Currently Viewing:
The American Journal of Managed Care September 2015
Do Patient or Provider Characteristics Impact Management of Diabetes?
Erin S. LeBlanc, MD, MPH; A. Gabriela Rosales, MS; Sumesh Kachroo, PhD; Jayanti Mukherjee, PhD; Kristine L. Funk, MS; and Gregory A. Nichols, PhD
The Utility of Cost Discussions Between Patients With Cancer and Oncologists
S. Yousuf Zafar, MD, MHS; Fumiko Chino, MD; Peter A. Ubel, MD; Christel Rushing, MS; Gregory Samsa, PhD; Ivy Altomare, MD; Jonathan Nicolla, MBA; Deborah Schrag, MD; James A. Tulsky, MD; Amy P. Abernethy, MD, PhD; and Jeffery M. Peppercorn, MD, MPH
Building Upon the Strong Foundation of National Healthcare Quality
Charles N. Kahn III, MPH, President and CEO, Federation of American Hospitals
Improving Partnerships Between Health Plans and Medical Groups
Howard Beckman, MD, FACP, FAACH; Patricia Healey, MPH; and Dana Gelb Safran, ScD
Innovative Approach to Patient-Centered Care Coordination in Primary Care Practices
Robin Clarke, MD, MSHS; Nazleen Bharmal, MD, PhD; Paul Di Capua, MD, MBA; Chi-Hong Tseng, PhD; Carol M. Mangione, MD, MSPH; Brian Mittman, PhD; and Samuel A. Skootsky, MD
Private Sector Risk-Sharing Agreements in the United States: Trends, Barriers, and Prospects
Louis P. Garrison, Jr, PhD; Josh J. Carlson, PhD; Preeti S. Bajaj, PhD; Adrian Towse, MA, MPhil; Peter J. Neumann, ScD; Sean D. Sullivan, PhD; Kimberly Westrich, MA; and Robert W. Dubois, MD, PhD
Developing Evidence That Is Fit for Purpose: A Framework for Payer and Research Dialogue
Rajeev K. Sabharwal, MPH; Jennifer S. Graff, PharmD; Erin Holve, PhD, MPH, MPP; and Robert W. Dubois, MD, PhD
Predicting Adherence Trajectory Using Initial Patterns of Medication Filling
Jessica M. Franklin, PhD; Alexis A. Krumme, MS; William H. Shrank, MD, MSHS; Olga S. Matlin, PhD; Troyen A. Brennan, MD, JD, MPH; and Niteesh K. Choudhry, MD, PhD
Reply to "Transforming Oncology Care": Advancing Value, Accessing Innovation
Rebecca Paradis, MPA
Payer Source Influence on Effectiveness of Lifestyle Medicine Programs
Joseph Vogelgesang, BS; David Drozek, DO; Masato Nakazawa, PhD; Jay H. Shubrook, DO
Currently Reading
High-Risk Centers and the Benefits for Lower-Risk Transplants
Schelomo Marmor, PhD, MPH; James W. Begun, PhD; Jean Abraham, PhD; and Beth A. Virnig, PhD, MPH
Socioeconomic Disparities Across Ethnicities: An Application to Cervical Cancer Screening
Brendan Walsh, PhD; and Ciaran O'Neill, PhD

High-Risk Centers and the Benefits for Lower-Risk Transplants

Schelomo Marmor, PhD, MPH; James W. Begun, PhD; Jean Abraham, PhD; and Beth A. Virnig, PhD, MPH
There does not appear to be any comparative advantage for low-risk hematopoietic cell transplantation patients to seek care from high-risk centers.
Objectives: Allogeneic hematopoietic cell transplantation (HCT) is the transplantation of stem cells from a donor and an effective treatment for many hematologic malignancies. We sought to compare allogeneic HCT survival outcomes and hazard of death among US centers that treat higher-risk patients versus those in centers that do not perform lower-risk HCT procedures. 

Study Design: We utilized 2008 to 2010 Center for International Blood and Marrow Transplant Research data. We categorized patients into 4 risk categories that align with factors shown in the literature to be associated with HCT survival. We stratified centers into those that do and do not conduct high-risk pre-transplant HCT.

Methods: To further evaluate the association between pre-transplant mortality risk and HCT survival by transplant center, we examined the association between risk category score and hazard of death using Cox proportional hazard modeling. 

Results: There were 12,436 HCT recipients at 147 transplant centers. Of the 147 centers, 74 performed HCT for patients ranging from the lowest risk category to the highest category, and 73 centers performed only lower-risk HCT. Adjusting for all other factors, lower-risk patients that underwent transplants in lower- or higher-risk centers had a similar relative hazard of death (P ≤.05). 

Conclusions: Low-risk patients had similar survival outcomes irrespective of whether they underwent transplant at higher- or lower-risk centers. Patient and payer policy implications could include initiatives that reduce travel for low-risk patients. Similarly, HCT center administrators and providers that manage higher-risk patients need not expect commensurate benefits in survival for lower-risk patients.  
Am J Manag Care. 2015;21(9):e509-e518
Take-Away Points
We sought to compare allogeneic hematopoietic cell transplantation (HCT) survival outcomes and hazard of death among US centers that treat higher-risk patients versus lower-risk centers that do not. Low-risk patients had similar survival outcomes regardless of whether they had a transplant performed at higher- or lower-risk centers. HCT center administrators and managers need not expect that the performance of higher-risk HCT provides benefits in survival for lower-risk patients.
  • Lower-risk patients who underwent transplants in either lower- or higher-risk centers had a similar relative hazard of death. 
  • There should be a reduction in policy emphasis on Centers of Excellence for lower-risk patients. 
  • We expect that health plans will increasingly use risk-stratified types of data to encourage lower-risk patients to restrict travel and receive comparable care at local transplant centers.
Hematopoietic cell transplantation (HCT) is a complex treatment procedure for various hematologic malignancies and other conditions that are often otherwise incurable. Each year, approximately 17,000 patients receive HCTs in the United States. This number has been steadily increasing since 20001 due to advances in transplantation over the last several decades that have resulted in an increasing number of healthcare facilities that are more willing to perform more complex and higher-risk transplants. Although mortality is a useful measure of transplant center quality, comparing outcomes among centers is challenging if estimates do not take differences in transplant populations into account.2,3 For instance, centers that perform transplants on a relatively larger percentage of high-risk patients with an intrinsically higher risk of mortality may be potentially perceived as poor performers compared with centers that perform transplants on a lower percentage of such patients.

Patient risk can affect overall center performance in several ways. Variation in HCT patient characteristics in centers that treat higher-risk patients could deplete human and financial resources for lower-risk patients. Alternatively, establishing processes to successfully manage high-risk patients could benefit lower-risk patients as well, thereby increasing quality and lowering the procedural mortality for all patients. There is a well-documented body of HCT literature indicating that certain patient characteristics—such as human leukocyte antigen (HLA) matching, comorbidities, age, and Karnofsky performance score—are determinants of a patient’s pre-transplant risk level and survival rates.4-18

In this paper we explore the effects on survival for lower-risk HCT patients undergoing transplant at HCT centers that do or do not perform high-risk transplants. We hypothesized that there were demonstrably superior survival results for low- and moderate-risk patients as transplant centers continue to explore new clinical successes with higher-risk patients. Specifically, we focused on potential spillover effects and benefits of higher-risk HCT performance on the low-risk patient population within the same centers. We seek to evaluate differences in outcomes among lower-risk HCT patients by risk-stratifying centers that perform HCT in high-risk patients versus centers that do not perform high-risk HCT.

Data Source

The data were obtained from the statistical center of the Center for International Blood and Marrow Transplant Research (CIBMTR), located at the Medical College of Wisconsin in Milwaukee, and from the National Marrow Donor Program. The CIBMTR is partially supported by Grant U24-CA76518 from the National Institutes of Health, and by the Health Resources and Services Administration (HRSA). CIBMTR is composed of a voluntary working group of more than 450 transplantation centers worldwide that contribute detailed data on consecutive HCT to its Statistical Center. In addition, the CIBMTR holds the contract for the Stem Cell Therapeutic Outcomes Database part of the C.W. Bill Young Transplantation Program from the HRSA. As part of this program, all transplant centers in the United States are mandated to report clinical outcomes data for HCT to the CIBMTR. We obtained a de-identified data set from the CIBTMR. The analysis has not been reviewed by the CIBMTR. Our study was deemed exempt from review by the Human Subjects Committee of the University of Minnesota’s Institutional Review Board.


Our cohort included patients 18 years or older who received transplants between January 1, 2008, and December 31, 2010, for whom data were reported to the CIBTMR. We excluded patients missing any of the 4 risk category (RC) criteria (N = 406 patients; see below for RC derivation) and centers that reported only 1 transplant from 2008 to 2010 (N = 15 centers).

Derivation of the Patient Risk Categories

The literature indicates that several patient characteristics are clinically important to the long-term survival of HCT recipients. To determine patient risk, we chose 4 characteristics that have been consistently reported across studies to be associated with survival following HCT: age at transplant, HLA match status, Karnofsky performance score, and comorbidities.4-17 We used binary risk indicators for the 4 patient characteristics to create a risk score for overall mortality that ranged from 0 to 4. Table 1 is a summary of characteristics included in the RCs.

Transplant recipients older than 40 years (score of 1) were considered to be higher risk than younger individuals (score of 0).5 The Karnofsky performance status is used to determine the recipient’s functional status and can range from 0 to 100. A Karnofsky performance score of 90 to 100 categorizes patients with the ability to carry on normal activity prior to transplant. CIBMTR codes Karnofsky performance score as a dichotomous variable of 90-100 (score of 0) and ≤80 (score of 1). Coexisting disease is a binary category of diseases collected by CIBMTR. CIBMTR codes HCT patients with any of 18 comorbidities as coexisting disease present (score of 1). Patients with no coexisting disease were scored as 0. The HLA match status of donors describes the degree of immunologic similarity between recipients and donors. HLA 8/8 match, well-matched unrelated, and HLA-matched HCTs were all scored as 0; any mismatched unrelated (HLA 6/8, or 7/8 matched, partially matched, or mismatched) or mismatched-related HCTs were scored as 1. Prior to consolidating related and unrelated transplant groups within our RC, we created separate RCs for each transplant group and independently verified our models for each group.

We considered patients with all 4 risk components—HLA mismatch, coexisting disease, age ≥40, and Karnofsky performance score <80—to be the highest pre-transplant risk within our analytical cohort (scored with an RC of 4). Conversely, transplant recipients with an RC score of 0—HLA match, no coexisting disease, aged <40, and Karnofsky performance score ≥90—were considered to have the lowest risk within our cohort. Patients with an RC score of 1 or 2 and 3 were considered to have moderate pre-HCT risk.

Overall survival for patients with scores of 1 and 2 were similar but different enough from a score of 3. We therefore combined scores 1 and 2 into a single group. We had a total of 4 risk groupings, including risk scores of 0, combined 1 and 2, 3, and 4. Although the data we received from CIBTMR were dichotomous and only broadly identified coexisting disease, HLA match, and Karnofsky performance score as either present or absent, we confirmed that our RCs were illustrative of distinct patient risk. Using Kaplan-Meier methods, we found that the difference in survival probability for our RCs was significant and observed distinct differences among all 4 groups (Figure 1). We tested different age cut-offs to ensure that our risk assumptions did not change our results. Given that our cut-off has been supported by the literature6 and that using other cut-offs in age does not change the magnitude and direction of our results, we found it beneficial to classify age as a binary variable. We also tested each of the RCs independently in our models to ensure that all individual components of the risk score behaved similarly. Under all assumptions, conclusions remain unchanged.

Center Characteristics

Based on our pre-HCT patient risk score categories, we categorized centers into either high- or low-risk. Centers that performed HCTs with only lower-risk patients (defined as risk scores = 0-3) were considered to be low-risk centers. Of the 147 centers, 73 (N = 1984 transplants) performed only lower-risk HCT. There were 74 high-risk centers, categorized as any center that performed HCT for patients with the highest RC score of 4 (N = 6864 transplants). We recorded other center characteristics that included a high-volume center indicator for centers conducting more than the mean number of transplants across the 3 years of observation. To adjust for possible high-risk center effects and the inclusion of unrelated donors in our cohort, a transplant center and related/unrelated donor indicator was added to all multivariate models as frailty variables. We also adjusted for region using the 10 HHS regions.

Statistical Analysis

We evaluated differences in center RCs and patient characteristics across all years. After assessing this unadjusted relationship, we evaluated the association between our RCs and 3-year hazard of death using Kaplan-Meier methods and Cox proportional hazards modeling. Kaplan-Meier methods were used to estimate unadjusted 3-year cumulative mortality across the RCs for our entire cohort, for a bifurcated cohort of centers that performed high-risk transplants, and for transplant centers that did not perform high-risk transplants within our analytical period. Three separate multivariable Cox models were used to compare the impact of pre-transplant risk factors independently for high- and low-risk centers and a combined model for all centers. Within the overall combined model, patients from high- and low-risk centers were included together. The model adjusted for patient factors and included a bivariate high-risk center indicator (yes/no) and a higher-risk unrelated donors indicator.

In all models, we performed several sensitivity analyses to ensure that the observed effects were not a product of our RCs and modeling decisions. We compared models that included each risk component measured separately to verify that there was statistical benefit to the creation and inclusion of our risk scores. Prior to consolidating related and unrelated transplant groups within our RC, we created separate RCs for each transplant group and independently verified our models for each group. We conducted sensitivity analyses utilizing the Sorror comorbidity score, an alternative comorbidity score for HCT, which did not produce different results.11 In addition, we restricted our analysis to the largest disease groupsacute myeloid leukemia (AML), acute lymphoblastic leukemia, non-Hodgkin lymphoma, Hodgkin lymphoma—to uncover broader trends. We created a more conservative cut-off point to define high-risk centers (eg, ≥5% high-risk patients treated) to ensure that we discovered no changes in our findings. We also conducted separate survival analyses with the use of hierarchical linear models to verify that accounting for transplant recipients being nested in centers did not produce results of different magnitude or direction. Under all assumptions, conclusions remain unchanged. SAS version 9.3 (SAS Institute, Cary, North Carolina) was used for all analyses. P values were 2-sided with a level of significance of ≤.05.

Our total cohort included 12,436 allogeneic transplants conducted in 147 centers. Over half of our patient cohort was aged 40 or older, and 36% of our cohort was determined to need special care to carry on normal activity prior to transplantation (Karnofsky performance scoring <80). Nearly 66% of our population was classified by CIBMTR to have coexisting disease prior to transplant (Table 2).

Copyright AJMC 2006-2020 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
Welcome the the new and improved, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up