The Hematologist

January-February 2012, Volume 9, Issue 1

The Value of Specialty Care in Malignant Hematology

John C. Byrd, MD

Published on: January 25, 2012

Dr. Bryd indicated no relevant conflicts of interest. 

Shanafelt RD, Kay NE, Rabe NE, et al. Hematologist/oncologist disease-specific expertise and survival: Lessons from chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL). Cancer. 2011. [Epub ahead of print].

Among colleagues who recently returned from the 2011 ASH Annual Meeting in San Diego, a common discussion point has been how much and how rapidly management of each disease entity in hematology changes as both diagnostic and therapeutic options grow in mass and complexity. For those of us who are specialists in a given disease, we marvel at how general hematologists and hematologist/oncologists strive to stay current with this information explosion while integrating the new developments with the “experience/art” component of specialty practice to provide outstanding patient care. And as a corollary, as specialists we have come to suspect that, for those areas of hematology outside of our focus, our ability to provide state-of-the-art care may be suboptimal. Until recently, however, there have been no objective studies addressing this issue. Now, a paper by Shanafelt and colleagues that retrospectively examined management of chronic lymphocytic leukemia (CLL) patients by “CLL specialists” and by other “hematology specialists” at the Mayo Clinic supports our intuitive impressions by showing that survival is greater for patients under the care of a hematology specialist.

The study by Shanafelt and colleagues showed that patients treated by CLL specialists were statistically more likely to undergo prognostic testing at diagnosis and more likely to experience a longer period of observation before beginning therapy (9.2 vs. 6.1 years [p< 0.001]). Patients treated by the CLL specialist were more likely both to receive purine analog-based therapy and to enroll in prospective clinical trials. The median overall survival for the group treated by a CLL specialist was 10.5 years compared with 8.8 years for those treated by non- CLL-specialist hematologists (p=.002). While there were slight differences in the characteristics of the patients assigned to each group of physicians, these differences are considered insufficient to account for the reported findings. Of note, in neither group was there a difference in outcome when comparisons were made between patients managed by an attending and fellow in training together compared with those managed by the attending alone, suggesting that the current model of training new hematologists does not adversely affect how patients fare.

The retrospective findings of Shanafelt and colleagues may have bearing both on allocation of resources and on the practice of hematology in academic medical centers as, unlike other studies that examined outcome of patients at small treatment centers to the outcome of patients at larger treatment centers, this study involved physicians within a single, well-respected academic institution. Based on both univariate and multivariate analysis, CLL/SLL patients benefit from specialty management. But important issues surrounding these findings need to be addressed. First, the results need to be prospectively validated, ideally in a separate medical center where specialists in CLL are prevalent but where patients with CLL are also managed by non- CLL/SLL-specialist hematologists. Secondly, the impact of similar specialty care on outcome in other diseases such as chronic myeloid leukemia, other lymphoproliferative disorders, and acute leukemia must also be assessed. Such prospective studies might choose to examine how often guidelines (e.g., those from NCCN) are followed to better understand how differences in outcome might emerge. Finally, as the study of Shanafelt et al. did not include comparison of outcomes involving either true general hematologists or hematologist/oncologists with those of the CLL specialist, the results should not be extrapolated to those two groups of physicians.

back to top