JMIR Publications

ABSTRACT

Background:

Scoliosis is a three-dimensional spinal deformity involving the coronal, sagittal, and axial planes. Traditional conservative treatments face challenges such as poor patient adherence, lack of continuous monitoring, and insufficient personalized feedback. The rapid advancement of digital health technologies offers new solutions for scoliosis rehabilitation, yet related research remains exploratory, lacking systematic analysis of different digital technology types, intervention characteristics, and clinical efficacy.

Objective:

To systematically evaluate the current application of digital health technologies in scoliosis treatment, categorize interventions based on technical characteristics, assess their clinical efficacy and safety, explore potential mechanisms of action, and provide evidence-based guidance for clinical practice.

Methods:

We searched PubMed, IEEE Xplore, Embase, and Web of Science databases (from inception to October 2025) to identify randomized controlled trials, non-randomized controlled studies, prospective cohort studies, and feasibility studies involving digital interventions for scoliosis. Methodological quality was assessed using the RoB 2 and ROBINS-I tools, while intervention reporting completeness was evaluated with the TIDieR checklist. Digital interventions were categorized into five types based on technical characteristics: synchronous telerehabilitation, asynchronous digital education, immersive VR/AR technology, wearable biosensing, and integrated digital health platforms. Intervention characteristics, clinical outcomes, and supervision models were systematically analyzed.

Results:

Twelve studies (507 patients) were included, comprising 7 RCTs, 2 prospective studies, 1 non-randomized controlled trial, and 2 feasibility studies from five countries. Quality assessment revealed that 42.9% of RCTs and 40% of non-randomized studies had low risk of bias, with intervention reporting completeness ranging from 75% to 100%. Evidence synthesis showed a positive trend toward improved Cobb angles with digital interventions. Specifically, integrated platforms and blended supervision models outperformed usual care, and multiple studies reached the minimal clinically important difference. Evidence for the Angle of Trunk Rotation improvement was heterogeneous, potentially influenced by variations in digital intervention protocols. Respiratory function and quality of life indicators showed improvement trends. Systems featuring bidirectional interaction and real-time feedback demonstrated higher adherence. Analysis of supervision models suggested that a hybrid approach combining ≥3 sessions per week with early in-person guidance may balance efficacy and feasibility. No serious adverse events were reported, indicating good safety.

Conclusions:

Digital health technologies demonstrate potential clinical value in improving spinal morphology, respiratory function, and quality of life for scoliosis patients. The three-tiered technology classification framework (Mature Application Tier, Emerging Technology Tier, and Frontier Exploration Tier) proposed in this study may guide clinical technology selection. However, existing evidence is limited by small sample sizes, short follow-up periods, and methodological heterogeneity. Future large-scale, long-term follow-up multicenter randomized controlled trials are needed to clarify optimal implementation strategies for digital interventions and their application value in precision rehabilitation for scoliosis. Clinical Trial: PROSPERO CRD420251250074