Background:
Pancreatic cancer outcomes are dismal, in part due to late presentation. Little is known about the barriers to optimal pancreatic cancer diagnosis and treatment in Australia, including the impacts of limited English proficiency (LEP).
Methods:
Clinicopathological data were extracted from the PURPLE Pancreatic Cancer registry for consecutive patients diagnosed at several tertiary Victorian hospitals between 2016 - 2020. LEP was defined by the use of an interpreter during hospital consultation and was obtained from hospital records. Factors were assessed for their impact on survival using Kaplan-Meier analysis and Cox regression.
Results:
Of 400 patients examined, 65 had LEP (16%) (n=65). LEP patients were more likely than English proficient patients (EP) to be female (60.0% vs 42.1%, p=0.008), older (median age 73.0 vs 69.0, p=0.017), to have ECOG performance status (PS) >= 2 (26.2% vs 12.8%, p = 0.006), and to present with pain as the primary symptom (40.0% vs 27.2%, p=0.037). LEP patients had similar rates of metastatic disease compared to EP patients (32.8% vs 29.9%, p=0.644), however those with local disease were less likely to be candidates for resection (27.9% vs 53.4%, p=0.002). LEP patients were ultimately more likely to receive therapy with palliative intent (78.5% vs 63.0%, p=0.016). The median survival of LEP trended lower overall (5.34 vs 7.69 months, p=0.08), most significant among patients who received only supportive care (2.32 vs 2.69 months, p=0.003).
Conclusions:
The survival outcome of patients with LEP is inferior to that of EP patients. Multiple factors are seen to contribute including demographics (older age) and being diagnosed with more advanced stages of disease. More advanced age and poorer PS are likely contributing to lower rates of active treatment. Further research is required to better understand potential delays in diagnosis and opportunities for intervention.