Special Poster Session 50th International Society for the Study of the Lumbar Spine Annual Meeting 2024

Introduction: "Whole person health" and "precision medicine" are approaches to disease prevention and intervention that are increasingly discussed as viable new directions for the management of patients with low back pain (LBP).  The whole person health approach recognizes the importance of addressing psychological, social, and behavioral factors in conjunction with the biologically targeted interventions to maximize treatment effectiveness.  Precision medicine strives to tailor a multimodal treatment to precisely match the individual patient's phenotype consisting of all factors contributing to LBP.  While the research in this direction, utilizing "big data" and machine learning methods, is emerging, it would be helpful to envision what would be required to implement such interventions strategies in the future.  Therefore, this study simulated various combinations of multimodal interventions using a comprehensive, collaborative model of LBP that synthesizes the opinions of LBP experts from diverse disciplines¹.

Methods: Clinicians and Scientists (n=29) with expertise in LBP (e.g., publications, contributions to societies, etc.), representing diverse disciplines (Basic Science (n=7), Chiropractic (n=4), Spine Surgery (n=2), Physical Medicine & Rehabilitation (n=2), Physical/Exercise Therapy (n=12), and Psychology (n=2)), contributed collaboratively to the model. Each participant underwent a structured one-on-one interview to construct a fuzzy cognitive map (Mental Modeler software, www.mentalmodeler.org) representing their understanding of factors involved LBP, their interaction, and how various interventions affect patient outcomes (pain, disability, quality of life (QoL)).  Individual maps were then converted to adjacency matrices and aggregated into one comprehensive model.  Various combinations of multimodal interventions (up to k=6 treatments) were simulated and their relative effects on pain, disability, and QoL were examined by comparing them to the effects of all listed treatments together.  Combinations of treatments with the largest effects were then identified for each k and outcome.

Results: The model consisted of 272 factors, 1429 connections (interactions) among them, and 41 treatments.  Simulations indicated that increasing the number of concurrent treatments results in the diminishing effects due to the significant overlapping among treatment pathways (Figure 1).  Exercise Therapy was present in every multimodal combination across each outcome (pain, disability, and QoL), suggesting it is the most general treatment (Table 1).  Other treatments in the most effective multimodal interventions were: Acceptance Therapy, Counseling and Education, Pain Relieving Intervention, Cognitive Behavioral Therapy, Nutritional Counseling, and Physical Therapy.

Discussion: The aggregated, comprehensive model of LBP was extremely complex.  The simulations provided outcomes that broadly agreed with published data, suggesting that Exercise Therapy, Acceptance Therapy, Counseling and Education, Pain Relieving Intervention, Cognitive Behavioral Therapy, Nutritional Counseling, and Physical Therapy are effective LBP treatments in multimodal interventions.  Based on these results, progressing towards "precision medicine" in LBP will require overcoming two major barriers: 1) Precisely determining an individual patient's phenotype, which might consist of many factors, could prove to be very difficult, and 2) Tailoring individual treatment, particularly when currently available treatments for LBP lack the precision because their mechanisms work via multiple overlapping pathways.  Future studies will examine the overall impact of multimodal interventions by simulating a larger LBP population of individuals with various phenotypes.

Figure 1