INTRODUCTION: Collaborative modeling has recently been used to investigate the most important factors related to low back pain (LBP) by documenting the perspective of experts from multiple disciplines using fuzzy-logic cognitive maps (FCMs).  FCMs represent an individual’s conceptual understanding (mental model) of a problem and have highlighted the perceived importance of psychological factors in LBP. Although clinicians may consider psychological factors when treating patients with LBP, the relative importance of such factors may differ between disciplines.  This study aimed to 1) investigate how osteopathic neuromusculoskeletal medicine (ONMM) and physical medicine & rehabilitation (PM&R) physicians conceptualize the factors and their interactions involved in LBP and 2) compare how these physician groups’ models of LBP are different (or similar) to a multidisciplinary cohort with expertise in LBP (EXP).

METHODS: ONMM (n=15) and PM&R (n=16) physicians were compared to a third EXP group of scientists/clinicians (n=29) that represented the following disciplines: Basic Science (n=3), Epidemiology(n=2), Biomechanics (n=3), Chiropractic (n=4), Spine Surgery (n=2), PM&R (n=2), Physical/Exercise Therapy (n=11), and Psychology (n=2). Each participant underwent a structured one-on-one interview to construct FCMs using online software (www.mentalmodeler.org). This process involved the nomination of factors contributing to patients’ outcomes (i.e., pain, disability, and quality of life) and the weighting of the connections (strength of the effect) between these factors. Factors from each model were allocated to one of eight categories: 1) Nociceptive Detection & Processing, 2) Behavioral/Lifestyle 3) Tissue Injury/Pathology, 4) Psychological, 5) Social/Work/Contextual, 6) Biomechanical, 7) Comorbidities and 8) Individual Factors.  The centrality of each category (sum of centrality for each factor within the designated category) was calculated to determine the importance of factors represented in the FCMs and was expressed as a percentage of the eight categories. FCMs were aggregated into a metamodel for each group to assess the collective opinion it represented.  The cognitive diversity index (CDI) was calculated to reflect how many different categories are represented in an FCM. Individual FCMs and metamodels were analyzed with descriptive statistics and ANOVA was used to detect a statistical difference in model metrics among the three groups (p<0.05).

RESULTS:  There were 1464 Factors and 4799 Connections generated for all FCMs; however, there was no significant difference in the number of Factors (p=0.67), Connections (p=0.27), or CDI (p=0.67) between the three groups. Psychological Factors was clearly the most prominent category in the EXP group accounting for 28.1% of the centrality, whereas Social/Work/Contextual Factors was the most prominent in the PM&R and ONMM groups accounting for 26.3% and 18.6%, respectively (Figure 1).  Social/Work/Contextual Factors was the second most prominent category in the EXP group, while Psychological Factors was second in the ONMM and PM&R groups (Figure 2).

DISCUSSION:  Although Social/Work/Contextual Factors was the most central/important category in the ONMM and PM&R groups, other categories had similar relative importance in the respective models. These findings speak to the perceived importance of psycho-social and contextual factors from the scientific and clinical community and highlight the unique perceptions of ONMM and PM&R physicians in the management of patients with LBP.