Skip Navigation

103 Issue 10 Garber Transcript

Kristine Crane: According to a JNCI study, patients often do not accurately report family cancer history. The Connecticut Family Health Study surveyed Connecticut residents about cancer diagnoses in first- and second-degree relatives with breast, prostate, lung and colorectal cancer. I talked with Dr. Judy Garber of the Dana Farber Cancer Institute, who co-authored an editorial on the study. So, let’s start by talking a little bit about the Connecticut Family Health Study.

Judy Garber: They looked at a study that included 1,019 participants, who were population-based and collected in 2001. And they asked about family history, and those people told them about more than 20,000 first- and second-degree relatives, so these are your closest relatives: parents, grandparents, siblings, children and aunts and uncles. So it’s reasonable to think that people-- the closest relatives are the ones you would expect people to know the most about. And they contacted a little over 10 percent of us to 2,600 people and asked them about their own history of breast, colon, prostate and lung cancer, so the four common cancers. And then they went back and tried to confirm what they were told, both by the people in the study and by the relatives, to see what could they find in medical records, so they looked at available databases to date: cancer registries, the National Death Index, death certificates, medical records that people gave permission for them to review. And then they said, “Now we know what the history actually is. How good were people in what they knew about their family?”

Kristine Crane: And what did they find?

Judy Garber: Well, what they showed was that most people really don’t know very much about their family, that they knew more about breast cancer and lung cancer than they did by far about colon cancer and even prostate cancer, so people were just not very accurate and they were more accurate for their closer relatives, their parents, siblings and children, as you would expect, than from more distant relatives, but they were not so great even at those. And I suppose this is not totally surprising. You would expect that people would know about breast cancer, and people were pretty good about breast cancer, not great but 80 percent; they were good, but they were not very accurate, particularly about prostate cancer or colon cancer. And the abdomen is always a challenge. Many people will tell you that relatives have stomach cancer, but they just don’t know which organ in the abdomen it was. So, basically what this says was that people knew accurately somewhere only between 30 and 60 percent accuracy in their report of common cancers in their close relatives. So that means when a primary care physician asks a patient for their cancer family history it’s not that much better than a flip of a coin whether they’re correct.

Kristine Crane: So have previous studies looked specifically at people known to have cancer in the family?

Judy Garber: They have. Studied have looked at cancer patients, for example, where you would expect people to have tried to figure out if other people in the family have the same cancer that I do, and they do a little bit better, but they don’t do much better as you might think. And even in studies in individuals with really remarkable family histories, people with inherited cancer syndrome, people are better but they are not very accurate in general at the specifics. So they may know that someone had breast cancer but they won’t know the age, and the rarer the tumor type, the less likely they are to get it correct.

Kristine Crane: Would the results also have looked differently a decade ago? Has there probably been some improvement in terms of what people are aware of?

Judy Garber: Well, we speculated in the editorial that people today might do a little better than this group, who was identified 10 years ago, because there’s more interaction potentially with extended family if you think that people are using electronic media or social media to contact their relatives. If nothing else, social media is used more by younger people and cancer is a disease of older people, so you might not keep up with that. There’s also been somewhat more emphasis in the public overall on knowing something about your family history. The surgeon general, the last one, had a big program for using Thanksgiving to collect family history, has a Web site for trying to record family history, so it’s possible that some of these things would have affected people’s accuracy, but I think it’s difficult to be sure.

Kristine Crane: And what about differences in terms of families’ socioeconomic status and race, ethnicity?

Judy Garber: Well, I think, as one might expect, that in families that have more complex structure, where parents have separated or divorced or some parts of the families are not in contact over time, where people have moved around a lot, that one would expect it to be even more difficult to have accurate information on relatives. In fact, I believe there are data that show that individuals more often are familiar with their maternal family histories than their paternal family histories, for obvious social reasons most of the time, so I think there could be some concern that there are socioeconomic differences because of the various socioeconomic pressures on families that could create circumstances, where it’s more difficult to know about relatives that are more distant and to maintain connections that would give you that information.

Kristine Crane: How relevant do you think that family history is going to become? What would be sort of the take-away message for the population at large and also specifically people with cancer in their family?

Judy Garber: Can you say that someday we’re going to be able to just test everybody and figure out what they have and who cares about family history? I think that there are two reasons to think that that’s not likely to be true, at least for quite a long time. One is that we still need family history to help us interpret certain kinds of genetic information, full sequencing that’s coming along with a thousand-dollar genome. You’re going to have your genes typed, but some of what you find is going to be easier to interpret if you know the context. That context may be the family history. And then the second reason that I think it’s not so clear is that not all cancer is about inheritance, so, I mean, in order to interpret which risk factors are most important. Whether they’re environmental or diet or genetic, you’re going to need to know how these genetic changes have affected perhaps other members the families so that you can try to sort out which elements are most important. I mean, these genetic risks are not all the same magnitude.

Kristine Crane: Is there ever the risk, then, of knowing too much, of having too much information at hand, I suppose if it’s not properly interpreted or put into context?

Judy Garber: I think it’s reasonable to think that families should have some idea of what goes on. Now, if you think that all of your relatives got lung cancer, perhaps that could help you realize that smoking could be a particularly bad idea for you, but it could backfire. You could have people who become totally terrified that they are destined to get a disease that family members have had, when, of course, inheritance doesn’t work that way; nothing is inevitable. You hope that people will use this for good. If you knew that everybody in the family had melanoma, hopefully that would mean that you would get yourself to a dermatologist or avoid the sun. I don’t think it really works that way, but that would be the-- you could hope that would be true.

Kristine Crane: That was Dr. Judy Garber, the co-director of the Friends of the Dana Farber Cancer Genetics and Prevention Program. For a transcript of this podcast or to listen to other JNCI podcasts, please visit our Web site at This is Kristine Crane. Thank you for listening.

#### End of Garber.mp3 ####