(186) - Identification of latent classes in chronic pain within low-income settings

Abstract

The majority of research examines associations between individual symptoms of chronic pain. Although this approach is important, chronic pain involves a combination of multiple concurrent symptoms. This study sought to identify groups with similar pain and psychosocial factors using latent profile analysis (LPA). Identified subgroups with similar pain profiles were then evaluated for health care differences between subgroups. A total of 290 participants were examined from the Learning About My Pain (LAMP) trial, a randomized comparative effectiveness study of group-based psychosocial interventions (PCORI Contract #941, Beverly Thorn, PI; clinicaltrials.gov identifier NCT01967342). LPA was used to classify subgroups based on participants' scores on pain-related variables (pain severity, pain interference, disability, and number of pain sites) and psychosocial variables (pain catastrophizing, depression, anxiety, anger, and sleep). A two-profile solution was supported statistically using model fit indices and substantively using conditional response means. Low Pain (n = 165, 56.9%) was characterized by below-average across all variables entered into the analysis. Conversely, High Pain (n = 125, 43.1%) was characterized by above-average scores across all variables. Age, poverty status, and primary literacy were significant covariates in profile membership, in which the “High” subgroup had younger participants, higher poverty status, and lower primary literacy. An ANCOVA was conducted to examine whether the profile groups differed in health care utilization when controlling for age, poverty status, and literacy. The High Pain group had significantly more health care visits. The results suggest that the heterogeneity of biopsychosocial characteristics of chronic pain within this low-income setting can be accounted for by two subgroups that differ on health care utilization. A pain profile approach may be helpful in identifying groups of patients with increased needs for specific interventions.