Table 5. Characteristics associated with SLK-listing among SLK-eligible liver candidates.
Odds ratios (ORs) represent likelihood of immediate SLK-listing; OR>1 indicates higher likelihood of immediate SLK listing. Time ratios (TRs) represent ratio of time to SLK-listing if candidates were not listed for SLK immediately; TR>1 indicates longer time and lower likelihood to SLK-listing.
| OR of immediate SLK-listing | P value | TR of later SLK-listing | P value | |
|---|---|---|---|---|
| Patient-level characteristics | ||||
| Age (per 10 year increments) | 0.50 0.55 0.60 | <0.001 | 1.25 1.39 1.56 | <0.001 |
| Male | 1.22 1.41 1.62 | <0.001 | 0.66 0.79 0.94 | 0.007 |
| African American | 1.75 2.23 2.84 | <0.001 | 0.40 0.55 0.75 | <0.001 |
| MELD (per 5 point increments) | 0.77 0.82 0.88 | <0.001 | 0.44 0.47 0.51 | <0.001 |
| eGFR (per 5 ml/min/1.73 increments) | 0.67 0.70 0.73 | <0.001 | 1.21 1.25 1.30 | <0.001 |
| Prior transplant | 1.30 1.68 2.17 | <0.001 | 0.81 1.15 1.63 | 0.4 |
| Diabetes | 1.18 1.35 1.56 | <0.001 | 0.62 0.74 0.89 | 0.001 |
| Hypertension | 1.83 2.13 2.48 | <0.001 | 0.68 0.82 1.01 | 0.06 |
|
| ||||
| Center-level characteristics | ||||
| Percentage of AA liver candidates | 0.90 1.01 1.14 | 0.8 | 0.99 1.13 1.28 | 0.07 |
| Liver transplant volume* | 0.94 0.96 0.97 | <0.001 | 1.02 1.03 1.06 | <0.001 |
| Kidney transplant volume* | 0.99 1.00 1.01 | 0.3 | 0.98 0.99 1.00 | 0.02 |
| SLK transplant volume* | 1.32 1.50 1.70 | <0.001 | 0.63 0.72 0.82 | <0.001 |
| Median MELD at transplant | 0.93 0.97 1.00 | 0.054 | 1.00 1.04 1.09 | 0.04 |
For liver, kidney, and SLK transplant volume, the adjusted OR is per 10 transplants. Liver, kidney, and SLK transplant volumes are cumulative transplant volumes over the study period.
We explored SLK-listing using 2 unique models based on the distribution of our observed data: (1) multilevel logistic regression for immediate listing and (2) multilevel mixed-effects parametric survival regression for listing after eligibility. For the logistic regression model that described immediate SLK-listing among SLK-eligible candidates, we reported an odds ratio for listing. For this population, an odds ratio (OR) of greater than one indicates a higher chance of SLK-listing. For the parametric survival regression model that described SLK-listing after becoming eligible, we reported a time ratio (TR) to listing. A time ratio of greater than one indicates a longer time to SLK-listing and thus, a lower likelihood of listing.