Journal of the National Cancer Institute Advance Access originally published online on June 27, 2007
JNCI Journal of the National Cancer Institute 2007 99(13):1050; doi:10.1093/jnci/djm015
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press 2007.
CORRESPONDENCE |
Re: Acute Myeloid Leukemia or Myelodysplastic Syndrome Following Use of Granulocyte Colony-Stimulating Factors During Breast Cancer Adjuvant Chemotherapy
Affiliation of authors: Health Services and Economics Branch, Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD
Correspondence to: Joan L. Warren, PhD, Health Services and Economics Branch, Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Executive Plaza North, Rm 4005, 6130 Executive Blvd, MSC 7344, Bethesda, MD 20892-7344 (e-mail: joan_warren{at}nih.gov).
The recent article in the Journal by Hershman et al. (1) reports on rates of acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS) among women with breast cancer following treatment with chemotherapy and granulocyte colony-stimulating factors. This analysis, based on Surveillance, Epidemiology, and End Results (SEER)–Medicare data, uses women treated with chemotherapy alone as the comparison group. Both the article and the accompanying Editorial (2) state that Medicare claims have not been validated and that it is not known how sensitive the claims information in the SEER Program database is for primary cancers. These statements could lead to a misunderstanding of the data and require clarification.
The SEER data include information reported from population-based registries. The registries abstract information from the medical record of persons who are newly diagnosed with cancer. SEER data undergo extensive quality assessment and have been found to have complete and reliable case ascertainment and staging (3). The SEER data include no health care claims and use International Classification of Diseases for Oncology codes to identify newly diagnosed cases.
To augment information collected by the registries, the National Cancer Institute has linked persons in the SEER data to Medicare enrollment files; 93% of persons aged 65 years and older in SEER have been linked to Medicare data. Procedure codes found on Medicare claims, including chemotherapy, have been shown to have good agreement with the medical record (4–7). Diagnostic codes on Medicare claims have been shown to have high positive predictive value for specific conditions, although the sensitivity of the data varies from low to high.
In their analysis, Hershman et al. (1) used diagnoses from the Medicare data to identify the occurrence of MDS and AML. To address concerns about the validity of the reporting of these conditions in the Medicare claims, the authors performed a sensitivity analysis, evaluating how requiring two or more claims of MDS or AML would affect their findings. They report that there was no change in the hazard ratio, although they do not report the specific point estimates, which would have allowed readers to judge the differences themselves. Findings from the earlier validations of Medicare data suggest that claims with MDS and AML codes have a high positive predictive value, with unknown sensitivity. If the sensitivity were low, the estimates reported in the study by Hershman et al. would be a lower bound.
Large population-based observational datasets, such as SEER–Medicare data, are not intended to provide definitive information about treatment outcomes. Rather, these data can be used to target more in-depth clinical evaluations. To use Medicare or SEER–Medicare data correctly and interpret the findings from analyses using these data, it is important that researchers and readers have to have an accurate understanding of the quality of the files.
REFERENCES
(1) Hershman D, Neugut AI, Jacobson JS, Wang J, Tsai WY, McBride R, et al. Acute myeloid leukemia or myelodysplastic syndrome following use of granulocyte colony-stimulating factors during breast cancer adjuvant chemotherapy. J Natl Cancer Inst (2007) 99:196–205.
(2) Touw IP, Bontenbal M. Granulocyte colony-stimulating factor: key (f)actor or innocent bystander in the development of secondary myeloid malignancy? J Natl Cancer Inst (2007) 99:183–6.
(3) Harlan LC, Hankey BF. The surveillance, epidemiology, and end-results program database as a resource for conducting descriptive epidemiologic and clinical studies. J Clin Oncol (2003) 21:2232–3.
(4) Fowles JB, Lawthers AG, Weiner JP, Garnick DW, Petrie DS, Palmer RH. Agreement between physicians’ office records and Medicare Part B claims data. Health Care Financ Rev (1995) 16:189–99.[Medline]
(5) Lawthers AG, McCarthy EP, Davis RB, Peterson LE, Palmer RH, Iezzoni LI. Identification of in-hospital complications from claims data. Is it valid? Med Care (2000) 38:785–95.[CrossRef][Web of Science][Medline]
(6) Javitt JC, McBean AM, Sastry SS, DiPaolo F. Accuracy of coding in Medicare part B claims. Cataract as a case study. Arch Ophthalmol (1993) 111:605–7.
(7) Warren JL, Harlan LC, Fahey A, Virnig BA, Freeman JL, Klabunde CN, et al. Utility of the SEER-Medicare data to identify chemotherapy use. Med Care (2002) 40(Suppl):IV-55–61.
Response to this Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 1051-1052.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||