O1044: The Impact of Language Barriers on Post-operative Management of Scoliosis Patients at a Tertiary Care Children's Hospital in Bronx, New York

Background: Language barriers between providers, patients, and families have long been associated with healthcare disparities. Inpatient post-operative care is particularly reliant upon open, clear communication between these groups, and thus is particularly vulnerable to disparities when barriers exist. Post-operative protocols, such as the Rapid Recovery Pathway for posterior spinal fusion of adolescent idiopathic scoliosis (AIS-PSF), have been created to mitigate potential variability in care. However, we do not know if highly-protocolized post-operative management, as seen for AIS-PSF patients, are enough to prevent differences in care from language barriers.

Methods: This retrospective cohort study was conducted among AIS-PSF patients who underwent surgery at a large, urban children’s hospital in the Bronx, New York following the implementation of a Scoliosis Rapid Recovery Pathway in 2017. The independent variable of interest was language of the patient’s family, categorized as English proficiency (EP) or limited-English proficiency (LEP). Patient-families were considered LEP if a non-English language was used for informed consent or if language interpreters were used while inpatient. Outcomes measured include length of stay (LOS), time spent on patient-controlled analgesia (PCA), amount of PCA consumed, number of PCA doses per hour on PCA, number of oral opioids requested per post-PCA day, total morphine milligram equivalents (MME) consumed while inpatient, and pain at six-weeks and six-months post-operatively. Bivariate analyses demonstrated relationships between independent variables, including language, and outcomes, from which multivariable regression models were built.

Results: 160 patients were included in the study cohort (39 LEP, 121 EP). Between LEP and EP patients, there were no significant differences in demographic, pre-, or intra-operative characteristics (Table 1). There were also no significant differences for inpatient outcomes, though non-significant signals demonstrated greater PCA consumption and higher rates of self-administered PCA doses among LEP patients (Table 2). After discharge, LEP patients were more likely to have clinically significant pain at six-weeks (p=0.045) and six-months (p=0.018) follow-up. Furthermore, when accounting for potentially confounding factors, such as pre-operative pain, Cobb angle, and number of vertebrae fused, LEP patients were still more likely to have pain at six-months post-operatively (p=0.032).

Conclusion: At our large, urban children’s hospital, we did not see significant differences in the inpatient post-operative pain management of AIS-PSF patients based on language. While it is possible that language barriers contribute to less adequate post-operative education or stronger reliance upon medications for managing inpatient pain by providers, these were not supported by our analyses. This could suggest that standardized post-operative protocols, such as our hospital’s scoliosis pathway, help ensure more equitable care. LEP patients did, however, experience greater rates of pain at outpatient follow-up. Future research will focus on why this has been observed and how we can further prevent this disparity.