Psychosocial Subgroups in Chronic Low Back Pain Identified by Latent Profile Analysis
Overview
This study used latent profile analysis to identify four distinct biobehavioral phenotypes among chronic low back pain patients, revealing heterogeneity in psychosocial and pain-related characteristics. The findings were validated across two large cohorts, supporting the model's robustness and clinical relevance for personalized treatment.
Background
Chronic low back pain (cLBP) is a highly prevalent condition and a leading cause of disability and healthcare utilization. Biobehavioral factors including psychological distress, coping behaviors, and social determinants contribute to the persistence and variability of cLBP. Latent profile analysis (LPA) is a statistical method that identifies homogeneous subgroups within heterogeneous populations based on multiple continuous variables. Prior research has demonstrated the utility of LPA in uncovering clinically meaningful chronic pain subgroups, which can inform tailored interventions.
Data Highlights
Class
Description
Train Set Size (N=2954)
Test Set Size (N=398)
Class 1
High Distress and Maladaptive Behaviors
701
127
Class 2
Resilient and Adaptive Coping
413
108
Class 3
Intermediate Maladaptive Patterns
893
95
Class 4
Emotionally Regulated with High Pain Burden
947
68
Model fit metrics: BACKHOME (train set) AIC = 77,792; BIC = 78,338; Entropy = 0.82. COMEBACK (test set) AIC = 72,437; BIC = 73,880; Entropy = 0.81. Significant differences between classes were observed in pain self-efficacy, fear avoidance, emotional awareness (P < .05), and longitudinal changes in pain severity and quality of life (P ≤ .001).
Key Findings
Four distinct psychosocial subgroups of cLBP patients were identified, reflecting varying levels of distress, coping, and pain burden.
Class 1 exhibited high anxiety, depression, and fear avoidance behaviors indicating maladaptive coping.
Class 2 showed low maladaptive behaviors and high pain self-efficacy, representing resilient coping strategies.
Class 3 had moderate psychological and behavioral challenges, an intermediate phenotype.
Class 4 demonstrated strong emotional regulation despite experiencing a high pain burden.
The model showed strong fit and generalizability across two independent cohorts, supporting clinical utility.
Clinical Implications
Recognition of these distinct psychosocial phenotypes in cLBP patients can guide clinicians in tailoring interventions to individual patient profiles. For example, patients in the high distress group may benefit from targeted psychological therapies addressing anxiety and fear avoidance, while resilient patients might require different support strategies. This personalized approach could improve treatment outcomes and quality of life.
Conclusion
This study advances understanding of chronic low back pain heterogeneity by identifying clinically meaningful psychosocial subgroups through latent profile analysis. These findings support the development of personalized treatment strategies to address the complex biopsychosocial nature of cLBP.
References
NIH HEAL BACPAC Consortium -- Identification of Psychosocial Subgroups in Chronic Low Back Pain Through Latent Profile Analysis of Biobehavioral Phenotypes
by Fatemeh Gholi Zadeh Kharrat, Prakruthi Amar Kumar, Wolf Mehling, Irina Strigo, Jeffrey Lotz, Thomas A Peterson, REACH Investigators, Jamie Ahn, Kristina Benirschke, Alexandra Bryson, Katherine Bunda, Briana Davis, Carolina Dorofeyev, Rosalee Espiritu, Pirooz Fereydouni, Aamna Haq, Nicholas Harris, Sara Honardoost, Gabriel Johnson, Jennifer Johnson, Edward Lingayo, Robert Miller, Phirum Nguyen, Christopher Orozco, Lindsay Ruiz-Graham, Kie Shidara, Kaitlyn Smith, John (Boyuan) Xiao, Michelle Yang
Burnout is easing. Sleep science is getting weird. And dental schools have been winging cadaver training for 50 years. This week's research is full of good news that immediately complicates itself.
