A statistical hub
Karen Emmons, PhD (left), and Glorian Sorensen, PhD, MPH (center), are leading a health-improvement study for underserved populations. Elizabeth Barbeau, ScD, MPH (right), is investigating whether certain groups are at heightened risk for hazards in the workplace.
Before data can be collected, investigators need to decide what questions to ask and what information can help supply the answers. "Our job is to provide scientific coordination and direction for this project — to help the member centers focus their research and find the best way of going about it," explains Paul Catalano, ScD, the Institute's associate chair of Biostatistical Science, who co-wrote the grant application with Harrington. Mary Beth Landrum, PhD, and Alan Zaslavsky, PhD, statisticians from the Depart-ment of Health Care Policy at Harvard Medical School, are collaborating on the project.
"The first task is to determine what information is relevant to the study: How do we measure the success of a particular treatment? How do we take account of other diseases that patients might have? How do we factor in patients' ages?" Catalano continues. "Then we decide what kind of tools we'll need — questionnaires, surveys, forms, and charts. Do the tried-and-true tools meet our needs, or should we create new ones? Before we use any of these instruments, we'll have to test them."
"By talking to a diverse group of workers, we'll be able to learn more about social inequalities and health and how to reduce them."
— Elizabeth Barbeau, ScD, MPH
Because the six collection sites wrote their own grant applications, their proposed goals, methods, and data-processing systems are apt to be somewhat different. Meetings to iron out those differences — a process Catalano likens to "rewriting the entire grant" — are under way, with Dana-Farber researchers taking the lead. "As the central data repository for this project, we need to make sure there are consistent standards for gathering, recording, and reporting patient information," notes Catalano.
While the study probes racial and ethnic disparities in cancer care and survival, researchers will be looking for differences in other categories as well. "Are elderly patients provided the same state-of-the-art care as younger patients, and are they accepting of such care? If not, why not?" Catalano asks. "Are there economic, demographic, or educational factors that influence treatment? These are the kind of issues the data may help us resolve."
"Are there economic, demographic, or educational factors that influence treatment? These are the kinds of issues the data may help us resolve."
— Paul Catalano, ScD
Another Dana-Farber investigator participating in the study is Jane Weeks, MD, MSc, director of the Institute's Center for Outcomes and Policy Research (See related story, Analyzing outcomes). She is heading up one of the six data collection sites — a national network of HMOs — to look for inequities in the outcomes of cancer treatment. One of her main interests is how racial or ethnic factors might bias the care patients receive.
"Some of the centers in the study treat large numbers of African-Americans," she says, while others see sizable populations of Asians and Pacific Islanders. "The data we collect should contribute to our knowledge of how good the quality of cancer care is — and how it varies for patients of different ethnicities."
Data gathering for the study is scheduled to begin early next year and continue until 2005.
- Next: New directions
- Page: 1 | 2 | 3
