A $1 billion funding windfall has brought newfound attention to comparative effectiveness research (CER) and broadened career opportunities in this growing field. The CER movement, which centers on the need to contain health care costs while maintaining quality, combines mathematical precision with political sensitivities to evaluate how medical treatments work in the real world. "CER is focused on actual choices that patients are having to face," says Kathryn McDonald, executive director of the Center for Health Policy at Stanford University.
"The broader need here is for these sorts of data to help us make reasonable decisions about how we provide health care. So it seems to me that [comparative effectiveness research] is a field whose time has come." --Scott Gazelle
Whereas clinical research is structured to meet federal regulatory requirements, CER is focused mainly on what happens after regulatory approval, McDonald says. For example, a clinical trial might determine whether a particular treatment for prostate cancer is safe and effective; a CER study might also consider cost and side effects such as frequency of urinary incontinence.
Scientists working in CER say the field is beginning to establish its own identity instead of being shoe-horned into areas such as health services research and public health. The influx of funding, courtesy of the American Recovery and Reinvestment Act, will help to solidify that identity while also creating opportunities for researchers interested in an area that embraces uncertainty and values ethics alongside data. The size of the government's commitment, and its importance for the future of a ballooning health care system, suggests that CER is here to stay.
"If there's anything faddish it's the name 'CER,' and the name may change," says Peter Neumann, director of the Center for the Evaluation of Value and Risk in Health at Tufts Medical Center, Boston. "But I don't think what's going on underneath it is a fad, and I don't think it's going away."
Igniting the spark
The science of generating and reviewing medical evidence is evolving, Neumann says. There is a lot of room for new methods to extract valuable information from the volumes of raw data generated within the medical research enterprise. This is what the new generation of CER research will do.
A new generation of researchers who think about how best to measure variables such as patient preferences and quality-of-life issues is coming of age now, McDonald adds. These rising stars in CER come from different fields, but all were motivated to enter the field by a passion for translating academic research into real-world decision-making.
"I always had an inherent interest in economics and its relationship to the well-being of people," says Ana Johnson, Canada research chair in health policy at Queens University in Kingston. A bachelor's degree in economics led to her graduate work in health economics at the University of Texas Health Science Center at Houston.
Johnson's research interests span the gap between CER and public policy. As a member of the Ontario Health Technology Advisory Committee, she advises the province about which nondrug medical technologies provide the most benefit for the cost. She and her colleagues recently published the decision-tree model, developed in Ontario, which combines science, economics, ethics, and public engagement in the decision-making process. She hopes that other municipalities -- or, in the United States, insurers -- will use their method to decide which medical treatments to pay for. Johnson's research integrates psychology, decision theory, and CER data with an eye toward public benefit.
"I am very interested in the interface between these comparative effectiveness results and then what is done with these results," she says. "I don't wish for them to just sit in a journal somewhere."
Johnson says that she believes the frontier of CER lies in developing new methods to extract useful information from large databases that were not originally designed with CER in mind. An example: Johnson uses Bayesian statistics, a statistical formalism that allows researchers to modify their hypotheses as data are collected, to tease out small differences in a therapy's effectiveness from clinical trials designed with the traditional accept-or-reject approach.
Defining comparative effectiveness research
Last year, a panel of experts came up with a draft definition for comparative effectiveness research. Here's the first line: "Comparative effectiveness research is the conduct and synthesis of systematic research comparing different interventions and strategies to prevent, diagnose, treat and monitor health conditions."
- CER researchers come from a range of professional disciplines including clinical medicine, epidemiology, bioinformatics, biostatistics, health services research, economics, methods research, decision and cognitive sciences, genomics, proteomics, library science, communications, as well as other areas. They may have medical or other clinical degrees, doctoral degrees in public health specialties, specific training in systematic reviews and clinical trials, and/or post-doctoral or master's-level training.
- The CER workforce needs individuals with expertise in designing and conducting clinical trials, statistical modeling, conducting systematic reviews and meta-analysis, quasi-experimental design and other observational methods, use and analysis of large datasets, cost-effectiveness analysis, clinical prediction rules, measurement of patient-reported and clinical outcomes, and communicating research findings to patients, providers and others.
Sources: Draft Definition of Comparative Effectiveness Research, Federal Coordinating Council for Comparative Effectiveness Research; Initial National Priorities for Comparative Effectiveness Research, Institute of Medicine.
Breaking from tradition
While Johnson works to extract data from traditional clinical trials, Anirban Basu, a health economist at the University of Chicago in Illinois, focuses on a new kind of clinical trial. "Adaptive" clinical trials allow investigators to modify who is treated with a particular drug or device on the fly based on who responds to the treatment and who doesn't. Often such trials include genetic markers for drugs that target particular pathways, such as a kinase inhibitor for cancer treatment. If patients who respond well have a particular molecular signature, investigators can include more people with those markers in the trial and stop treatment for those who don't respond well. This approach could supplement the traditional questions explored during clinical trials -- Does it work? Is it safe? -- by asking who does it work for, and who is it safe for?
Basu's research focuses on establishing the value of this personalized approach to medicine, which he believes, and his research has shown, will be more cost-effective and valuable long term than the current approach to clinical trials. Toward that end, Basu and his colleagues have developed methods to calculate the benefit of individualized treatment versus the one-size-fits-all approach.
The future of personalized medicine, Basu believes, depends on an influx of researchers who can think beyond traditional research methods. "We need a cohort of people who are not tied to old methods but will really think of new methods in this field," he says.
Indeed, CER seems to attract people who see a need for change in the current system. A radiologist by training, G. Scott Gazelle was drawn into CER over frustration at not knowing which technologies were right for his patients. During a postgraduate medical fellowship, he developed radio-frequency ablation techniques to remove cancerous tumors and he wanted to know how well his technique compared with traditional surgery. As he got involved in comparative studies, he began to see a need for new, systematic methods for evaluating new medical technologies.
So in 1997, he founded what is now known as the Institute for Technology Assessment (ITA) at Massachusetts General Hospital (MGH) in Boston to do these types of evaluations. Researchers at MGH ITA conduct studies in clinical economics, cost-effectiveness, and quality of life issues for diseases such as cancer and cardiovascular disease. They do studies for the federal government and other organizations, such as the World Health Organization. "The broader need here is for these sorts of data to help us make reasonable decisions about how we provide health care," Gazelle says. "So it seems to me that this is a field whose time has come."
CER professionals work mainly at academic institutions, but increasingly they are employed by the pharmaceutical industry, clinical research organizations, and insurers, McDonald says. Even states are starting to do CER studies to maximize value in Medicare and Medicaid spending, she says, pointing to California as an example.
Because CER is still coming into its own as a field, very few departments and programs do true CER. Many of the National Institutes of Health - sponsored Clinical and Translational Science Award programs report having a CER component, but a recent survey of CTSA awardees revealed that about half of the 33 institutes that responded had complete courses that encompass the most common CER components such as health economics and health informatics.
Basu recommends that anyone who is interested in CER take coursework in epidemiology, biostatistics, and research methods. His background in pharmacy didn't have a strong quantitative element, he says, so he took extra coursework in statistics then got a Ph.D. in public policy. The key to a career in CER is a diversity of training, he adds.
Stanford's McDonald also notes that diverse training is important in CER, but a lot of the research happens in teams with a diversity of expertise. "There are a lot of nuances to each clinical question, so sometimes it's best to have a sub-specialist clinician who's trained in these kinds of methods do the research, but ultimately it's teamwork," says McDonald, who is also president of the Society for Medical Decision Making, a professional group that includes many who conduct CER studies. Often such teams include clinicians, statisticians, computer programmers, and data analysts.
With the recent emphasis on translational and outcomes research, it should not be difficult to find a training program in fields related to CER, says Milton Weinstein, director of Harvard University's Program on the Economic Evaluation of Medical Technology and co-author of the book Decision Making in Health and Medicine: Integrating Evidence and Values. The Agency for Healthcare Research and Quality maintains a list of its fellowships in health services research; and the National Institutes of Health, which, along with AHRQ, is a major benefactor of the federal money given to CER, plans to fund new training programs in the field. The Society for Medical Decision Making maintains an international list of medical decision-making training programs.
Gazelle recommends that Ph.D. students in the traditional sciences look for a postdoctoral fellowship in health services research or perhaps think about getting a master's degree in public health. For medical students, he says quantitative training is key.
"You can't just learn about this in a classroom or by reading a book," Gazelle says. "The best opportunities are going to come by going to places where there are fellowship training programs or where you can join a team, bring whatever skills you have, and learn new skills."
Karyn Hede is a freelance writer in Chapel Hill, North Carolina.