Abstract
Background/Aims Cutaneous melanoma is among the most rapidly increasing malignancies in the United States, afflicting approximately 77,000 persons and causing over 9,000 deaths annually. Over half of incident melanoma and melanoma-related deaths occurs in patients older than age 65. Barriers to accessing dermatologic care may impede timely surgical therapy for patients with melanoma, potentially resulting in psychological or physical harm. Accordingly, given the disproportionate burden of melanoma borne by the elderly and limited availability of dermatologic care, we sought to determine timeliness of surgical resection of melanoma and identify factors associated with time to surgery in the Medicare population.
Methods We conducted a population-based, retrospective cohort study of melanoma cases from years 2000 to 2009 using linked data from the Surveillance, Epidemiology, and End Results (SEER)-Medicare database. We identified cases of melanoma using ICD-O-3 and ICD-9 codes. Our primary outcome of interest was time to surgery (dichotomized at 21 days), ascertained by dates of ICD-9 and CPT skin biopsy and surgical procedure codes from Medicare administrative claims. Multivariate logistic regression with clustering of cases by patient and fixed effects for hospital referral region was used to evaluate the impact of type of surgery (Mohs versus non- Mohs technique), patient-level factors (age, gender, race, comorbidities, marital status, income, and prior history of melanoma), and tumor-level factors (stage and anatomic location).
Results We identified 32,666 cases of melanoma (1.04 cases per patient) during our study period. The median time to surgery was 27 days (IQR 16–42). Significant predictors of surgical delay (P<0.05) included: Mohs surgery (OR 1.10), Elixhauser comorbidity index of 3 or greater (OR 1.16), unmarried status (OR 1.17), regional disease (OR 1.44), and anatomic localization to trunk (OR 0.67) and extremities (OR 0.74). Age, gender, race, income, and history of prior melanoma were not associated with time to surgery.
Conclusions Time to surgery exhibits substantial variation among the Medicare population, with longer delays associated with select patient-, provider-, and tumor-level risk factors. Future research is needed to better understand the role these factors may play as possible barriers to dermatologic surgical care and their potential impact on patient morbidity and mortality.
