Skip to main content
Screening Guidelines May Fail to Catch 10 Percent of Colon Cancers

You are listening to Health Library:

Screening Guidelines May Fail to Catch 10 Percent of Colon Cancers

Nov 22, 2013

Utah’s unique resources of genealogy and cancer data, combined with records from two major health care organizations, made it possible for researchers to determine that the current screening guidelines allow a large number of colon cancer cases to go undetected. Dr. N. Jewel Samadder, of the Huntsman Cancer Institute, discusses how the findings can affect colon cancer screening guidelines and explains that because of the state’s unique storehouses of genealogy and medical records, a study covering such a broad spectrum of the population couldn’t have happened anywhere but Utah. Samadder also discusses the future of combining records from multiple sources to provide consumers more accurate health care information.

Episode Transcript

Announcer: Medical news and research from University of Utah physicians and specialists you can use for a happier and healthier life. You're listening to The Scope.

Host: Colon cancer screening guidelines might fail to catch 10% of colon cancers according to some recent research, but that's only part of this story. What makes this story so unique is how a wide array of very unique data was combined in a way that has not been done before in a study of this magnitude. We're here with Dr. N. Jewel Samadder, gastroenterologist at Huntsman Cancer Institute. Tell us a little bit about this study and what made it so unique.

Dr. Samadder: Our study looked at 127,000 patients who underwent colonoscopy at the University of Utah Health Sciences system and Intermountain Healthcare. These two health systems account for roughly 85% of all health care in the state of Utah. We were able to look at the colonoscopy records and identify those patients who had polyps found at colonoscopy as well as patients who did not have polyps found at colonoscopy, and by using the unique genealogical records that exist in the state of Utah and the linkage with the Utah Cancer Registry, the statewide cancer registry, look at what is the risk of polyps and colorectal cancer in the relatives of patients who are found to have polyps at colonoscopy.
In most states where data like this exists it exists in separate silos that you can not easily or ever combine together. However, in Utah we're very unique. One, we have two major health systems that account for the majority of health care in the state. Also, we have a statewide cancer registry, the Utah Cancer Registry, which has been part of the SEER network of registries since 1973. State law requires that all cancer diagnoses, including colorectal cancer, from any hospital or health system in this state is reported to the Utah Cancer Registry.
Third, we have genealogical records that are also linked into the Utah Population Database. Finally, we have a unique ability to link all electronic medical records from the University of Utah Health Sciences system and Intermountain Healthcare into the Utah Population Database and thereby into these genealogical records and the cancer registry records.

Interviewer: Truly, if I'm getting this right, there's no other place that this could've happened other than here in Utah at this moment in time.

Dr. Samadder: Exactly. There is no other U.S. state or province in Canada that could do this type of research. The closest thing that comes to this is there are a few Scandinavian countries that have genealogical records, but they're much more limited and really only in first degree relatives and can not go out to more distant relatives.

Interviewer: This type of data, how does it change your trust in the results that you get from it?

Dr. Samadder: That's a great question. One of the important things when we look at such large population base studies is to look at does it represent all of the patients or the majority of patients we would see. Well, our patient population we used here is cared for by the two health systems that account for a majority of the state. We have 127,000 patients nearly in our colonoscopy file that we were able to look at. Again, that is a broad cross-section of the state of Utah, versus limiting it to a single hospital or a single region, a single urban center for example.
A second thing is many studies that have tried to examine this in the past have been limited by patient recall of family history data. By using the genealogical records we take those biases all out of the picture.

Interviewer: So, if I was to ask how confident are you in these results?

Dr. Samadder: We are extremely confident in the results.

Interviewer: That's got to feel good to be able to say that when it comes to doing a study like this.

Dr. Samadder: Yes, yes. I think it has the potential to really inform screening guidelines and colon cancer prevention with the use of colonoscopy and appropriately targeting populations that are at higher risk, those with a family history of polyps or colon cancer.

Interviewer: You take a look at currently what the screening guidelines are. You put that up next to your research. How does that change things?

Dr. Samadder: We still recommend the current screening guidelines, that people with no risk factor everyone should undergo some type of colon cancer screening at age 50. The current preferred screening option is a colonoscopy, and that should be repeated every ten years.
People who have a risk factor such as a family history of colon cancer or polyps at a younger age in their close relative, below age 60, start colonoscopy at age 40 or 10 years before the earliest advanced polyp or colon cancer in that family, and repeat more frequently - every 5 years instead of every 10 years.
However, the guidelines currently say that if your relative developed colon cancer or advanced polyps over age 60 you can start at age 50 which is the standard guidelines. Our evidence and study would support maybe considering starting those people at an earlier age as well such as age 40.

Interviewer: How do you even go about changing the guidelines, then? If this is something you truly believe in, you truly believe your research is solid, what's the next step for you in this process?

Dr. Samadder: I think the next step in this process is generating more data. We have now also looked at colorectal cancer in this colonoscopy population, and the risk of adenomas or polyps and colorectal cancer in their relatives and showing that the risk, again, is increased regardless of the age in which the initial colon cancer is diagnosed.
I think when that study is also published that'll provide more fuel for us to then approach the different screening guideline organizations and say we have evidence from multiple studies in our population that support this increased risk of polyps and colon cancer in relatives regardless of what age the initial patient developed advanced polyps or colon cancer at.

Interviewer: What kind of man hours work went into bringing all this data together? Paint the picture for me. How many computers did you need? How many people did you need looking at it? How Herculean of a task was it?

Dr. Samadder: This was really a collaborative venture between the Huntsman Cancer Institute, the Utah Population Database, and Intermountain Healthcare. It was having staff at all three sites that allowed us to do this project.
For example, we had programmers at Intermountain Healthcare who identified all of the colonoscopies done over this time frame of 1995 to 2009 generating an electronic list from that. That was securely transmitted to the Utah Population Database with the identified medical record information and then merged with data from the University of Utah Health Sciences system, where a programmer also similarly in parallel extracted all colonoscopy records and securely transmitted it to the Utah population database. Finally, that data was then linked with the genealogical records and the Utah Cancer Registry records.
In terms of manpower or person power of effort, this is 10 to 15 people working at all 3 sites putting in hundreds of hours of work in programming time just to get all of these databases linked. Then, thinking of the question and analyzing the data is also hundreds of hours of programming time to do the statistical analysis.

Interviewer: It's not just as easy as taking some data from here, taking some data from there, putting it together, and seeing what you get.

Dr. Samadder: No, unfortunately not.

Interviewer: How much computing time did it take? Do you have any idea? Is that something that you're even aware of? Once they get all this stuff in and then they...

Dr. Samadder: It's not that long, actually. It's not like the old days where you have these supercomputers.

Interviewer: So that part's the easy part.

Dr. Samadder: That's the easy part.

Interviewer: The hard part was extracting what you need and then figuring out how do we get this database to talk to this database.

Dr. Samadder: Yeah, yeah. It's probably, like, 10 minutes, 20 minutes of actually the computer thinking it and doing it. That's not the limitation.

Interviewer: Yeah, yeah. For this type of research, what are the biggest challenges that need to be overcome to make it easier in the future so we can do more of this great research more easily?

Dr. Samadder: Probably the most important thing is what we have in Utah, these unique linkages that occur between our health system, our cancer registry, and the Utah Population Database to allow these data pieces to come together.
Why that's important is we want to look at the entire population. We want the data we generate to represent not just the care at a university hospital, not just the care at a managed care organization, not just the care at an urban center like Salt Lake City, but it can be expanded to generalize or summarize the type of care that's provided throughout the entire state of Utah in multiple different health care settings in multiple different urban and rural settings.

Interviewer: Your next study, how long do you think before that's going to be out?

Dr. Samadder: Our next study, looking at the risk of polyps and colorectal cancer in relatives of patients with colorectal cancer, is currently at a journal and hopefully will get published soon and will be the next one.

Announcer: We're your daily dose of science, conversation, medicine. This is The Scope, University of Utah Health Sciences Radio.