[PPT] - Metrics Subcommittee Data Presenter: John R. Lake, MD 1 Request PowerPoint Presentation

SLIDE 1

Metrics Subcommittee Data Request Results: Supply/ Demand Ratios and Proximity Points Modeling

Presenter: John R. Lake, MD1

Authors: David Schladt, MS2 Joshua Pyke, PhD2 Sommer Gentry, PhD3 Ajay K. Israni, MD2 Bertram L. Kasiske, MD2 John R. Lake, MD1 Susan N. Leppke, MPH2 Jessica Zeglin, MPH2

1. University of Minnesota; SRTR Senior Staff
2. Scientific Registry of Transplant Recipients
3. United States Naval Academy; SRTR Senior Staff

SLIDE 2

2

Liver Redesign Timeline (to date)

Dec 2011: Initial committee discussions. Broader sharing alone not sufficient to decrease disparity

in access to transplant (actually worsens disparity)

Nov 2012: Problem definition. OPTN board resolution on disparity, alternatives discussed by

committee

Jan 2013: Redistricting selected as conceptual approach to decrease disparity
Mar 2013: Metrics and constraints chosen by committee
Dec 2013: Optimal map designs developed based on supply/demand information, possible results

modeled using LSAM

Mar 2014: Additional analysis of possible impact on transport and cost, analysis of relationship of

redistricting to OPO performance

Sept 2014: First Liver Forum to discuss redistricting concepts and related issues
Dec 2014: Ad Hoc Subcommittees form to discuss Metrics, Finance, and Logistics of possible

redistricting

Feb 27, 2015: Data request from Liver & Intestinal Committee to examine additional approaches to

redistricting

May 29, 2015: Draft report submitted to OPTN on above data request
June 3-4, 2015: Draft report information presented to Liver & Intestinal Committee and Ad Hoc

Subcommittees

June 15, 2015: Final report submitted to OPTN on above data request
June 22, 2015: Second Liver Forum to discuss redistricting concepts and related issues

SLIDE 3

3

Optimized Districts

Optimized districts are based on 3 things:

1.

The number of donors recovered in each DSA

2.

The number of candidates in each DSA

3.

District constraints as determined by the Committee

Optimized districts match (1) the number of donor organs

recovered in each DSA with (2) the number and MELD score of candidates listed in each DSA, with district features meeting (3) the committee constraints.

SLIDE 4

4

Constraints for Optimized Districts

The Liver and Intestinal Organ Transplantation Committee set

specific constraints for optimized districts:

Contiguous districts
Follow existing DSA boundaries
Develop options for between 4 and 8 districts
Minimum of 6 transplant centers within every district
No increase of waiting list deaths
The maximum median transplant-volume-weighted

transport time between DSAs is 3 hours.

SLIDE 5

5

Data for Optimized Districts

Optimization used data only on counts of donors and candidates:
the number of donors per DSA, and
the number of new liver transplant candidates at various MELDs

(supply/demand per DSA)

Districts under consideration were designed using 2010 data
These counts are stable: maps designed using 2006 data also

significantly reduced geographic disparity and reduced waiting list and total deaths when applied to 2010 simulations (Gentry et al., AJT 2013)

These counts seem to be stable even since Share-35
LSAM was not used to generate the districts

SLIDE 6

6

Optimal 4 District Map

SLIDE 7

7

Optimal 8 District Map

SLIDE 8

8

Supply / Demand Ratio Analysis

SLIDE 9

9

Committee Request

Committee requested that SRTR provide an analysis of

supply/demand ratios nationwide and by DSA for the current 11 regions, 4 district, and 8 district maps

Requested supply and demand metrics included:
Supply
Actual liver donors
Eligible deaths
Total deaths
Demand
Total waitlisted candidates
Total waitlisted candidates with MELD/PELD > 15

SLIDE 10

10

Note

Supply/demand ratios will not vary depending on organ

allocation or distribution policies. Modeling of allocation/distribution will not affect these metrics.

Supply metrics are based on deaths in an area
Demand metrics are based on waitlisted candidates in an

area

This analysis uses only observed data.

SLIDE 11

11

Study Population

US population data for 2013 from US Census Bureau and OPTN datasets

(2013 was the most recent data available from the US Census Bureau)

Population for supply metrics:
2,571,164 total deaths
6,893 eligible deaths
6,225 actual liver donors
Population for demand metrics:
25,200 waitlisted candidates
16,747 waitlisted candidates with allocation MELD/PELD > 15
Population includes all active waitlisted candidates in 2013 including

children, adults, and candidates listed at all allocation MELD/PELD scores and at status 1A and status 1B.

52 DSAs had active liver transplant programs during 2013

SLIDE 12

12

Analytic Approach

Specifications for variables include:
Actual liver donors
Deceased liver donors, including donors whose livers were recovered

for transplant but not transplanted. Sourced from OPTN data.

Eligible deaths
Deceased individuals meeting eligibility criteria, i.e., their donor

referral classification is “Eligible”. Sourced from OPTN data, as reported by OPOs.

Total deaths
Total deaths in a geographic region. Sourced from US Census Bureau

data.

Using the population and specifications described above, we

described supply/demand metrics for the US in 2013.

SLIDE 13

13

Results: National Supply/ Demand ratios

Across the country in 2013 there were:
For all active waitlisted liver candidates:
0.25 actual liver donors per 1 waitlisted candidate
0.27 eligible deaths per 1 waitlisted candidate
102.03 deaths per 1 waitlisted candidate
For all active waitlisted liver candidates with

allocation MELD/PELD > 15:

0.37 actual liver donors per 1 waitlisted candidate
0.41 eligible deaths per 1 waitlisted candidate
153.53 deaths per 1 waitlisted candidate

SLIDE 14

14

Results: Supply/ Demand Metric Correlation

Metric Type Metric 1 Metric 2 R2* Supply Eligible deaths Actual deceased donors 0.97 Supply Total deaths Actual deceased donors 0.80 Supply Eligible deaths Total deaths 0.82 Demand Total waitlisted candidates Waitlisted candidates with MELD/PELD > 15 0.98 Supply/demand ratio Eligible deaths/waitlisted candidates > 15 Actual donors/waitlisted candidates > 15 0.99 Supply/demand ratio Total deaths/waitlisted candidates > 15 Actual donors/waitlisted candidates > 15 0.90 Supply/demand ratio Eligible deaths/waitlisted candidates > 15 Total deaths/waitlisted candidates > 15 0.88

*R2 is the coefficient of determination. This number estimates how well data fit a statistical model. R2 ranges from 0 to 1, where 0 indicates no correlation between the elements in the model, and 1 indicates complete correlation.

SLIDE 15

15

E ligible deaths / WL candidates ME LD/ PE LD > 15

SLIDE 16

16

E ligible deaths / WL candidates ME LD/ PE LD > 15

52 DSAs 11 Regions 4 Districts 8 Districts

SLIDE 17

17

Total deaths / WL candidates ME LD/ PE LD > 15

SLIDE 18

18

Total deaths / WL candidates ME LD/ PE LD > 15

52 DSAs 11 Regions 8 Districts 4 Districts

SLIDE 19

19

Supply / Demand Ratio Summary

Analysis of supply/demand ratios indicates that liver transplant supply and

demand vary widely across the country by DSA.

In 52 DSAs with active liver transplant programs, the ratio of eligible

deaths/waitlisted candidates with allocation MELD/PELD > 15 ranged from 0.16 to 4.08.

The grouping of DSAs affects supply/demand ratios across regions
In current 11 regions, the ratio of eligible deaths/waitlisted candidates

with M/P > 15 ranged from 0.24 to 0.62, a 2.5-fold difference.

In conceptualized 8 districts, this range decreased to

0.37 to 0.51, a 1.4-fold difference.

In conceptualized 4 districts, this range further decreased to 0.37 to

0.43, 1.16-fold difference.

This same pattern was seen across many different variations of

supply/demand metrics, including for waitlisted patients with allocation MELD/PELD > 24 or lab MELD/PELD > 15.

SLIDE 20

20

Proximity Points Introduction and Methods

SLIDE 21

21

Committee Request

Committee requested Liver Simulated Allocation Model

(LSAM) analysis of proximity points within circles surrounding donor hospitals for current 11 regions, 4 districts, and 8 districts.

SLIDE 22

22

Proximity Point Parameters

Committee requested the use of three sets of proximity point parameters:

Proximity circle radii of 150 or 250 miles from the donor hospital
Awarding 3 or 5 additional MELD/PELD points to candidates

within the proximity circle

Including candidates outside of the donor’s district but inside

the proximity circle in the first level of regional sharing (“out- district”) or limiting the first level of sharing to candidates within the district (“in-district”) (Proximity points were awarded to all candidates in the proximity circle in both cases.)

SLIDE 23

23

Allocation Ordering for Simulations

The following allocation orderings were used for all simulations except the current policy

scenario.

Adult donors:

1.

District status 1A

2.

District status 1B

3.

District MELD ≥ 15

4.

National status 1A

5.

National status 1B

6.

National M/P ≥ 15

7.

District M/P < 15

8.

National M/P < 15

Child donors (age 0-10):

1.

District pediatric status 1A

2.

National pediatric status 1A (0-11)

3.

District adult status 1A

4.

District pediatric status 1B

5.

District any PELD

6.

District MELD >=15, age 12 to 17

7.

District MELD >= 15 , age 18+

8.

District MELD < 15, age 12 to 17

9.

District MELD < 15, age 18+

10.

National status 1A, age 12-17

11.

National status 1A, age 18+

12.

National status 1B, age 0-17

13.

National PELD

14.

National MELD, age 12-17

15.

National MELD, age 18+

Adolescent donors (age 11-17):

1.

District pediatric status 1A

2.

District adult status 1A

3.

District pediatric status 1B

4.

District any PELD

5.

District MELD >= 15, age 12-17

6.

District MELD >= 15, age 18+

7.

District MELD < 15, age 12-17

8.

District - MELD < 15, age 18+

9.

National pediatric status 1A

10.

National adult status 1A

11.

National pediatric status 1B

12.

National any PELD

13.

National any MELD, age 12-17

14.

National any MELD, age 18+

SLIDE 24

24

Proximity Point Parameters: In/ Out District

Rectangle: region/district
Circle: proximity radius
X: donor center
A-D: transplant centers

Allocation groupings: 1. A + C (A has points) 2. B + D (B has points) Allocation groupings: 1. A + B + C (A, B have points) 2. D

1. Layout
2. In District
3. Out District

SLIDE 25

25

Study Population

Analysis based on actual patient data including:
Transplant candidates listed on liver waiting lists as of December

31, 2006

Candidates added to those waiting lists between January 1,

2007 and December 31, 2011

Organs donated between January 1, 2007 and December 31,

2011

Software combines these actual patient data into independent

donor and candidate populations used in each of the multiple LSAM iterations involved in simulating each allocation scenario

51 DSAs had active liver transplant programs during the sampled

time period (2007-2011)

SLIDE 26

26

Analytic Approach

To assess the impact of proximity points within circles surrounding

donor hospitals, we simulated 28 different allocation scenarios with LSAM and compared the results.

Each simulation was repeated 10 times to provide an estimate of

variability.

Each of the 10 iterations for each scenario used independent sets of
rgan and waitlist arrivals and distinct random number seeds.
Each scenario simulated 5 years of transplants.
Allocation MELD/PELD scores were capped at 40 for all simulations.
Results reported include a mean estimate and minimum to

maximum values of estimates across all 10 scenarios

SLIDE 27

27

Outcome Metrics – Disparity and Summative

Disparity Metrics:
Variance of median allocation MELD/PELD at transplant across DSAs
Variance in waitlist mortality rates across DSAs
Variance in transplant rates across DSAs
Variance in overall mortality rates for candidates with MELD/PELD of ≥ 20

across DSAs

Summative Metrics:
Pretransplant deaths prevented (including waitlist and removal deaths)
Posttransplant deaths prevented
Overall mortality counts and rates
Waitlist mortality counts and rates stratified by MELD/PELD ≥ 35 (including

status 1A and 1B), 29-34, 15-28, < 15

MELD/PELD score at transplant by DSA

SLIDE 28

28

Outcome Metrics – Cost and Transport

Transport Metrics:
Percentage of transplants that are local (within DSA)
Percentage of transplants that are regional (within district)
Median transport time (hours)
Median transport distance (miles)
Percentage of organs flying
Cost Metrics:
Estimated average transportation cost per transplant
Cost and Transport results will be reviewed during the Logistics and

Transportation Ad Hoc Subcommittee and the Finance Ad Hoc Subcommittee presentations

SLIDE 29

29

Proximity Points Results: Disparity Metrics

SLIDE 30

30

Figure Conventions

Each figure displays the values for a given metric across the 28 tested scenarios.
Each scenario was simulated 10 times. The vertical lines on the plot display the

range of variability extending from the minimum to maximum value for that metric.

The point along the line marks the mean value of the metric across the 10

simulated iterations for each of the scenarios.

The left most scenario in each plot is the simulation of current policy. To the right
f this are simulations with broader sharing in the current 11 regions, then in 4

districts, and finally in 8 districts.

Baseline scenarios with no proximity points are marked by a circular point.
Scenarios with 3 proximity points are marked by a square point.
Scenarios with 5 proximity points are marked by a triangular point.
Scenarios with 150 mile circles are marked by a solid point.
Scenarios with 250 mile circles are marked by a hollow point.

SLIDE 31

31

Disparity: Variance in median ME LD/ PE LD at transplant

SLIDE 32

Median ME LD/ PE LD at Transplant

current policy 11R 3P 150Mi In 4D 3P 150Mi In 8D 3P 150Mi In

SLIDE 33

33

Disparity: Variance in pretransplant mortality rates

SLIDE 34

Pretransplant Mortality Rate

current policy 11R 3P 150Mi In 8D 3P 150Mi In 4D 3P 150Mi In

SLIDE 35

35

Disparity: Variance in transplant rates

SLIDE 36

Transplant Rate

current policy 11R 3P 150Mi In 8D 3P 150Mi In 4D 3P 150Mi In

SLIDE 37

37

Disparity: Variance in overall* mortality rates

nce candidates reach a ME

LD/ PE LD of 20

* Includes both pre- and posttransplant mortality

SLIDE 38

Overall Mortality Rate

current policy 11R 3P 150Mi In 8D 3P 150Mi In 4D 3P 150Mi In

SLIDE 39

Proximity Points Results: Summative Metrics

SLIDE 40

40

Summative: Pretransplant deaths prevented per year

SLIDE 41

41

Summative: Posttransplant deaths prevented per year

SLIDE 42

42

Summative: Overall mortality rates

SLIDE 43

43

Summative: Waitlist mortality rates for

MELD/PELD ≥ 35

SLIDE 44

44

Summative: Waitlist mortality rates for ME LD/ PE LD 29-34

Very little estimated difference between scenarios is seen in this figure. Waitlist mortality rates for M/P 15-28 and < 15 show a similar pattern, with no appreciable differences between scenarios.

SLIDE 45

45

Additional metrics: Transplant Rates

SLIDE 46

46

Subgroup Analyses: E ffects on Vulnerable Populations

Pediatric patients have similar outcomes across the simulated

scenarios in most cases. Exceptions:

Transplant rates increase in 4- and 8-district in-district scenarios
Transport distances and times increase in 4- and 8-district

scenarios

Racial/ethnic subgroups have similar outcomes across the simulated

scenarios in most cases. Exceptions:

Transplant rates decreased the most for Caucasian patients in

redistricted scenarios

Transplant rates decrease very slightly for African-American

patients in redistricted scenarios

Transplant rates increase for Hispanic and Asian patients in

redistricted scenarios

SLIDE 47

47

Proximity Points Summary

Across the wide range of scenarios and metrics studied, a few overall trends stood out:

The number of distribution units (11 regions, 4 or 8 districts) and the sharing

configuration for proximity circles (in-district or out-district) had the largest effects on variation between scenarios.

Out-district proximity circles were similar to each other across 11 region, 4-district, and

8-district scenarios. This may indicate that out-district circles perform less like regional allocation and more like concentric circle allocation.

Largest decreases in disparity metrics were seen among 4-district baseline and in-

district circle scenarios, followed by 8-district baseline and in-district circle scenarios. Out-district circle scenarios decreased disparity metrics compared to current policy, but not as much as 4- and 8-district baseline and in-district scenarios.

Very little variation in mortality rates was observed across any of the scenarios, with the

exception of the MELD ≥ 35 group where waitlist mortality decreased slightly in the 4- and 8-district scenarios.

Adding proximity circles to 4- and 8-district scenarios increased the local transplant

percentage while decreasing transport times, distances, and flying percentages (shown in transport presentation)

SLIDE 48

Q & A

SLIDE 49

49

Supplemental Slides: Data Request 1.1

SLIDE 50

50

Actual donors/ WL candidates ME LD/ PE LD > 15

52 DSAs 11 Regions 8 Districts 4 Districts

SLIDE 51

51

E ligible deaths / WL candidates lab ME LD/ PE LD > 15

52 DSAs 11 Regions 8 Districts 4 Districts

SLIDE 52

52

E ligible deaths / WL candidates ME LD/ PE LD > 24

52 DSAs 11 Regions 8 Districts 4 Districts

SLIDE 53

53

Supplemental Slides: Data Request 1.2

SLIDE 54

54

Simulated Scenarios

Run Number Distribution Number

f Points

Radius Candidate Designation 1 11 Regions None None None 2 4 Districts None None None 3 8 Districts None None None 4 11 Regions 3 150 In district 5 11 Regions 3 150 Out of district 6 11 Regions 3 250 In district 7 11 Regions 3 250 Out of district 8 11 Regions 5 150 In district 9 11 Regions 5 150 Out of district 10 11 Regions 5 250 In district 11 11 Regions 5 250 Out of district 12 4 Districts 3 150 In district 13 4 Districts 3 150 Out of district 14 4 Districts 3 250 In district Run Number Distribution Number

f Points

Radius Candidate Designation 15 4 Districts 3 250 Out of district 16 4 Districts 5 150 In district 17 4 Districts 5 150 Out of district 18 4 Districts 5 250 In district 19 4 Districts 5 250 Out of district 20 8 Districts 3 150 In district 21 8 Districts 3 150 Out of district 22 8 Districts 3 250 In district 23 8 Districts 3 250 Out of district 24 8 Districts 5 150 In district 25 8 Districts 5 150 Out of district 26 8 Districts 5 250 In district 27 8 Districts 5 250 Out of district 28 Current system None None None

SLIDE 55

55

LSAM Strengths and Limitations

Strengths

Draws on real transplant

data

Simulates up to 5 years
Multivariable acceptance

and survival models

Can compare multiple

allocation and distribution systems

Limitations

Predicts direction of change

between alternatives, not necessarily the magnitude

f change
Cannot account for changes

in listing or acceptance behavior

Cannot predict outcomes on

a center-by-center basis

Most recent input data files

use data through 2011

SLIDE 56

56

Summative: Overall mortality counts per year

SLIDE 57

57

Summative: Annual waitlist mortality counts for ME LD/ PE LD ≥ 35

SLIDE 58

58

Summative: Annual waitlist mortality counts for ME LD/ PE LD 29-34

SLIDE 59

59

Summative: Annual waitlist mortality counts for ME LD/ PE LD <15

SLIDE 60

60

Summative: Waitlist mortality rates for ME LD/ PE LD 15-28

SLIDE 61

61

Summative: Waitlist mortality rates for ME LD/ PE LD <15

SLIDE 62

62

Additional Metrics: Transplant Counts

SLIDE 63

63

Additional Metrics: Pretransplant Mortality Counts

SLIDE 64

64

Additional Metrics: Pretransplant Mortality Rates

SLIDE 65

65

Disparity Metrics Table

Variance in median MELD at transplant Variance in pretransplant mortality rates Variance in transplant rates Variance in mortality rates after MELD 20 Current 6.2 (5.5-6.9) 0.00058 (0.00043-0.00073) 0.116 (0.105-0.128) 0.00035 (0.00024-0.00052) 11 Regions 8.2 (7.4-9.1) 0.00049 (0.00043-0.00058) 0.1 (0.09-0.111) 0.00031 (0.00021-0.0004) 11R 3P 150Mi In 8.3 (7.5-9.7) 0.00052 (0.0004-0.00063) 0.106 (0.09-0.13) 0.00032 (0.00019-0.00059) 11R 3P 250Mi In 8.3 (7.2-9.3) 0.00053 (0.0004-0.00068) 0.108 (0.092-0.124) 0.00034 (0.00022-0.00058) 11R 5P 150Mi In 7.7 (6.9-8.6) 0.00055 (0.00046-0.00063) 0.107 (0.091-0.128) 0.00034 (0.00024-0.00074) 11R 5P 250Mi In 7.9 (6.8-8.9) 0.00051 (0.00042-0.00067) 0.113 (0.1-0.126) 0.00038 (0.00025-0.0007) 11R 3P 150Mi Out 4.2 (3.8-4.6) 0.00061 (0.00048-0.0007) 0.062 (0.051-0.071) 0.00031 (0.0002-0.00041) 11R 3P 250Mi Out 3.8 (3.5-4.7) 0.00061 (0.00048-0.0007) 0.062 (0.053-0.069) 0.0003 (0.00016-0.00056) 11R 5P 150Mi Out 4.1 (3.6-4.6) 0.00061 (0.0005-0.00075) 0.062 (0.052-0.068) 0.0003 (0.00021-0.00043) 11R 5P 250Mi Out 3.8 (3.3-4.6) 0.00059 (0.00045-0.0007) 0.061 (0.051-0.07) 0.00031 (0.00021-0.00044) 4 Districts 2.1 (1.6-2.7) 0.0004 (0.00031-0.00047) 0.024 (0.021-0.029) 0.00023 (0.00018-0.00029) 4D 3P 150Mi In 2.1 (1.6-2.7) 0.00042 (0.00031-0.0005) 0.027 (0.023-0.032) 0.00025 (0.00014-0.00032) 4D 3P 250Mi In 2.1 (1.5-2.5) 0.00043 (0.00034-0.00056) 0.029 (0.025-0.035) 0.00024 (0.00016-0.00048) 4D 5P 150Mi In 2.3 (1.8-2.8) 0.00046 (0.00039-0.00057) 0.031 (0.026-0.036) 0.00026 (0.00014-0.00047) 4D 5P 250Mi In 2.4 (1.9-3.2) 0.00047 (0.00039-0.00057) 0.032 (0.027-0.038) 0.00026 (0.00014-0.0004) 4D 3P 150Mi Out 4 (3.6-4.6) 0.0006 (0.00045-0.00078) 0.06 (0.046-0.072) 0.00031 (0.00022-0.00045) 4D 3P 250Mi Out 3.7 (3.1-4.5) 0.00057 (0.00049-0.00069) 0.063 (0.054-0.076) 0.00033 (0.00019-0.00066) 4D 5P 150Mi Out 4.1 (3.4-4.7) 0.00061 (0.00047-0.00074) 0.062 (0.052-0.07) 0.00027 (0.00021-0.00038) 4D 5P 250Mi Out 3.7 (3.2-4.2) 0.00058 (0.00049-0.00067) 0.06 (0.053-0.069) 0.00027 (0.00018-0.00044) 8 Districts 2.8 (2.1-3.5) 0.00047 (0.00036-0.00058) 0.027 (0.021-0.031) 0.00029 (0.00018-0.0004) 8D 3P 150Mi In 2.9 (2-3.6) 0.00046 (0.00036-0.00057) 0.029 (0.025-0.033) 0.00026 (0.00019-0.00032) 8D 3P 250Mi In 3 (2.1-3.6) 0.00046 (0.00036-0.00054) 0.03 (0.025-0.034) 0.00026 (0.00015-0.00051) 8D 5P 150Mi In 3 (2.2-3.8) 0.00048 (0.0004-0.00059) 0.032 (0.025-0.04) 0.0003 (0.00016-0.00039) 8D 5P 250Mi In 3 (2.1-3.9) 0.00051 (0.00041-0.00067) 0.033 (0.028-0.036) 0.00031 (0.00023-0.00062) 8D 3P 150Mi Out 4.1 (3.6-5) 0.00061 (0.0005-0.00074) 0.059 (0.049-0.067) 0.0003 (0.00023-0.00051) 8D 3P 250Mi Out 3.8 (3.2-4.9) 0.00058 (0.00046-0.00084) 0.065 (0.055-0.075) 0.00028 (0.00022-0.00039) 8D 5P 150Mi Out 4.1 (3.6-4.8) 0.00062 (0.00046-0.00074) 0.061 (0.047-0.072) 0.00028 (0.00019-0.00044) 8D 5P 250Mi Out 3.6 (3-4.5) 0.00058 (0.00045-0.00069) 0.061 (0.054-0.069) 0.0003 (0.00016-0.00078)

SLIDE 66

66

Summative Metrics Table

Pretransplant deaths per year prevented Posttransplant deaths per year prevented Overall deaths per year Overall death rates per patient-year 11 Regions

4.8 (-27.8-12.6)

4.7 (-18.4-32.6) 3608.1 (3567.8-3658.4) 0.1 (0.1-0.1) 11R 3P 150Mi In 4.9 (-12.4-26) 0.6 (-25.2-32.4) 3608.2 (3565.6-3652.8) 0.1 (0.1-0.1) 11R 3P 150Mi Out 31.9 (17.2-48.2)

53.3 (-79.8--40.8)

3602.6 (3558.4-3652.8) 0.1 (0.1-0.1) 11R 3P 250Mi In 5.1 (-13.2-24)

0.1 (-33.4-20)

3603.1 (3571.8-3637.2) 0.1 (0.1-0.1) 11R 3P 250Mi Out 17.9 (-1.2-27.2)

39.2 (-63.4--1.8)

3601.1 (3563.4-3648.8) 0.1 (0.1-0.1) 11R 5P 150Mi In 4.1 (-12-19.2) 2.9 (-21.2-21) 3613.5 (3578.6-3644) 0.1 (0.1-0.1) 11R 5P 150Mi Out 32.7 (21-53.6)

49.7 (-75--32.2)

3629.5 (3600-3672.8) 0.1 (0.1-0.1) 11R 5P 250Mi In

4.3 (-26.8-15.2)
1.1 (-20.8-13)

3629.4 (3581.2-3680.4) 0.1 (0.1-0.1) 11R 5P 250Mi Out 19.6 (6.8-32.2)

37.3 (-66.6--8.8)

3625.1 (3592.4-3682.8) 0.1 (0.1-0.1) 4 Districts 118.8 (100.4-134.6) 1.6 (-42.6-29) 3625.8 (3595-3656.4) 0.1 (0.1-0.1) 4D 3P 150Mi In 116.4 (97.2-133.6)

11.5 (-41.8-23.4)

3487.8 (3468.2-3507.4) 0.1 (0.1-0.1) 4D 3P 150Mi Out 39.7 (18.8-59.6)

33 (-54--17)

3503.2 (3471.2-3537.2) 0.1 (0.1-0.1) 4D 3P 250Mi In 113.7 (97-132)

13.3 (-38.8-6.6)

3507.7 (3469.4-3540.6) 0.1 (0.1-0.1) 4D 3P 250Mi Out 17.7 (-4-34.8)

14.2 (-38.8-11.4)

3508.2 (3483.8-3561.4) 0.1 (0.1-0.1) 4D 5P 150Mi In 103.1 (81.6-117.8)

3.2 (-19.6-14.4)

3520.6 (3482.8-3554.4) 0.1 (0.1-0.1) 4D 5P 150Mi Out 42.4 (27.8-59.2)

44.9 (-61.2--29.8)

3601.3 (3575.6-3637.6) 0.1 (0.1-0.1) 4D 5P 250Mi In 97.9 (88-111)

10.3 (-30.4-5.8)

3604.6 (3562.2-3654.8) 0.1 (0.1-0.1) 4D 5P 250Mi Out 16.2 (-1.8-33)

6.3 (-24.6-10.2)

3610.6 (3582.2-3643.2) 0.1 (0.1-0.1) 8 Districts 59.8 (39.8-86.6) 1.2 (-17.8-27.6) 3598.2 (3565-3649.4) 0.1 (0.1-0.1) 8D 3P 150Mi In 55.4 (35-74.2)

3.3 (-22.4-18.4)

3547 (3523.6-3567.2) 0.1 (0.1-0.1) 8D 3P 150Mi Out 38.3 (15.8-54.4)

46.3 (-75.6--29.4)

3556 (3525.4-3589.6) 0.1 (0.1-0.1) 8D 3P 250Mi In 51.6 (30.2-63.8) 5.4 (-19-25.4) 3551.1 (3528-3595.2) 0.1 (0.1-0.1) 8D 3P 250Mi Out 21.9 (-1.8-43.6)

28.7 (-52.6-4.2)

3562.9 (3538.2-3604.4) 0.1 (0.1-0.1) 8D 5P 150Mi In 53.5 (36.2-69.2)

8.3 (-23.6-7)

3558.2 (3521.8-3599) 0.1 (0.1-0.1) 8D 5P 150Mi Out 37.3 (9.8-60.4)

48.1 (-78.8--33.8)

3616.1 (3585.4-3668.6) 0.1 (0.1-0.1) 8D 5P 250Mi In 50.6 (35.8-68.8)

0.7 (-30-27.8)

3614.9 (3590-3645.2) 0.1 (0.1-0.1) 8D 5P 250Mi Out 17.1 (2.2-30)

23.7 (-38.2--12.8)

3618.9 (3590.8-3684.4) 0.1 (0.1-0.1) Current 0 (0-0) 0 (0-0) 3614.7 (3596.6-3655.6) 0.1 (0.1-0.1)

SLIDE 67

67

Mortality Counts by ME LD Category Table

≥35 29-34 15-28 <15 Current 93.8 (85-100.4) 38.6 (34.6-46.4) 267.8 (257.4-278) 279.7 (270.8-284.8) 11 Regions 89.4 (83-95.2) 33.1 (29.8-38.8) 275.9 (267.6-288.6) 283.3 (278.2-288.6) 11R 3P 150Mi In 88.7 (80.4-98.4) 34.8 (31.6-41.2) 274.1 (263.8-285.6) 283.3 (276.2-289) 11R 3P 250Mi In 89 (84.6-94.2) 32.9 (28.2-38.8) 275.1 (263.4-292) 282.5 (275.4-290) 11R 5P 150Mi In 90.3 (81-98) 33.8 (30.6-38.2) 273.8 (267-288.6) 281.8 (274-287.6) 11R 5P 250Mi In 91.1 (84.6-98.2) 33.9 (31-40) 274.5 (261.4-290) 282.8 (274.8-289.2) 11R 3P 150Mi Out 79.7 (75-84.4) 33.1 (27.6-38.6) 273.1 (261.4-286) 279.7 (271.4-285.8) 11R 3P 250Mi Out 79.2 (74-85.6) 31.5 (25.4-36.8) 276.7 (266-289) 282.1 (275.6-285.8) 11R 5P 150Mi Out 78.3 (71.4-84.2) 34.6 (29-41.4) 271.9 (262.4-285.6) 279.6 (273.2-285) 11R 5P 250Mi Out 80.5 (73.6-88) 31 (23.4-37.2) 277.1 (267.8-293.6) 281.5 (273-288.4) 4 Districts 48.1 (44.4-52) 22.6 (19-27.6) 277.3 (266.6-288.6) 284.1 (276-291) 4D 3P 150Mi In 49.2 (45.8-52) 22.6 (18.4-25.6) 280 (271.6-291.6) 284.3 (275.6-292.2) 4D 3P 250Mi In 48.2 (43.2-52.2) 22.8 (17.6-26.6) 280.7 (272.6-291.6) 284 (276.2-289) 4D 5P 150Mi In 49.7 (45.2-54.2) 23.9 (20.6-30.8) 279.6 (273-291.6) 284.6 (276.6-291.6) 4D 5P 250Mi In 49.6 (44.6-58) 25.1 (20.6-30.4) 281.5 (271.6-294) 284.1 (276.2-289) 4D 3P 150Mi Out 78.5 (72.2-82.6) 33.6 (30.2-40) 272.9 (263-281.4) 279.6 (272.6-285.6) 4D 3P 250Mi Out 79.6 (72.8-88.6) 31.7 (27.4-37.2) 276.5 (263.2-286.2) 282.8 (276.6-289.2) 4D 5P 150Mi Out 78.8 (73.4-86) 34.1 (27.6-41.2) 271.7 (262.4-285) 280.1 (272.4-287) 4D 5P 250Mi Out 82 (74.2-88.2) 31.2 (26.8-34.8) 276.3 (266.2-285.8) 282.9 (275.6-289) 8 Districts 71.1 (65.2-76) 27.1 (23.4-32) 276.1 (266.6-287.8) 282.8 (275-289.2) 8D 3P 150Mi In 72.6 (67.6-77.2) 28.6 (24.2-34.6) 276.5 (267-288.2) 282.4 (273.8-288) 8D 3P 250Mi In 72.1 (67.4-76) 27.9 (24-34.6) 277.6 (270.2-289.4) 282.5 (276.8-289.8) 8D 5P 150Mi In 72 (67.6-79.8) 29.5 (23.8-35.6) 275.3 (267-290.8) 282.6 (275.2-288.4) 8D 5P 250Mi In 72.5 (67.2-77.4) 29.5 (25-33.6) 277.3 (267.2-286.4) 281.7 (273.2-287.4) 8D 3P 150Mi Out 78.2 (71.4-85.4) 33.6 (29.2-40.2) 272.1 (262.6-282.2) 280.9 (272.4-285.8) 8D 3P 250Mi Out 79.1 (74.4-85.2) 31.2 (26.6-37) 274.7 (268.8-284.8) 282.7 (274.8-287.6) 8D 5P 150Mi Out 79.7 (75-86.4) 33.8 (30.6-41.2) 273.6 (265.4-285.4) 279.1 (271.4-286) 8D 5P 250Mi Out 81.3 (73.8-86) 32.4 (25.6-37.8) 277.2 (269-291) 282.9 (277.8-290.2)

SLIDE 68

68

Waitlist Mortality Rates by ME LD Category Table

35+ 29-34 15-28 <15 Current 1.249 (1.073-1.412) 0.23 (0.211-0.258) 0.063 (0.06-0.065) 0.028 (0.027-0.028) 11 Regions 1.261 (1.118-1.398) 0.265 (0.229-0.329) 0.061 (0.06-0.064) 0.028 (0.027-0.028) 11R 3P 150Mi In 1.247 (1.115-1.324) 0.269 (0.241-0.304) 0.062 (0.06-0.064) 0.028 (0.027-0.028) 11R 3P 250Mi In 1.249 (1.096-1.374) 0.257 (0.234-0.315) 0.062 (0.059-0.064) 0.028 (0.027-0.028) 11R 5P 150Mi In 1.262 (1.075-1.397) 0.254 (0.23-0.297) 0.062 (0.061-0.063) 0.028 (0.027-0.028) 11R 5P 250Mi In 1.286 (1.132-1.428) 0.259 (0.232-0.294) 0.062 (0.059-0.064) 0.028 (0.027-0.028) 11R 3P 150Mi Out 1.279 (1.17-1.418) 0.264 (0.224-0.318) 0.062 (0.06-0.063) 0.027 (0.027-0.028) 11R 3P 250Mi Out 1.222 (1.09-1.323) 0.275 (0.224-0.331) 0.061 (0.059-0.063) 0.027 (0.027-0.028) 11R 5P 150Mi Out 1.256 (1.153-1.374) 0.274 (0.232-0.332) 0.062 (0.06-0.063) 0.027 (0.027-0.028) 11R 5P 250Mi Out 1.265 (1.108-1.401) 0.271 (0.21-0.348) 0.061 (0.059-0.063) 0.027 (0.027-0.028) 4 Districts 1.088 (0.955-1.211) 0.295 (0.26-0.334) 0.055 (0.054-0.057) 0.027 (0.027-0.028) 4D 3P 150Mi In 1.112 (1.022-1.19) 0.284 (0.244-0.327) 0.057 (0.056-0.059) 0.027 (0.027-0.028) 4D 3P 250Mi In 1.072 (0.954-1.183) 0.282 (0.228-0.332) 0.057 (0.056-0.06) 0.027 (0.027-0.028) 4D 5P 150Mi In 1.091 (0.997-1.225) 0.28 (0.241-0.339) 0.058 (0.057-0.06) 0.028 (0.027-0.028) 4D 5P 250Mi In 1.083 (0.938-1.242) 0.287 (0.25-0.356) 0.058 (0.057-0.061) 0.027 (0.027-0.028) 4D 3P 150Mi Out 1.24 (1.07-1.403) 0.267 (0.237-0.34) 0.062 (0.06-0.064) 0.027 (0.027-0.028) 4D 3P 250Mi Out 1.238 (1.047-1.352) 0.277 (0.239-0.316) 0.061 (0.058-0.063) 0.027 (0.027-0.028) 4D 5P 150Mi Out 1.267 (1.176-1.414) 0.271 (0.225-0.324) 0.062 (0.06-0.063) 0.027 (0.027-0.028) 4D 5P 250Mi Out 1.287 (1.178-1.363) 0.273 (0.24-0.314) 0.061 (0.059-0.064) 0.027 (0.027-0.028) 8 Districts 1.178 (1.099-1.274) 0.28 (0.244-0.336) 0.058 (0.057-0.06) 0.027 (0.027-0.028) 8D 3P 150Mi In 1.202 (1.092-1.383) 0.289 (0.246-0.327) 0.06 (0.058-0.061) 0.027 (0.027-0.028) 8D 3P 250Mi In 1.201 (1.069-1.259) 0.281 (0.252-0.321) 0.06 (0.058-0.061) 0.027 (0.027-0.028) 8D 5P 150Mi In 1.185 (1.083-1.328) 0.282 (0.231-0.321) 0.06 (0.059-0.063) 0.027 (0.027-0.028) 8D 5P 250Mi In 1.208 (1.071-1.321) 0.285 (0.248-0.333) 0.06 (0.058-0.062) 0.027 (0.027-0.028) 8D 3P 150Mi Out 1.267 (1.122-1.389) 0.266 (0.23-0.304) 0.062 (0.06-0.064) 0.028 (0.027-0.028) 8D 3P 250Mi Out 1.236 (1.148-1.333) 0.272 (0.228-0.337) 0.061 (0.06-0.062) 0.027 (0.027-0.028) 8D 5P 150Mi Out 1.276 (1.16-1.468) 0.268 (0.237-0.308) 0.062 (0.061-0.064) 0.027 (0.027-0.028) 8D 5P 250Mi Out 1.261 (1.071-1.42) 0.287 (0.221-0.358) 0.061 (0.06-0.063) 0.027 (0.027-0.028)

SLIDE 69

69

Summary: 11 Region Broader Sharing

Disparity metrics:
Variance in median MELD at transplant increased for baseline and in-

district circles scenarios but decreased for out-district circles as compared to current policy

Variance in transplant rates did not differ from current policy in

baseline and in-district scenarios, decreased in out-district scenarios

Variance in waitlist mortality and overall mortality did not appreciably

differ from current policy

Summative metrics:
Mortality metrics were broadly similar to the simulation of current

policy across 11 region scenarios

Transport metrics:
11 region scenarios overall showed the smallest increase in median

transport time, flying percentage, and transport cost compared with current policy

11 region out-district scenarios did not differ from 4- or 8-district out-

district circle scenarios in transport patterns

SLIDE 70

70

Summary: 4 District Broader Sharing

Disparity metrics:
4-district baseline and in-district scenarios showed largest projected reduction of all

scenarios (more than 2 times) in variance in median MELD at transplant from current policy, largest reduction in variance in waitlist mortality and transplant rates.

4-district out-district circle scenarios also showed reductions in these disparity metrics

compared to the current system, but less so than other 4-district scenarios.

Summative metrics:
4-district baseline and in-district circle scenarios had greatest impact of all scenarios on

pretransplant deaths prevented per year, posttransplant deaths prevented per year in the same range as current policy, and greatest decrease in overall mortality and waitlist mortality counts and rates per year.

4-district out-district circle scenarios did not appreciably differ from current policy, 11

region, and 8-district out-district circle scenarios

Transport metrics:
4-district baseline scenario showed highest transport time, distance, and highest percent
flying. 4-district in-district circle scenarios reduced this effect.
4-district out-district circle scenarios did not differ from 11 region or 8-district
ut-district circle scenarios in transport patterns

SLIDE 71

71

Summary: 8 District Broader Sharing

Metrics for 8-district scenarios fell between the 11-region and 4-district results in most cases
Disparity metrics:
Variance in median MELD at transplant decreased by a factor of nearly 2 for 8-district baseline

and in-district circle scenarios (though not as far as 4-district baseline and in-district scenarios). Variance in transplant rates was similar to 4-district scenarios

8-district out-district circle scenarios also showed reductions in these disparity metrics

compared to the current system, but less so than other 8-district scenarios

Summative metrics:
8-district scenarios showed some advantage over current policy in pretransplant deaths

prevented per year.

For posttransplant deaths prevented per year, 8-district baseline and in-district scenarios did

not differ from current policy, out-district scenarios decreased prevented deaths

For overall mortality, 8-district baseline and in-district scenarios showed a decrease, but out-

district scenarios did not appreciably differ from current policy

Transport metrics:
8-district baseline and in-district scenarios had slightly less transport time, distance, and

percent flying than 4-district in-district scenarios. 8-district out-district scenarios did not differ from 11 region or 4-district out-district scenarios.

SLIDE 72

Supplemental Slides: Subgroup Analysis

SLIDE 73

Subgroup Analysis: Transplant Counts (Pediatrics)

SLIDE 74

Subgroup Analysis: Transplant Counts (Gender)

SLIDE 75

Subgroup Analysis: Transplant Counts (Race/ E thnicity)

SLIDE 76

Subgroup Analysis: Transplant Rates (Pediatrics)

SLIDE 77

Subgroup Analysis: Transplant Rates (Gender)

SLIDE 78

Subgroup Analysis: Transplant Rates (Race/ E thnicity)

SLIDE 79

Subgroup Analysis: Pretransplant Deaths Prevented (Pediatrics)

SLIDE 80

Subgroup Analysis: Pretransplant Deaths Prevented (Gender)

SLIDE 81

Subgroup Analysis: Pretransplant Deaths Prevented (Race/ E thnicity)

SLIDE 82

Subgroup Analysis: Posttransplant Deaths Prevented (Pediatrics)

SLIDE 83

Subgroup Analysis: Posttransplant Deaths Prevented (Gender)

SLIDE 84

Subgroup Analysis: Posttransplant Deaths Prevented (Race/ E thnicity)

SLIDE 85

Subgroup Analysis: Overall Mortality Counts (Pediatric)

SLIDE 86

Subgroup Analysis: Overall Mortality Counts (Gender)

SLIDE 87

Subgroup Analysis: Overall Mortality Counts (Race/ E thnicity)

SLIDE 88

Subgroup Analysis: Overall Mortality Rates (Pediatrics)

SLIDE 89

Subgroup Analysis: Overall Mortality Rates (Gender)

SLIDE 90

Subgroup Analysis: Overall Death Rates (Race/ E thnicity)

SLIDE 91

Subgroup Analysis: Median Transport Time (Pediatrics)

SLIDE 92

Subgroup Analysis: Median Transport Time (Gender)

SLIDE 93

Subgroup Analysis: Median Transport Time (Race/ E thnicity)

SLIDE 94

Subgroup Analysis: Median Transport Distance (Pediatrics)

SLIDE 95

Subgroup Analysis: Median Transport Distance (Pediatrics)

SLIDE 96

Subgroup Analysis: Median Transport Distance (Race/ E thnicity)

SLIDE 97

Subgroup Analysis: Pretransplant Mortality Counts (Pediatrics)

SLIDE 98

Subgroup Analysis: Pretransplant Mortality Counts (Gender)

SLIDE 99

Subgroup Analysis: Pretransplant Mortality Counts (Race/ E thnicity)

SLIDE 100

Subgroup Analysis: Pretransplant Mortality Rates (Pediatrics)

SLIDE 101

Subgroup Analysis: Pretransplant Mortality Rates (Gender)

SLIDE 102

Subgroup Analysis: Pretransplant Mortality Rates (Race/ E thnicity)

SLIDE 103

Subgroup Analysis: Posttransplant Mortality Counts (Pediatrics)

SLIDE 104

Subgroup Analysis: Posttransplant Mortality Counts (Gender)

SLIDE 105

Subgroup Analysis: Posttransplant Mortality Counts (Race/ E thnicity)

SLIDE 106

Subgroup Analysis: Posttransplant Mortality Rates (Pediatrics)

SLIDE 107

Subgroup Analysis: Posttransplant Mortality Rates (Gender)

SLIDE 108

Subgroup Analysis: Posttransplant Mortality Rates (Race/ E thnicity)

SLIDE 109

Subgroup Analysis: Variance in Median ME LD/ PE LD at Transplant (Pediatrics)

SLIDE 110

Subgroup Analysis: Variance in Median ME LD/ PE LD at Transplant (Gender)

SLIDE 111