Predictors of Asylum Approval: Insights from a Decade of Forensic Medical Evaluations at a Student-Run Clinic
摘要
Asylum seekers face significant challenges in substantiating their claims of trauma and persecution. Forensic medical evaluations (FMEs) support these claims and improve case outcomes. This study investigates which demographic, traumatic, and medical factors predict asylum approval within a student-run asylum clinic. We performed a retrospective study of clinic FMEs (2014–2024). Of 289 completed evaluations, 154 cases with known outcomes were analyzed. The outcome variable was asylum granted (1) vs. not granted (0). 29 binary predictors (10 demographic, 10 trauma, 9 medical) were abstracted from affidavits. We conducted univariate and multivariable logistic regressions. Results were compared with national data from Physicians for Human Rights. 57.1% cases were granted asylum, exceeding the national average of 39.3%. In univariate analyses, higher approval odds were observed with lacerations (OR = 9.30), scars (OR = 3.51), burns (OR = 7.66), any psychiatric diagnosis (OR = 2.81), PTSD (OR = 5.40), depression (OR = 4.22), kidnapping (OR = 12.0), torture (OR = 3.56), having a reported income (OR = 3.24), having children (OR = 1.97), and having a physical (OR = 3.11) and psychological (OR = 4.34) FME conducted (p < 0.05). Gait abnormality was inversely associated (OR = 0.30). In the multivariable model, PTSD (aOR = 6.67), kidnapping (aOR = 12.5), lacerations (aOR = 5.97), and income (aOR = 2.86) remained associated with approval. A decade of data from a student-run asylum clinic confirmed the importance of FMEs. Findings suggest the medical content of FMEs, especially clear, well-documented injuries that match the client’s narrative and select trauma exposures (e.g., kidnapping), carries strong evidentiary weight, while demographic indicators (income) may also play a role. Results support maintaining capacity for both physical and psychological FMEs and funding for student-run clinics. Larger multi-site studies are needed to validate.