Is Social Deprivation Associated with PROMIS Outcomes After Upper Extremity Fractures in Children?

BACKGROUND: We previously found that social deprivation was associated with worse perceived function and pain among children presenting with upper extremity fractures. We performed the current study to determine whether this differential in outcome scores would resolve after children received orthopaedic treatment for their fractures. This was needed to understand whether acute pain and impaired function were magnified by worse social deprivation or whether social deprivation was associated with differences in health perception even after injury resolution. QUESTIONS/PURPOSES: Comparing patients from the least socially deprived national quartile and those from the most deprived quartile, we asked: (1) Are there differences in age, gender, race, or fracture location among children with upper extremity fractures? (2) After controlling for relevant confounding variables, is worse social deprivation associated with worse self-reported Patient-Reported Outcomes Measurement Information System (PROMIS) scores before and after the treatment of pediatric upper extremity fractures? (3) Is social deprivation associated with PROMIS score improvements as a result of fracture treatment?
METHODS: In this this retrospective, comparative study, we considered data from 1131 pediatric patients (aged 8 to 17 years) treated nonoperatively at a single tertiary academic medical center for isolated upper extremity fractures between June 2016 and June 2017. We used the Area Deprivation Index to define the patient's social deprivation by national quartiles to analyze those in the most- and least-deprived quartiles. After excluding patients with missing zip codes (n = 181), 18% (172 of 950) lived in the most socially deprived national quartile, while 31% (295 of 950) lived in the least socially deprived quartile. Among these 467 patients in the most- and least-deprived quartiles, 28% (129 of 467) were excluded for lack of follow-up and 9% (41 of 467) were excluded for incomplete PROMIS scores. The remaining 297 patients were analyzed (107 most-deprived quartile, 190 least-deprived quartile) longitudinally in the current study; they included 237 from our initial cross-sectional investigation that only considered reported health at presentation (60 patients added and 292 removed from the 529 patients in the original study, based on updated Area Deprivation Index quartiles). The primary outcomes were the self-completed pediatric PROMIS Upper Extremity Function, Pain Interference, and secondarily PROMIS Peer Relationships computer-adaptive tests. In each PROMIS assessment, higher scores indicated more of that domain (such as, higher function scores indicate better function but a higher pain score indicates more pain), and clinically relevant differences were approximately 3 points. Bivariate analysis compared patient age, gender, race, fracture type, and PROMIS scores between the most- and least-deprived groups. A multivariable linear regression analysis was used to determine factors associated with the final PROMIS scores.
RESULTS: Between the two quartiles, the only demographic and injury characteristic difference was race, with Black children being overrepresented in the most-deprived group (most deprived: white 53% [57 of 107], Black 45% [48 of 107], other 2% [2 of 107]; least deprived: white 92% [174 of 190], Black 4% [7 of 190), other 5% [9 of 190]; p < 0.001). At presentation, accounting for patient gender, race, and fracture location, the most socially deprived quartile remained independently associated with the initial PROMIS Upper Extremity (β 5.8 [95% CI 3.2 to 8.4]; p < 0.001) scores. After accounting for patient gender, race, and number of days in care, we found that the social deprivation quartile remained independently associated with the final PROMIS Upper Extremity (β 4.9 [95% CI 2.3 to 7.6]; p < 0.001) and Pain Interference scores (β -4.4 [95% CI -2.3 to -6.6]; p < 0.001). Social deprivation quartile was not associated with any differential in treatment impact on change in PROMIS Upper Extremity function (8 ± 13 versus 8 ± 12; mean difference 0.4 [95% CI -3.4 to 2.6]; p = 0.79) or Pain Interference scores (8 ± 9 versus 6 ± 12; mean difference 1.1 [95% CI -1.4 to 3.5]; p = 0.39) from presentation to the conclusion of treatment.
CONCLUSION: Delivering upper extremity fracture care produces substantial improvement in pain and function that is consistent regardless of a child's degree of social deprivation. However, as social deprivation is associated with worse perceived health at treatment initiation and conclusion, prospective interventional trials are needed to determine how orthopaedic surgeons can act to reduce the health disparities in children associated with social deprivation. As fractures prompt interaction with our health care system, the orthopaedic community may be well placed to identify children who could benefit from enrollment in proven community health initiatives or to advocate for multidisciplinary care coordinators such as social workers in fracture clinics. LEVEL OF EVIDENCE: Level III, therapeutic study.