Use of machine learning to identify prognostic variables for outcomes in chronic low back pain treatment: a retrospective analysis.
Carolyn Cheema, Jonathan Baldwin, Jason Rodeghero, Mark W Werneke, Jerry E Mioduski, Lynn Jeffries, Joseph Kucksdorf, Mark Shepherd, Carol Dionne, Ken Randall
Author Information
Carolyn Cheema: College of Allied Health, Department of Rehabilitation Sciences, The University of Oklahoma Health Sciences Center, Tulsa, OK, USA. ORCID
Jonathan Baldwin: College of Allied Health, Department of Medical Imaging and Radiation Sciences, The University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA.
Jason Rodeghero: Department of Public Health & Community Medicine, School of Medicine, Tufts University, Boston, MA, USA.
Mark W Werneke: Net Health Systems, Inc, Pittsburgh, PA, USA.
Jerry E Mioduski: Net Health Systems, Inc, Pittsburgh, PA, USA.
Lynn Jeffries: College of Allied Health, Department of Rehabilitation Sciences, The University of Oklahoma Health Sciences Center, Tulsa, OK, USA.
Joseph Kucksdorf: Bellin Health, Orthopedics and Sports Medicine, Green Bay, WI, USA.
Mark Shepherd: Physical Therapy Department Bellin College, Green Bay, WI, USA.
Carol Dionne: College of Allied Health, Department of Rehabilitation Sciences, The University of Oklahoma Health Sciences Center, Tulsa, OK, USA.
Ken Randall: College of Allied Health, Department of Rehabilitation Sciences, The University of Oklahoma Health Sciences Center, Tulsa, OK, USA.
OBJECTIVES: Most patients seen in physical therapy (PT) clinics for low back pain (LBP) are treated for chronic low back pain (CLBP), yet PT interventions suggest minimal effectiveness. The Cochrane Back Review Group proposed 'Holy Grail' questions, one being: 'What are the most important (preventable) predictors of chronicity' for patients with LBP? Subsequently, prognostic factors influencing outcomes for CLBP have been described, however results remain conflicting due to methodological weaknesses. METHODS: This retrospective observational cohort study examined prognostic risk factors for PT outcomes in CLBP treatment using a sub-type of AI. Bootstrap random forest supervised machine learning analysis was employed to identify the outcomes-associated variables. RESULTS: The top variables identified as predictive were: FOTO™ predicted functional status (FS) change score; FOTO™ predicted number of visits; initial FS score, age; history of jogging/walking, obesity, and previous treatments; provider education level; medication use; gender. CONCLUSION: This article presents how AI can be used to predict risk prognostic factors in healthcare research. Improving predictive accuracy helps clinicians predict outcomes and determine most appropriate plans of care and may impact research attrition rates.
