Artificial intelligence-guided therapy: Clinic versus home users. Long-term differences in patient-reported outcomes in patients with low back pain
Background: Low back pain (LBP) is a common cause of disability in the United States and contributes to a great economic burden on both the individual and healthcare system level. Previous studies have investigated home-based therapy as an alternative or supplement to clinical care. In a previous study, we found comparable positive outcomes in patients who use an artificial-intelligence (AI) device to manage LBP in both home and clinical settings. In this study, we plan to compare long-term outcomes throughout 1-year post-intervention. Methods: This is a randomized clinical trial of 52 subjects. The investigation compared outcomes in LBP patients who used AI-device resistance therapy in either clinical or home settings. Outcomes of interest were pain, functional status, and kinesiophobia. Results: Group (Home vs. Clinic) was significantly associated with all three PRO parameters (PROMIS PI, PROMIS PF, and TSK scores) (P < 0.007). Home subjects, on average, had lower PROMIS PI and TSK scores (P < 0.007) and higher PROMIS PF scores (P < 0.0001); however, these differences were clinically insignificant. Time and the interaction of Time∗Group were not significant factors that affected PRO scores (P > 0.66). Conclusion: Patients who used AI-resistance training in either home or clinical settings had similar improvement by the end of the intervention and maintained this similarity throughout 1 year of follow-up. AI-resistance training has the potential to be a cost-effective supplement to clinical therapy when managing LBP. Irb approval number: 20-007281.
