Pain Phenotyping Indicators in Older Adults with Chronic Low Back Pain: A Secondary Analysis of a Randomized Controlled Trial

Purpose: A multimodal approach to clinical care is often recommended for chronic low back pain (cLBP); however, treatment responses are highly variable. Phenotyping could help determine subgroups of patients, allowing for targeted and tailored interventional approaches. Patients and Methods: 263 (Mage=69.8 (7.2); 64.6% female) individuals with cLBP participated in the parent study, a 3-arm, randomized clinical trial examining auricular point acupressure treatment outcomes. Parent study participants were randomized (1:1:1) to APA ear points targeted to cLBP (T-APA, n=92), non-targeted to cLBP (NT-APA, n=91), or education control (n=89). The current study used latent class analysis to identify clustering for pain severity (intensity, neuropathic pain) and pain impact (anxiety, depression, pain catastrophizing, fatigue, sleep disturbance) and determine if these classes were related to treatment outcomes (pain and disability reduction). Bayesian Information Criterion (BIC) was used for model selection. Post-LCA, ANOVA and Fisher’s exact tests examined potential subgroup differences. Results: The three-class model emerged as the best fit due to relatively low BIC (−12105.46) and good patient distribution per class; class 1 n=79, class 2 n=109, class 3 n=75. Latent class 1 had moderate pain severity and pain impact, class 2 had high pain severity and pain impact, and class 3 had low pain severity and pain impact. No significant differences between classes were seen for age, sex, or BMI (p-value>0.05); however, latent class 3 had the highest physical functioning, lowest fear of physical activity, and disability, and significantly lower unemployment rate. There were no significant differences in treatment outcomes among the three classes. Conclusion: Three distinct clusters of factors related to pain and psychological function for individuals with cLBP were found. These clusters align with previous work and add to the literature by providing important associations with demographic and clinical factors that have not been previously examined.

Duke Scholars

Author Constance Margaret Johnson School of Nursing