February 13-15, 2012
Key Bridge Marriott Hotel
Arlington, Virginia

Cosponsors: Scientific Registry of Transplant Recipients (SRTR), Organ Procurement and Transplantation Network (OPTN)

Background: Organ transplant programs in the US must submit data to the OPTN. These data are used by SRTR to calculate each transplant program’s observed and expected outcomes, which are then reported regularly to the OPTN Membership and Professional Standards Committee (MPSC), the Centers for Medicare & Medicaid Services (CMS), and the general public. When a transplant program fails to achieve expected outcomes, it becomes the target of further scrutiny by both the MPSC and CMS. In addition, private insurance providers may use these same data to determine where their patients are referred for transplantation.

Purpose of the conference: To discuss the methods used by the SRTR in the surveillance of solid-organ transplant programs and to make recommendations for improvements.

Participants: 115 attendees from transplant programs, the SRTR, the OPTN, Centers for Medicare and Medicaid Services (CMS), and the Health Resources and Services Administration (HRSA).

Download them All

Agenda

Program Specific Report Action Items

Overview of Program Specific Reports – Bert Kasiske
The mandate for PSRs and surveillance of organ transplant programs originates in the Final Rule. The PSRs are produced by the Scientific Registry of Transplant Recipients (SRTR) under contract from the Health Resources and Services Administration (HRSA). The data for PSRs are collected by the Organ Procurement and Transplantation Network (OPTN). The SRTR publishes observed and expected outcomes every 6 months using a 30-month, rolling, national cohort and Cox proportional hazards models to adjust for risk. The OPTN Membership and Professional Standards Committee (MPSC) uses performance measures applied to the PSRs to identify transplant programs that appear to be underperforming and therefore require further scrutiny. The Centers for Medicare and Medicaid Services (CMS) separately scrutinizes programs using OPTN data and methods similar to those used by the MPSC.

The OPTN Perspective – Alan Reed

The CMS Perspective – Thomas Hamilton

The Program’s Perspective – Barry Friedman

The Private Insurers’ Perspective – Richard Migliori

The Public’s Perspective – David Howard

Unintended Consequences of PSRs – Jesse Schold
Many risk factors for poor outcomes may not be taken into account in PSR models, and transplanting patients with those risk factors may have an adverse effect on center-specific outcomes. This could discourage innovation and deny access to transplantation for high-risk patients that may nevertheless benefit from transplantation. Is there evidence that this may be occurring?

Analysis of the Effect of PSRs on Center Behaviors – Jon Snyder
An SRTR analysis of changes in transplant center outcomes and behaviors before and after CMS policy changes that may have created a disincentive for centers to transplant higher risk patients will be presented.

What Can We Learn From Other Experiences – David Naftel
The National Institutes of Health-sponsored Interagency registry for Mechanically Assisted Circulatory Support (INTERMACS) database has been used to assess the safety and efficacy of cardiac assist devices. Risk assessment has been key to understanding outcomes in this multicenter database.

What Can We Learn From Other Experiences – Douglas Rizzo
The Center for International Blood and Bone Marrow Transplant Research (CIBMTR) has conducted research and quality assurance for autologous and allogeneic hematopoietic cell transplants. Methods for operating this database and its use in outcomes surveillance will be described.

CUSUM – David Axelrod
A cumulative sum time-series technique, CUSUM, may be able to detect a change in outcomes at a transplant center sooner than currently used PSR methods. CUSUM can be adjusted for risk, can be used for different outcomes, and can be calculated at any time for use as a quality assurance measure. CUSUM is used in the United Kingdom to monitor outcomes at organ transplant centers.

Other Analytical Techniques – Nick Salkowski
Both the current PSR and CUSUM methodologies have advantages and disadvantages. Extrapolating short term to estimate long term outcomes can be accomplished with different techniques and variable accuracy. In the end, more than one technique could be used to better assess a variety of center-specific outcomes.

Framing Questions: Limitation of Current System – Ajay Israni
Not surprisingly, the PSRs and their application have been controversial. A number of potential weaknesses have become apparent.

Methods – Nick Salkowski
Should the current methods be modified?

• Use of Cox models with statistically significant covariates
• Use of 30 month moving cohorts How should centers with few transplants be handled?
• Remove smaller programs from analysis (if so at what threshold)
• Use larger cohorts/intervals for smaller programs Should different or additional methods be used?
• CUSUM
• Projected long term outcomes, e.g. half-lives, period analysis

Risk Adjustment – Mike Abecassis
What process should be used to determine risk adjustment? How should risk be adjusted?

• Covariates in statistical models
• Excluded from analysis What major risk factors should be included?
• Cardiovascular disease risk, e.g. AMI, revascularizations
• Immunological risk, e.g. DSA (before and after desensitization)
• Other recipient risk factors
• Donor risk

Outcomes – Larry Hunsicker
Are the current outcomes appropriate? What additional outcomes should be assessed?

• Outcomes on the waiting list, e.g. rates of death, acceptance, transplant, removal
• Composite pre-transplant metrics, e.g. acceptance, mortality, and transplant rates
• Long-term follow-up of post-transplant patient and graft survival
• Life-Years from Transplantation (LYFT)
• Quality of life
• Cost (to Medicare)
• Acute rejections
• Hospitalizations

Data – Stuart Sweet
Is there a need for additional data to be collected? What process should be used to determine additional data that should be collected? What criteria should be used to determine additional data that should be collected? What additional organ-specific data should be collected?

• for kidney transplant recipients?
• for liver transplant recipients?
• for pancreas transplant recipients?
• for heart transplant recipients?
• for lung transplant recipients?

Break-Out Groups
Methods – Chairs: Robert Gaston & Stuart Sweet
Risk Adjustment – Chairs: Mitch Henry & Dennis Irwin
Outcomes – Chairs: Alan Reed & Nancy Metzler
Data – Chairs: John Roberts & Danielle Cornell