Factors Predicting Disposition After Lower Extremity Bypass

Abstract

Introduction: Disposition to post-acute care services (PACS) after lower extremity bypass (LEB) is a measure of resource utilization. Due to comorbidities and impaired ambulation many require PACS to return to independent living. Utilization of home health services (HHS), inpatient rehabilitation facilities (IRF), skilled nursing facilities (SNF), and long-term care hospitals (LTCH) varies by demographic and insurance status. Financial incentives to decrease LOS may increase PACS use and shift costs from acute care hospitals. The objective of this study is to identify predictors of disposition after LEB. Methods: The Nationwide Inpatient Sample 2003-2007 was queried for adult discharges containing diagnoses for claudication or critical limb ischemia and LEB. Patients who died in hospital or were transferred to acute care hospitals were excluded. The outcome was defined as disposition to home, HHS, IRF, or SNF/LTCH. The Elixhauser method was used to adjust for comorbidities. Significant predictors in bivariate polychotomous weighted logistic regression for the dependent multi-level outcome of disposition were entered into a multivariate model. Predictors of LOS were analyzed with weighted logistic regression for the outcome of LOS>5 days. Results: A weighted sample of 374,937 discharges met criteria for analysis. The mean age was 67.8 yrs (±0.1), 40.9% (±0.2) were female, and 55.4% (±1.7) were white. 65.3% (±0.4) were Medicare beneficiaries. Median LOS was 4.9 days (±0.1). 57.4% (±0.7) were discharged home, 19.5% (±0.4) to HHS, 3.7% to IRF (±0.2), and 19.4% (±0.5) to SNF/LTCH. Compared to the privately insured, Medicare beneficiaries had longer LOS (OR 1.74; P<.001) and greater PACS utilization (HHS vs home OR 5.40, P<.001; IRF vs home OR 3.23, P<.001; SNF/LTCH vs home OR 2.17, P<.001). Increased LOS was related to disposition (HHS vs home OR 10.49, P<.001; IRF vs home OR 13.09, P<.001; SNF/LTCH vs home OR 3.94, P<.001). Significant relationships were found in multivariate analyses between demographics, insurance, amputation and disposition. Patients who were older or female, were Medicare beneficiaries, or had severe CLI or amputations had greater PACS utilization (Table). Conclusions: 43% of this nationally representative sample utilized PACS after LEB. As expected, longer LOS, older age, and amputation were significant predictors of PACS utilization. Private insurance is assumed to provide similar to better coverage of PACS than Medicare. However, in multivariate analysis, Medicare beneficiaries had greater utilization. Potential unmeasured confounders (social support, disability, or employment) may explain these findings. Further study of appropriate referral to PACS after LEB is needed. Tabled 1 OR: HHS vs home (95% CI; P value) OR: IRF vs home (95% CI, P value) OR: SNF/LTCH vs home (95% CI, P value) Age ≥ 69 years 1.62 (1.52, 1.74; <.001) 2.34 (2.07, 2.64; <.001) 3.51 (3.22, 3.84; <.001) Female 1.20 (1.14, 1.27; <.001) 1.51 (1.36, 1.67; <.001) 1.61 (1.51, 1.72; <.001) White Reference Black 1.13 (0.99, 1.29; 0.073) 1.36 (1.13, 1.63; <.001) 1.13 (0.99, 1.29; 0.068) Hispanic 1.02 (0.86, 1.21; 0.845) 0.56 (0.40, 0.77; <.001) 0.81 (0.65, 1.01; 0.067) Asian 1.07 (0.71, 1.61; 0.753) 0.75 (0.30, 1.90; 0.547) 0.71 (0.40, 1.26; 0.244) Native American 1.09 (0.66, 1.78; 0.743) 0.98 (0.40, 2.40; 0.968) 1.34 (0.70, 2.56; 0.374) Other race 0.66 (0.52, 0.85; <.001) 1.03 (0.70, 1.52; 0.893) 0.82 (0.63, 1.08; 0.161) Private Insurance Reference Medicare 1.57 (1.45, 1.69; <.001) 2.00 (1.70, 2.35; <.001) 2.74 (2.47, 3.05; <.001) Medicaid 1.18 (1.04, 1.35; 0.013) 0.73 (0.52, 1.03; 0.069) 1.37 (1.13, 1.66; 0.001) No insurance 0.68 (0.55, 0.84; <.001) 0.49 (0.34, 0.69; <.001) 0.49 (0.37, 0.64; <.001) Northeast Reference Midwest 0.43 (0.33, 0.54; <.001) 0.64 (0.46, 0.88; 0.006) 0.41 (0.33, 0.51; <.001) South 0.53 (0.44, 0.64; <.001) 0.49 (0.37, 0.64; <.001) 0.37 (0.30, 0.46; <.001) West 0.26 (0.21, 0.32; <.001) 0.87 (0.61, 1.24; 0.443) 0.36 (0.27, 0.47; <.001) Severe CLI 1.56 (1.46, 1.67; <.001) 1.61 (1.42, 1.83; <.001) 1.91 (1.77, 2.07; <.001) No amputation Reference Minor Amputation 1.01 (1.80, 2.26; <.001) 2.35 (1.95, 2.83; <.001) 2.76 (2.45, 3.10; <.001) Major Amputation 1.40 (1.02, 1.90; 0.034) 21.84 (16.03, 29.75; <.001) 7.14 (5.39, 9.45; <.001) LOS ≥ 5 days 2.79 (2.58, 3.02; <.001) 6.88 (5.95, 7.96; <.001) 5.85 (5.20, 6.58; <.001) Expression levels normalized to beta-actin and scaled by 103 except for FOXP3 which was normalized to T-cell receptor-alpha levels and is not scaled. Expression levels normalized to beta-actin and scaled by 103 except for FOXP3 which was normalized to T-cell receptor-alpha levels and is not scaled. Open table in a new tab Expression levels normalized to beta-actin and scaled by 103 except for FOXP3 which was normalized to T-cell receptor-alpha levels and is not scaled. Expression levels normalized to beta-actin and scaled by 103 except for FOXP3 which was normalized to T-cell receptor-alpha levels and is not scaled.