Health Disparity Indices in the Assessment of Adolescent Idiopathic Scoliosis: Are All Indices Similar?

Background: Adolescent idiopathic scoliosis (AIS) is a common spinal deformity. Previous studies have demonstrated that healthcare disparities affect the initial presentation of AIS patients. However, studies have utilized various socioeconomic status (SES) indices with no previous study assessing the congruency of these indices. Methods: A single-center, retrospective review of adolescent patients presenting for initial evaluation of scoliosis was performed at a tertiary care pediatric hospital based in the central United States. Patient demographic data were recorded in addition to radiographic data consisting of coronal curve magnitude at presentation and skeletal maturity indices. Socioeconomic variables were collected, including geographic area of residence (urban vs rural), distance from the treating hospital, and SES indices using the Childhood Opportunity Index (COI), Index of Concentration at the Extremes (ICE), and Area Deprivation Index (ADI) based on home address. SES indices were based on national standards and were subdivided into quintiles. Socioeconomic status (SES) disparity was defined as the two worst quintile scores. Socioeconomic status indices were compared for correlation, as well as significant differences in the definition of SES disparity. Results: A total of 368 patients presenting for initial evaluation of scoliosis were identified (mean 15.0 ±1.6 years, 75.8% female). The mean curve magnitude at presentation was 25.9±12.3° (range 10-81.0°). Using the three SES indices, disparity was present in between 5% and 36% of patients. There were significant differences (P < .001) in the definition of SES disparity between the indices: ADI 36.4% (N = 134), COI 19.8% (N = 73), and ICE 5.4% (N = 20). Overall, there was a 52.2% rate of loss to follow-up (N = 192), which was significantly associated with disparity as defined by COI (23.9% vs 15.3%, P = .038) but not ADI (33.1% vs 39.6%, P = .2) or ICE (6.8% vs 3.9%, P = .2). Conclusions: Significant differences exist in the definition of SES disparity using the COI, ICE, and ADI indices. Although these indices are all significantly correlated, ICE produces a more conservative estimate of disparity, whereas COI is more responsive and may be more advantageous in identifying patients at risk for SES disparity as it pertains to the management of spinal deformity. Key Concepts: (1)Previous studies have shown that healthcare disparities influence how AIS patients initially present. However, these studies have used various socioeconomic status (SES) indices, and no previous research has evaluated the consistency of these measures indices.(2)Socioeconomic status indices may not be predictive of coronal curve magnitude at initial presentation.(3)The increased Childhood Opportunity Index disparity was correlated with higher rates of loss to follow-up.(4)ICE produces a more conservative estimate of disparity. In contrast, COI and ADI are more responsive. They may be more advantageous in identifying patients at risk for SES disparity as it pertains to the management of spinal deformity, with COI providing a less extreme assessment for SES disparity.(5)Future studies should aim to determine how SES indices can help prospectively identify patients at risk for loss to follow-up, facilitating targeted resource allocation for these individuals. Level of Evidence: III, Retrospective Review and Analysis.