Do Patients of Different Levels of Affluence Receive Different Care for Pediatric Osteosarcomas? One Institution's Experience

Clin Orthop Relat Res. 2024 Oct 30. doi: 10.1097/CORR.0000000000003299. Online ahead of print.

Abstract

Background: The published reports examining socioeconomic factors and their relationship to osteosarcoma presentation and treatment suggest an association between lower socioeconomic status and a worse response to chemotherapy and lower survivorship. However, the driving factors behind these disparities are unclear. The Child Opportunity Index was developed by diversitydatakids.org (https://www.diversitydatakids.org/) in 2014 to cumulatively quantify social determinants of health in an index specifically tailored toward a pediatric population and organized by census tract. The Childhood Opportunity Index can be used to explore the relationship between a patient's socioeconomic background and disparities in osteosarcoma presentation, treatment, and outcomes.

Question/purposes: Are differences in a child's Childhood Opportunity Index score associated with differences in (1) time from symptom onset to first office visit for osteosarcoma, (2) timing of chemotherapy or timing and type of surgical resection, or (3) initial disease severity, development of metastatic disease, or overall survival?

Methods: A retrospective therapeutic study was conducted using data drawn from the institutional records of a large pediatric tertiary cancer center located in the Mid-Atlantic region of the United States from the years 2006 to 2022. Our main site is in an urban setting, with ample access to public transit. Patients were excluded from analysis if they were seeking a second opinion or our institution was not the main point of orthopaedic care (24% [54 of 223]), had incomplete electronic medical records (4% [9 of 223]), resided in an international country (3% [7 of 223]), presented after relapse (3% [7 of 223]), or lacked 2 years of follow-up at our institution (2% [4 of 223]). A total of 113 pediatric patients (children younger than 18 years) met the inclusion criteria. The Child Opportunity Index is a composite index derived from three domains (education, health and environment, and social and economic) and 29 indicators within the domains that serve to capture the cumulative effect of disparities on child well-being. National Childhood Opportunity scores were collected and scored from 1 to 100. Each score represents an equal proportion of the US population of children 18 years of age or younger. A higher number indicates higher levels of socioeconomic opportunity. The overall Childhood Opportunity Index score was then broken down into three groups representative of the child's relative socioeconomic opportunity: lowest tertile for scores < 34, middle tertile for scores between 34 and 66, and highest tertile for scores > 66. Means, ranges, medians, IQRs, and percentages were used to describe the study sample. Data analysis was conducted across the three groups (lowest tertile, middle, and highest), assessing differences in time to presentation, treatment variations, disease severity, and overall survivorship. Chi-square and Fisher exact tests were applied to compare categorical variables. Mann-Whitney U tests compared continuous data. Kaplan-Meier survival analysis, stratified by Childhood Opportunity Index tertile, was performed for a 5-year period to evaluate the development of metastatic disease and overall survivorship. A log-rank test was applied to evaluate statistical significance. Due to the small sample size, we were unable to control potential confounders such as race and insurance. However, the three domains (education, health and environment, and social and economic) encapsulated by the Childhood Opportunity Index data indirectly account for disparities related to race and insurance status.

Results: There was no association between lower levels of socioeconomic opportunity, as expressed by the lack of difference between the Childhood Opportunity Index tertiles for the interval between symptom onset and first office visit (mean ± SD lowest tertile 77 ± 67 days [95% confidence interval (CI) 60 to 94], middle tertile 69 ± 94 days [95% CI 50 to 89], and highest tertile 56 ± 58 days [95% CI 41 to 71]; p = 0.3). Similarly, we found no association between lower levels of socioeconomic opportunity, as expressed by the lack of difference between the Childhood Opportunity Index tertiles and the time elapsed from the first office visit to the first chemotherapy session (lowest tertile 19 ± 12 days [95% CI 12 to 26], middle 19 ± 14 days [95% CI 11 to 26], and highest 15 ± 9.7 days [95% CI 8.4 to 21]; p = 0.31), the time to surgical resection (lowest tertile 99 ± 35 days [95% CI 87 to 111], middle 88 ± 28 days [95% CI 77 to 99], and highest 102 ± 64 days [95% CI 86 to 118]; p = 0.24), or the type of surgical resection (limb-sparing versus amputation: 84% [21 of 25] in lowest tertile, 83% [24 of 29] in the middle tertile, and 81% [48 of 59] in the highest tertile received limb-sparing surgery; p = 0.52). Finally, we found no differences in terms of disease-free survival at 5 years (lowest tertile 27% [95% CI 7.8% to 43%], middle 44% [95% CI 23% to 59%], and highest 56% [95% CI 40% to 67%]; p = 0.22), overall survival (lowest 74% [95% CI 58% to 95%], middle 82% [95% CI 68% to 98%], and highest 64% [95% CI 52% to 78%]; p = 0.27), or in terms of survivorship of the cohort, excluding patients who presented with metastatic disease (lowest 84% [95% CI 68% to 100%], middle 91% [95% CI 80% to 100%], and highest 68% [95% CI 55% to 83%]; p = 0.10).

Conclusion: In our single-center retrospective study of 113 children who presented with osteosarcoma, we did not find an association between a patient's national socioeconomic opportunity and their time to presentation, chemotherapy treatment, time to and type of surgical resection, or disease-free and overall survival. Prior work has shown an association between socioeconomic background and disparities in osteosarcoma treatment. It is possible that these findings will be similar to those from other hospitals and geographic areas, but based on our findings, we believe that proximity to providers, access to public transit, and regional insurance policies may help diminish these disparities. Future multicenter studies are needed to further explore the role that regional variations and the aforementioned factors may play in osteosarcoma treatment to help inform the direction of public policy.

Level of evidence: Level III, therapeutic study.