Case Example 14Selecting Data Elements for a Registry

DescriptionThe Dosing and Outcomes Study of Erythropoiesis-stimulating Therapies (DOSE) Registry was designed to understand anemia management patterns and clinical, economic, and patient-reported outcomes in oncology patients treated in outpatient oncology practice settings across the United States. The prospective design of the DOSE Registry enabled data capture from oncology patients treated with erythropoiesis- stimulating therapies.
SponsorCentocor Ortho Biotech Services, LLC
Year Started2003
Year Ended2009
No. of Sites71
No. of Patients2,354


Epoetin alfa was approved for patients with chemotherapy-induced anemia in 1994. In 2002, the U.S. Food and Drug Administration approved a second erythropoiesis-stimulating therapy (EST), darbepoetin alfa, for a similar indication. While multiple clinical trials described outcomes following intervention with ESTs, little information was available on real-world practice patterns and outcomes in oncology patients. The registry team determined that a prospective observational effectiveness study in this therapeutic area was needed to gain this information. The three key challenges were to make the study representative of real-world practices and settings (e.g., hospital-based clinics, community oncology clinics); to collect data elements that were straightforward so as to minimize potential data collection errors; and to collect sufficient data to study effectiveness, while ensuring that the data collection remained feasible and time efficient for outpatient oncology clinics.

Proposed Solution

The registry team began selecting data elements by completing a thorough literature review. Because this would be one of the first prospective observational studies in this therapeutic area, the team wanted to ensure that study results could be presented to health care professionals and decisionmakers in a manner consistent with clinical trials, of which there were many. The team also intended to make the data reports from this study comparable with clinical trial reports. To meet these objectives, data elements (e.g., baseline demographics, dosing patterns, hemoglobin levels) similar to those in clinical trials were selected whenever possible, based on a thorough literature review.

For the patient-reported outcomes component of the registry, the team incorporated standard validated instruments. This decision allowed the team to avoid developing and validating new instruments and supported consistency with clinical trial literature, as many trials had incorporated these instruments. To capture patient-reported data, the team selected two instruments, the Functional Assessment of Cancer Therapy–Anemia (FACT-An) and the Linear Analog Scale Assessment (LASA) tool. The FACT-An tool, developed from the FACT-General scale, had been designed and validated to measure the impact of anemia in cancer patients. The LASA enables patients to report their energy level, activity level, and overall quality of life on a scale of 0 to 100. Both tools are commonly used to gather patient-reported outcomes data for cancer patients.

Following the literature review, an advisory board was convened to discuss the registry objectives, data elements, and study execution. The advisory board included representatives from the medical and nursing professions. The multidisciplinary board provided insights into both the practical and clinical aspects of the registry procedures and data elements. Throughout the process, the registry team remained focused on both the overall registry objectives and user-friendly data collection. In particular, the team worked to make each question clear and unambiguous in order to minimize confusion and enable a variety of site personnel, as well as the patients, to complete the registry data collection.


The registry was launched in 2003 as one of the first prospective observational effectiveness studies in this therapeutic area. Seventy-one sites and 2,354 patients enrolled in the study. The sites participating in the registry represented a wide geographic distribution and a mixture of outpatient practice settings.

Key Point

Use of common data elements, guided by a literature review, and validated patient-reported outcomes instruments enhanced data generalizability and comparability with clinical trial data. A multidisciplinary advisory board also helped to ensure collection of key data elements in an appropriate manner from both a clinical and practical standpoint.

For More Information

  1. Larholt K, Burton TM, Hoaglin DC. et al. Clinical and patient-reported outcomes based on achieved hemoglobin levels in chemotherapy-treated cancer patients receiving erythropoiesis-stimulating agents. Commun Oncol. 2009;6:403–8.
  2. Larholt K, Pashos CL, Wang Q. et al. Dosing and Outcomes Study of Erythropoiesis-Stimulating Therapies (DOSE): a registry for characterizing anaemia management and outcomes in oncology patients. Clin Drug Invest. 2008;28(3):159–67. [PubMed: 18266401]

From: Chapter 5, Data Elements for Registries

Cover of Registries for Evaluating Patient Outcomes: A User's Guide
Registries for Evaluating Patient Outcomes: A User's Guide. 2nd edition.
Gliklich RE, Dreyer NA, editors.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.