ARRA IMPACT REPORT:
Comparative Effectiveness Research for Cancer
Public Health Burden
Cancer is the second leading cause of death in the United States after heart disease. In 2013, it is estimated that nearly 1.66 million new cases of invasive cancer will be diagnosed in this country, and more than 580,000 Americans will die of cancer.
Comparative Effectiveness Research Overview
To improve the overall quality and value of health care, the Institute of Medicine (IOM) recommended development of comparative effectiveness research (CER) to inform health care decisions. CER compares health outcomes, lifestyle implications, and value of competing approaches to diagnosis, treatment, or disease management to determine their relative effectiveness. CER (also known as patient-centered outcomes research) empowers clinicians to make informed health care decisions to optimize patient outcomes and improve overall health care value.
How ARRA Funding Supported Cancer-Related CER
The American Recovery and Reinvestment Act (ARRA) supported efforts to build capacity and accelerate scientific progress in cancer-related CER. The funds enabled NCI to support approximately $70 million in research and training in this area. Research funds were used to address a broad spectrum of cancer issues including cancer-screening methods, interventions to alter risk behaviors, surgery and treatment options, community cancer surveillance, and genomic medicine. ARRA funds also have contributed to the CER infrastructure by improving methods and resources for CER research and by providing training opportunities for the next generation of CER investigators. Through these efforts, NCI built teams of interdisciplinary researchers, leveraged and integrated existing data and health system research resources, and advanced measurement techniques and methodology.
Cancer-Related CER Progress
Progress in CER requires a robust infrastructure. A 2009 IOM report on CER priorities1 noted that new methods, databases, registries, and researchers were needed to develop a sustained national CER effort. This infrastructure would allow CER investigators to accomplish recommended research and identify future priorities. Through ARRA, NCI funded multiple projects that have helped advance the development of cancer-related CER infrastructure.
- Genome-enabled Electronic Medical Records: At the University of Virginia, researchers led by Principal Investigator Dr. William Knaus are creating a genome-enabled electronic medical record that will provide standardized information about a patient’s genome and family history. Such records will form the cornerstone for future use of genome and family history information to identify optimal prevention strategies and therapies for an individual. These records will be used in two proof-of-principle CER studies: 1) one to evaluate the clinical utility of family history data collection and individualized recommendations on risk reduction behaviors for cancer prevention and 2) the other to evaluate the impact of genetic counseling for patients at high risk for hereditary cancer. The use of these records ultimately will allow patients and physicians to make informed, personalized decisions about prevention and treatment options, thereby potentially reducing the use of ineffective approaches and improving the overall value of the health intervention.2
- Breast Cancer Surgical Outcomes (BRCASO) Database: Researchers at Van Andel Research Institute in Grand Rapids, Michigan, identified an information gap in the evaluation of breast cancer surgical outcomes. They found that the common measures of surgical quality—morbidity and mortality—were insufficient to inform breast cancer surgical best practices. Principal Investigator Dr. Laurence McCahill and colleagues developed the Breast Cancer Surgical Outcomes (BRCASO) database to identify quality measures that allow for meaningful comparisons of breast cancer surgical outcomes among a diverse population of patients, providers, and facilities. This database provides the infrastructure for future CER to evaluate a variety of surgical quality measures, including mastectomy rates, positive margin rates, and partial mastectomy reexcision rates. To date, the researchers have identified substantial variation among surgeons and institutions in reexcision rates following partial mastectomy. These observations will help physicians identify future best practices for breast cancer surgery.3
- Advances in Language Processing to Improve Access to Clinical Information in Electronic Medical Records: At Group Health Cooperative in Seattle, Washington, researchers led by Principal Investigator Dr. David Carrell are developing new technologies that use natural language processing to extract clinical information about disease processes and treatments from electronic medical records.4 This technology will allow researchers to provide structure to historical clinical data and to build comprehensive databases for the study of disease progression, treatment effectiveness, and long-term outcomes. In addition, this project has integrated a de-identification tool to allow researchers access to clinical information without compromising privacy. Collectively, these efforts have paved the way for cross-institution collaborations in CER and have spurred at least three successful disease-specific grant applications.5
- CER Career Development Program: Dr. Victoria Fraser is Principal Investigator and Director of the KM1 CER Career Development Program at Washington University in St. Louis, Missouri, which trains and mentors researchers in the emerging field of CER. This CER training program provides hands-on experience in the use of clinical research, administrative, and electronic health records data to investigators from diverse backgrounds; develops and implements new CER curricula, classes, and short-term training toward master of science degrees; creates new CER scholar positions in the Clinical Research Training Center at Washington University to support postdoctoral fellows and junior and mid-career faculty focused on CER; and establishes a comprehensive system to evaluate the CER Career Development Program. In addition, the program has promoted CER by establishing a CER-focused website (http://cer.wustl.edu), a seminar series, and an annual symposium. To support future training in CER, Washington University will be a mentor institution for other institutions interested in developing CER training programs. Ultimately, these efforts will help develop a sustained community of CER researchers and mentors.6
Contributing NIH Institutes & Centers
- National Cancer Institute (NCI)