Examining Care Fragmentation After PAD Interventions: The Readmission Event
Olamide Alabi1, Nader Nabile Massarweh2, Xinyan Zheng3, Jialin Mao3, Yazan Duwayri1
1Emory University School of Medicine, Altanta, GA;2Atlanta VA HealthCare System, Altanta, GA;3Cornell University, New York, NY
INTRODUCTION: Lower extremity revascularization (LER) for peripheral artery disease (PAD) is complicated by the frequent need for readmission either at the same facility or a different one than where the index LER was performed. Prior work using Vascular Quality Initiative (VQI) data linked to Medicare claims has demonstrated readmission rates after LER are 18% and 31% at 30-days and 90-days, respectively. It is unclear, however, if readmission at a non-index LER facility compared to the index LER facility is associated with worse outcomes. METHODS: This was a national cohort study of older adults who underwent open, endovascular, or hybrid LER for PAD (January 1, 2010 - December 31, 2018) in the VQI with subsequent readmission within 90-days. This dataset was linked to Medicare claims and the American Hospital Association Annual Survey. LER performed for acute limb ischemia or aneurysm as well as those performed in an office-based setting were excluded. The primary outcome was 90-day mortality and secondary outcomes were major amputation at 30- and 90-days after LER. The primary exposure was the first readmission after LER to the index LER facility versus non-index LER facility. Multivariable logistic regression was used to assess the association between 90-day mortality and readmission location adjusting for standard demographics, pre-procedure symptoms and ambulatory status, procedural type, discharge location, readmission facility-level characteristics, and information regarding access to care. Readmission diagnosis codes were reviewed to determine if the readmission was procedure-related or not. All Medicare designated ‘planned readmissions’ and those discharged from index procedure to hospice (<1% of the cohort) were excluded. RESULTS: Among 13,299 patients who were readmitted within 90 days of LER for PAD, 27.3% of these patients were readmitted to a non-index LER facility. While there were no differences noted in comorbidity burden based on readmission site, compared to patients readmitted to the index LER facility, those readmitted to a non-index facility had a lower proportion of procedure-related reasons for readmission (21.5% vs 50.1%, p<0.001). Most of the patients who were readmitted to a non-index LER facility lived further than 31 miles from the index LER facility (39.2% vs 19.7%, p<0.001) and were readmitted to a facility with a total bed size under 250 (60.3% vs 12.1%, p<0.001). Readmission to a non-index LER facility within 90-days was not associated with 90-day mortality (odds ratio [OR] 1.08 [95% CI, 0.92-1.26]). Readmission to a non-index LER facility was also not associated with increased odds of 30-day or 90-day amputation. However, readmission for a procedure-related complication was associated with major amputation (30-day amputation: OR 3.58 [95% CI, 3.00-4.27]; 90-day amputation: OR 3.33 [95% CI, 2.93-3.80]). CONCLUSIONS:While care fragmentation and readmission to a different facility after LER for PAD is not associated with amputation or death within 90 days, readmission for a procedure-related complication is significantly associated with 30- and 90-day amputation. Quality improvement efforts should focus on understanding the reasons care fragmentation are associated with procedure-related failure to better understand the role that discontinuity of care plays on limb salvage.
Back to 2023 Abstracts