Geographic variation in the treatment of spinal disorders: association with health care professional availability, and population socioeconomic status, race, and ethnicity. A retrospective cohort study

Abstract

AbstractImportanceSpinal disorders are common and associated with high rates of low value care. Efforts to improve guideline concordance and value benefit from understanding the influence of population and environmental factors.ObjectiveQuantify geographic variation in population race/ethnicity, socioeconomic status, and health care professional (HCP) availability, and associated variation in service utilization and total cost for management of spinal disorders.Design, Setting, and ParticipantsThis retrospective cohort study examines a national sample of 1,534,280 complete episodes of a spinal disorder experienced in 2017-2019 using zip code and episode of care. Risk ratio (RR) and 95% confidence interval (CI), and ridge regression were used to examine associations between independent and dependent measures.ExposuresZip code level measures of population race/ethnicity and socioeconomic status were the primary independent measures.Main Outcomes and MeasuresAvailability of and access to 17 types of HCPs, use of 14 types of health care services, and total cost were the primary dependent measures.Results1,075,204 continuously insured individuals aged 18 years and older from 29,318 zip codes were associated with 1,534,280 episodes of a spinal disorder involving 531,115 HCPs generating $2,022,124,695 in expenditures. Compared to those in primarily white, middle income zip codes, individuals in non-white, disadvantaged zip codes were more likely to initially contact an emergency medicine physician (RR 2.23, 95% CI 2.11-2.36) or primary care provider (RR 1.40, 95% CI 1.38-1.42) and less likely to contact a chiropractor (RR 0.36, 95% CI 0.34-0.37). These individuals had higher exposure to prescription NSAIDs (RR 1.99, 95% CI 1.95-2.02), skeletal muscle relaxants (RR 1.57, 95% CI 1.52-1.59), opioids (RR 1.18, 95% CI 1.15-1.20), and CT scans (RR 1.94, 95% CI 1.84-2.04).Conclusions and RelevanceIndividuals in affluent, primarily non-Hispanic white zip codes have an abundance of options for managing a spinal disorder. Use of first line non-pharmacological and non-interventional options should be reinforced before second- and third-line services are considered. Individuals in low-income, non-white zip codes have less availability of non-pharmacological options, leading to greater use of emergency department and primary care and resulting in pharmaceutical management of spinal disorders. Sustainable models that increase availability of non-pharmaceutical options warrant further study.Key PointsQuestionCompared to the middle-income zip codes with a primarily non-Hispanic white population, do individuals with a spinal disorder living in zip codes characterized by socioeconomic disadvantage and a primarily non-white population have different care experiences?FindingsIndividuals with low back pain living in non-white, low-income zip codes have less availability of chiropractors, physical therapists and licensed acupuncturists and are more likely to seek initial treatment from an emergency department or primary care provider. These same individuals are more likely to receive pharmaceutical management, including prescription opioids, and less likely to receive guideline recommended non-pharmaceutical and non-interventional first line treatments.MeaningThese findings suggest the need to create economically sustainable models that ensure access to guideline concordant non-pharmaceutical treatment options in non-white, low-income communities.