Association of insurance status with timing of hemodialysis access placement

OBJECTIVE Patients without adequate outpatient follow-up often present requiring emergency hemodialysis and then undergo permanent dialysis access placement at a later time. We sought to examine the relationship between type of insurance and whether a patient was already on dialysis at time of surgery. METHODS The Vascular Quality Initiative Hemodialysis Access registry was queried for all adult patients undergoing first time permanent hemodialysis access between January 2015 and September 2019. Patient and procedural characteristics were examined in patients split by private insurance-Medicare more than 65 years of age, Medicare less than 65 years of age, and Medicaid. The primary outcome was whether patients were on dialysis at the time of surgery. RESULTS There were 19,307 adult patients that underwent first time placement of an arteriovenous fistula or graft. Of these patients, 9729 (50%) had Medicare, 7179 (37%) had private insurance, and 2399 (12%) had Medicaid. The patients with Medicare were subgrouped by age with 2968 (31%) being less than 65 years of age and 6761 (69%) being more than 65 years of age. Patients with Medicare and less than 65 were the most likely to be on dialysis at the time of surgical access placement at 67%, whereas 59% of Medicaid patients were on dialysis, and 53% each group of patients with Medicare and more than 65 years of age and private insurance were on dialysis. After adjustment for patient characteristics, patients with Medicare who were less than 65 and more than 65 years of age were both significantly more likely to be on dialysis at time of surgery compared with private insurance with odds ratio (OR) of 1.64 (95% confidence interval [CI], 1.49-1.80; P < .001) and an OR of 1.11 (95% CI, 1.03-1.20; P = .007), respectively. After adjustment, patients with Medicaid were no longer significantly more likely to be on dialysis. Secondary outcomes demonstrated, after adjustment, no difference in the association between a surgical fistula vs graft in any insurance groups; however, patients with Medicare and who were less than 65 years of age were more likely to have a nonradial artery used for anastomosis with an OR of 1.18 (95% CI, 1.04-1.34; P = .011). CONCLUSIONS Certain types of insurance are correlated with being on dialysis at the time of access placement. Although associations were seen between insurance type and surgical access characteristics, these were associations predominantly insignificant when patient demographics and status of dialysis were controlled for. These potential gaps in care represent an area for improvement that deserves further exploration.