REINEKE FORD   ||   NEWS UPDATES

Arlington native’s research helps estimate cancer risk

By SARA ARTHURS
Staff Writer
An Arlington native who is a physician and researcher at the Cleveland Clinic has developed a new method of determining the risk of colorectal cancer, which he hopes will help doctors better assess who needs to be screened for the disease.
The tool, called CRC-PRO, is designed to allow physicians to quickly and accurately predict an individual’s risk of colorectal cancer. Details of the project were recently published in the Journal of the American Board of Family Medicine.
Dr. Brian Wells is an associate staff member in quantitative health sciences at the Cleveland Clinic. He is also a family physician who practices one day a week.
Wells graduated from Arlington High School in 1990 and completed his undergraduate education at Ohio State University, followed by medical school at Wright State University. He spent five years in South Carolina, where he completed residency and training, then came to the Cleveland Clinic in 2004.
As part of his thesis for his doctorate in epidemiology at Case Western Reserve University, Wells analyzed data from a study on colorectal cancer completed at the University of Hawaii.
Wells led other researchers in analyzing data on more than 180,000 patients who were followed for up to 11.5 years to determine which factors were highly associated with the development of colorectal cancer. Factors such as dietary and exercise habits were examined, and the researchers looked at which patients ended up developing colorectal cancer.
Wells then analyzed the data to create two CRC-PRO calculators, one for men and one for women. He called them “another tool in our arsenal.”
The Hawaiian study did a better job than previous studies of including people of multiple ethnic backgrounds, he said. This is important as certain groups such as African-Americans and people of Japanese descent are at a much higher risk of colon cancer, but some earlier studies of colon cancer were conducted primarily with white patients.
Wells put in all the variables that he thought might be important in predicting a person’s risk.
Currently the guidelines for screening are “essentially an age cutoff,” which recommend that people get screening starting at age 50. Wells said that’s a crude way to determine who needs screening.
“There’s many other factors besides just age that affect your risk,” he said.
The screening for colorectal cancer is mostly done through a colonoscopy. Wells said there can be a slight risk of complications, and in addition the procedure is expensive and may not be widely enough available. Also “very few people actually do it when they should,” he said. All of these factors mean it would be helpful to determine who is most at need of screening.
Wells said doctors may recommend screening high-risk patients earlier than age 50 but delay or forego screening in low-risk individuals.
In addition to improving the efficiency of colorectal cancer screening, the hope is that these tools will help lower health care costs by cutting down on unnecessary testing.
People who are older, who smoke and who are overweight are most likely to be at risk, Wells said. He said behaviors like exercising, avoiding too much alcohol and not smoking, ‘just being healthy in general,” can help prevent cancer.
Wells said that although the calculator can identify who is most at risk it does not predict cause and effect. That is, his research shows that anti-inflammatory medicines like aspirin may make a person less likely to get cancer but this does not mean that everyone should start taking aspirin, as it can have side effects. When it comes to preventing cancer, it’s best to talk to one’s physician, he said. The risk calculator “is best used by physicians with patients at a visit,” he said.
In addition, CRC-PRO is “not designed for certain high-risk people” such as those who have a history of colon cancer, a hereditary cancer syndrome in their family, or inflammatory bowel disease.
Wells said there’s no particular group that should definitely not get a colonoscopy but there is a trade-off in risks and benefits. Someone who has a history of gastrointestinal bleeding or who has a shortened life expectancy because of age or other health problems may not find a colonoscopy beneficial, he said.
Almost all colon cancers start with benign polyps that become cancerous 10 to 20 years after they first appear, Wells said. So, someone who has a shortened life expectancy may not benefit from colon cancer screening.
Wells’ work involves quantitative science. He said many decisions in medicine are made “looking at one factor in isolation.”
“We think that you can make much more efficient decisions by considering all the important factors simultaneously,” he said.
A computer or mathematical model can do this work, he said.
Wells sees the future of medicine moving toward more tools like this calculator.
The increased use of electronic health records is creating a push toward these methods, he said. Doctors are sometimes hesitant to use tools that could be time-consuming but “if we can start to integrate these tools with the electronic health record,” the calculations of a person’s risk could be done automatically in the background.
It’s a struggle, however, “getting physicians to accept that the computer is smarter than they are,” Wells said.
Wells and his colleagues are providing their work for free for other researchers who may wish to build their own calculators on other medical conditions. The tool automatically creates the interface and is designed to be streamlined with electronic health records.
He is also involved in a project designed to calculate the risk of heart disease in someone who doesn’t already have it. Some of this research includes recording visits between patients and physicians. Wells said research in this field is often changing, such as guidelines that just came out that change the recommendations on who should be on the cholesterol-lowering medications known as statins.
Arthurs: 419-427-8494 Send an E-mail to Sara Arthurs

Comments

comments

About the Author