Person-Centred AI Support in Interdisciplinary Rehabilitation for Chronic Pain

N/ANot Yet RecruitingNCT07081737

Bjorn Ang400 enrolled

Overview

This cluster randomized controlled trial evaluates whether a person-centred, AI-supported Clinical Decision Support System (CDSS) can improve outcomes and cost-effectiveness in interdisciplinary rehabilitation for people with complex chronic pain. The CDSS is designed to assist clinicians in making personalized treatment decisions within standard interdisciplinary treatment (IDT). It has been developed using machine learning models trained on real-world data from over 100,000 patients in the Swedish Quality Registry for Pain Rehabilitation (SQRP), linked to several national registers, including the National Patient Register, the Prescribed Drug Register, the Social Insurance Agency database (MiDAS), and the Cause of Death Register. This enables individualized predictions of treatment outcomes, work ability, and healthcare utilization. The trial includes 400 adult patients with chronic pain, enrolled at 20 IDT clinics randomized to either CDSS-supported or standard IDT. The study has three phases: feasibility, effectiveness, and implementation. The primary outcome is a patient-prioritized composite single-index of health-related well-being, based on domains such as pain, sleep, physical and mental health, emotional distress, and work ability. Patients prioritize these domains together with their clinical team, enabling a person-centred assessment. Secondary outcomes include HRQoL (EQ-5D, SF-36), emotional distress (HADS), and work ability (WAI), measured at baseline, post-treatment, 6- and 12-month follow-up. A parallel mixed-methods process evaluation will examine implementation outcomes such as usability, clinician adherence, and workflow integration, using logs, surveys (e.g., S-NoMAD), and interviews. Normalization Process Theory guides the analysis. Cost-utility will be assessed using QALYs and ICERs from a societal perspective, with long-term projections using simulation models. Results will be reported in peer-reviewed publications.

This project consists of three integrated phases aimed at evaluating a machine learning-based Clinical Decision Support System (CDSS) to improve interdisciplinary rehabilitation for individuals with complex chronic pain. The evaluation encompasses feasibility, clinical effectiveness, cost-utility, and implementation in routine care. The results will be reported in multiple peer-reviewed scientific publications. Phase 1: Development, validation, and feasibility By the end of 2025, the CDSS-developed in an ongoing project-will be ready for clinical testing. It is based on predictive models trained on registry-linked data from over 100,000 patients in the Swedish Quality Registry for Pain Rehabilitation (SQRP), linked to several national registers, including the National Patient Register, the Prescribed Drug Register, the Social Insurance Agency database (MiDAS), and the Cause of Death Register. The system provides personalized forecasts for treatment outcomes, long-term work ability, and healthcare use. A pilot cluster-RCT will be conducted at 10 clinics (5 patients per site) to evaluate feasibility outcomes such as recruitment, retention, usability, data completeness, and workflow fit. These will be assessed using structured surveys, usage data, and interviews. Outcome measures will be collected at baseline, immediately after the intervention (i.e., up to 18 weeks after baseline), and at 12-month follow-up. While a typical interdisciplinary rehabilitation program lasts 6-8 weeks, some clinics may extend the intervention up to 18 weeks (with less treatment occasions per week) due to their ordinary and existing treatment procedures at that specific clinic. Published results indicate however no significant differences in treatment outcomes based on such extended program duration (Tseli et al., 2020). No major changes to the CDSS algorithm or interface are planned during the trial. Phase 2: Clinical effectiveness and health economic evaluation The full evaluation will be conducted through a non-registry-based cluster randomized controlled trial involving 400 patients across 20 interdisciplinary rehabilitation clinics. Outcomes will be analyzed using linear mixed-effects models adjusted for time, group, clustering, and covariates. The primary endpoint is at 12-month follow-up. Secondary outcomes will be assessed at baseline, up to 18 weeks after baseline (i.e., immediately post intervention), and at 6- and 12-month follow-up. Health economic analyses will include within-trial cost-utility evaluation (QALYs from EQ-5D and SF-36) and longer-term modelling using Markov or microsimulation methods. Both direct (healthcare) and indirect (productivity loss) costs will be included. Sensitivity analyses will address uncertainty and robustness. Phase 3: Implementation research A mixed-methods process evaluation will examine real-world adoption, scalability, and sustainability. Data will include system logs (e.g., reach, fidelity), survey responses (S-NoMAD), and interviews with clinicians and decision-makers. Analysis is guided by Normalization Process Theory, focusing on coherence (understanding), cognitive participation (engagement), collective action (integration), and reflexive monitoring (clinical utility). This structure enables a rigorous, practice-oriented evaluation of AI support in pain rehabilitation, integrating clinical, economic, and implementation perspectives to guide responsible and scalable integration into healthcare.

Eligibility

Sex: ALLMin age: 18 YearsMax age: 67 Years

Medical Language ↔ Plain English

Inclusion Criteria: * Aged 18 to 67 years * Diagnosed with chronic non-malignant pain persisting longer than 3 months * Pain condition includes, but is not limited to: fibromyalgia, widespread pain, back pain, neck pain, or shoulder pain * Eligible for and referred to interdisciplinary rehabilitation (IDT) at a participating clinic * Willing and able to participate in digital assessment and follow-up procedures * Able to communicate and complete study materials in Swedish * Provides written informed consent Exclusion Criteria * Pain caused by malignancy or cancer-related treatment * Pain caused by systemic diseases such as rheumatoid arthritis, lupus, or other autoimmune or inflammatory conditions * Severe psychiatric conditions interfering with study participation (e.g., untreated psychosis or severe depression requiring immediate psychiatric care) * Documented cognitive impairment limiting ability to understand study participation or complete self-reported measures * Currently enrolled in another interventional clinical trial that may confound the outcomes of this study * Not expected to remain in the clinic's follow-up system for the duration of the study

Outcomes

Primary Outcomes

Change from Baseline in Patient-Prioritized Health-Related Well-being Composite at 12 Months

A person-centred composite score based on eight validated domains: pain intensity (NRS; 0-10), sleep problems (ISI; 0-28), physical health (SF-36 PF; 0-100), mental health (SF-36 MH; 0-100), depression and anxiety (HADS-A/D; 0-21), work ability (WAI single item; 0-10), and pain interference (single item; 0-10). At baseline, participants prioritize these domains together with the interdisciplinary treatment (IDT) team. The Clinical Decision Support System (CDSS) stores these weights. Each domain is normalized to 0-100 and combined into a weighted composite. Higher scores indicate better health-related well-being. Although composed of several scales, the outcome is reported as a single, aggregated primary measure.

Time frame: (1) Baseline, (2) up to 18 weeks after baseline*, and (3) 12-month follow-up. (*Most programs last 6-8 weeks, but some clinics extend to 18 weeks (fewer sessions/week) due to existing routine practice).

Secondary Outcomes

Change from Baseline in Pain Intensity (NRS) at 12 Months

Measured using the Numeric Rating Scale (NRS, 0-10), where 0 = no pain and 10 = worst imaginable pain. Participants report average pain over the last seven days. A reduction in score indicates clinical improvement.

Change from Baseline in Sleep Problems Measured by the Insomnia Severity Index (ISI) at 12 Months

The ISI assesses perceived sleep difficulties through seven items. Total scores range from 0 to 28, with higher values indicating more severe insomnia. A lower score reflects improved sleep quality.

Change from Baseline in Physical Health Functioning (SF-36 PF) at 12 Months

Measured using the SF-36 Physical Functioning subscale (10 items). Scores range from 0 to 100, with higher scores representing better physical functioning.

Change from Baseline in Mental Health (SF-36 MH) at 12 Months

Measured using the SF-36 Mental Health subscale (5 items). Scores range from 0 to 100; higher values indicate better mental well-being.

Change from Baseline in Emotional Distress Measured by the Hospital Anxiety and Depression Scale (HADS) at 12 Months

The HADS includes two 7-item subscales for anxiety and depression. Each subscale ranges from 0 to 21; higher scores indicate greater emotional distress.

Change from Baseline in Work Ability Measured by the Work Ability Index (WAI) Single Item at 12 Months

Assessed via the single-item WAI: "Current work ability compared to lifetime best", scored from 0 (completely unable to work) to 10 (work ability at its best).

Change from Baseline in Pain Interference in Daily Life at 12 Months

Measured using validated items adapted from the Multidimensional Pain Inventory or equivalent, reflecting the extent to which pain disrupts everyday activities. Scores are standardized from 0 (no interference) to 100 (maximum interference).

Change from Baseline in Physical Activity Measured by the Exercise Vital Sign (EVS) at 12 Months

The EVS includes two questions: (1) days per week of moderate-to-strenuous activity, and (2) average minutes per day. The product gives total minutes/week. Validated against objective activity measures.

Treatment Expectations at Baseline

A single-item self-report question regarding the individual's expectation of treatment benefit, rated on a Likert scale from "no improvement expected" to "complete recovery expected".

Time frame: At enrollment (baseline) only.

Change from Baseline in Sick Leave Status at 12 Months

Self-reported current sick leave status (full-time, part-time, or not on leave). Changes over time reflect return-to-work or ongoing sick leave status.

Change from Baseline in Self-Reported Medication Use at 12 Months

Captured through a validated web-based tool. Participants report medication name (text), dose (mg), frequency (time/day), and form (e.g. tablet). Variables are analyzed separately using descriptive statistics and longitudinal methods to assess changes in medication use over time.

Change in Cost-Utility at 24 Months (EQ-5D-5L)

Measured in Quality-Adjusted Life Years (QALYs) derived from the EuroQol 5-Dimension 5-Level (EQ-5D-5L) instrument using standard value sets. The Incremental Cost-Effectiveness Ratio (ICER) will be calculated as the ratio of the difference in costs to the difference in health outcomes (QALYs gained), comparing the intervention to the control. ICER provides a summary measure of the additional cost required to gain one additional unit of health benefit. EQ-5D index scores typically range from below zero (e.g., -0.594 in the UK value set) to 1.000, or from 0.296 to 1.000 using the Swedish experience-based value set, where higher values indicate better health-related quality of life.

Time frame: (1) Up to 18 weeks after baseline, (2) 12 months, and (3) 24 months after treatment completion.

Change in Cost-Utility at 24 Months (RAND SF-36)

Measured using the RAND 36-Item Health Survey (RAND SF-36), which assesses physical and mental health across eight domains. Each domain score ranges from 0 to 100, with higher scores indicating better health status. A cost-utility analysis will also be performed using utility estimates derived from RAND SF-36, and an ICER will be calculated accordingly. ICER will be interpreted as the additional cost per unit of utility gained, based on mapped utility values. (Note: Mapping algorithms may be applied to derive utility weights from RAND SF-36, subject to validation.)

Time frame: (1) Up to 18 weeks after baseline, (2) 12 months, and (3) 24 months after treatment completion.

Person-Centred AI Support in Interdisciplinary Rehabilitation for Chronic Pain

Overview

Conditions

Interventions

Eligibility

Locations (1)

Outcomes

Primary Outcomes

Secondary Outcomes

Central Contacts