficulty of using it, and the suggested improvements.	Immediately after intervention
Adverse events	Adverse events related to the intervention, such as dizziness, fatigue, or headache from smartphone use, will be monitored and recorded through participant self-reports.	During the intervention
Adherence Rate	The adherence rate will be monitored through backend data, defined as the percentage of participants who complete the intervention.	Immediately after intervention

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Psychological Distress	This will be measured by the Depression Anxiety Stress Scales for Youth. Participants will be asked to rate on four-point Likert scale from 0 (never) to 3 (almost always). Subscale scores can be calculated by adding the scores for the individual items, with higher scores indicating greater levels of distress.	Pre-intervention, immediately after intervention and at 1-month follow-up
Psychological Well-Being	Participants' well-being will be assessed using the WHO-5 Well-Being Index, which is a brief five-item self-reported measure. Participants will be asked to report their experiences in the past two weeks on a six-point Likert scale, ranging from 0 (at no time) to 5 (all of the time). The raw well-being score theoretically ranges from 0 (absence of well-being) to 25 (maximal well-being). The WHO-5 has shown both clinical and psychometric validity and has been previously integrated in psychological health-related mobile applications.	Pre-intervention, immediately after intervention and at 1-month follow-up
Perceived Social Support	The study will use the Multidimensional Scale of Perceived Social Support, comprising three domains: family support, friend support, and other support, totaling 12 items. This seven-point Likert scale assesses perceived social support, with a score range of 12 to 36 indicating low support, 37 to 60 indicating moderate support, and 61 to 84 indicating high support. The Chinese version has demonstrated strong psychometric properties when used with adolescents.	Pre-intervention, immediately after intervention and at 1-month follow-up
Scoliosis-Related Quality of Life	The refined Scoliosis Research Society 22-item questionnaire will be employed to measure the quality of life. It consists of five domains: pain, self-image/appearance, function/activity, mental health, and management satisfaction. Each item is scored from 1 (worst) to 5 (best).	Pre-intervention, immediately after intervention and at 1-month follow-up

Collaborators and Investigators

• Sponsor: The Hong Kong Polytechnic University

Study record dates

Study Major Dates

• Study Start (Estimated): August 15, 2025
• Primary Completion (Estimated): February 1, 2026
• Study Completion (Estimated): February 1, 2026

Study Registration Dates

• First Submitted: November 15, 2024
• First Submitted That Met QC Criteria: November 19, 2024
• First Posted (Actual): November 21, 2024

Study Record Updates

• Last Update Posted (Actual): June 10, 2025
• Last Update Submitted That Met QC Criteria: June 5, 2025
• Last Verified: November 1, 2024

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Bone Diseases; Musculoskeletal Diseases; Spinal Diseases; Spinal Curvatures; Scoliosis
• Other Study ID Numbers: HSEARS20240919007

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No