BACKGROUND AND OBJECTIVES: Up to 70% of individuals experience neck pain at some point in their lives, and approximately 30% of adults encounter it each year. This study examines the healthcare provider utilization for neck pain management based on patient factors such as age, race, and risk indices. METHODS: Using data from a quaternary academic center's health plan, claims of 12 615 patients were analyzed. Uniform manifold approximation and projection with K-means clustering and multinomial logistic regression of clusters identified global trends in the data. Multinomial logistic regression highlighted significant relationships between predictors, initial/subsequent visit providers, and imaging utilization. RESULTS: Analysis of 62 055 claims from 12 615 patients revealed significant disparities in provider utilization and cost of care. Uniform manifold approximation and projection identified a distinct cluster of patients with older age (mean 72.54 years) and high Charlson Comorbidity Index (mean 1.85) that predominantly visited emergency medicine initially but did not follow-up with a subsequent provider. Multivariate regression showed higher odds of seeing no subsequent provider in patients first visiting emergency medicine (odds ratio 67.41). Patients under Medicaid (log odds 1.76, P < .001) and Medicare (log odds 1.3, P < .001) were more likely to initiate care at the emergency department compared with those with commercial insurance. African American patients had greater odds of presenting initially to emergency medicine than Caucasian patients (log odds 1.38, P < .001). Men (β = 193.72, P < .001) and older age (β = 3.78, P < .01) were associated with higher opioid prescription costs. Opioids with greatest abuse potential showed decreased prescription costs (β = -439.13, P < .001). CONCLUSION: This study reveals healthcare utilization differences linked to demographic, insurance, and pharmaceutical factors in patients seeking care for chronic neck pain. Minority, male, and frail (high comorbidity burden) patients were more frequently seen in emergency settings, with a higher likelihood of being lost to follow-up, and slower transitions to additional care.
Mitha et al. (Mon,) studied this question.