Family medical histories: How accurate are they and what information might be important to know?

When someone in the family has breast cancer, it is likely that many of the other adults in the family will know about it. But this does not hold true for all types of cancer. If a family member has developed bowel cancer, for example, it is quite common for the other family members to not know anything about it. It is not clear whether or not preparing a cancer family medical history will improve health outcomes, nor is it apparent what the best way is to collect the information.

It is a common question from doctors and in disease risk-assessment questionnaires: “Does any close family member have breast cancer? Heart disease? Diabetes?” This question is part of taking down family background information on a disease, what is called a family medical history. For some diseases, such as breast cancer and bowel (colon or colorectal) cancer, your risk of getting the disease is definitely known to be higher if a close relative has it.

But research results are much less clear for many other diseases. And in fact, there is no standard or universal way of collecting this information. While some of this information might be very important, some of it could be misleading or even harmful. For example, if the information is wrong, this means you could be over-estimating or under-estimating your risk of the disease. Collecting this information could be distressing for some family members.

Some people develop family trees of major illnesses in their families, and the information can be very important when there are inherited diseases in the family. How far these trees should go, though, and whether they are needed for more common diseases like diabetes or heart disease, is unclear.

Researchers from McMaster University in Canada were commissioned by the U.S. Agency for Healthcare Research and Quality to analyze the research on collecting family medical histories for certain types of cancer. They looked for studies that examined a number of questions, including how accurate family cancer histories are and what information could be helpful to collect when you need to know your own medical history.

Some of what they found could be relevant to the collection of family medical histories for other diseases. For example, they clarified some definitions. Close family, then, are first and second degree relatives. First degree relatives are people who share one-half of their genetic information with you. Or to use a more common way of saying it: you have at least half the same blood as they do. Your first degree relatives include

• your full siblings (brothers and sisters),
• your birth parents, and
• your birth children.

Second degree relatives are those you share one-quarter of the same genetic information with. This means that your second degree relatives include

• your grandparents (the birth parents of your birth parents),
• your grandchildren (birth children of your birth children),
• your blood-related uncles and aunts, and
• your half-siblings.

This is called a “three-generation family tree”: your generation plus one generation above and below you. The researchers concluded that information higher or lower than this becomes less and less relevant. Especially as our communities become more mobile and family members lose touch with each other, with each step further away, the information reported during collection of family histories might get less accurate: they can no longer say for sure what diseases their relatives have had.

The core information items in such a family tree include the age or year of birth of each person, their ethnic background, as well as the cause and age of death for those who have died. In case of cancer, the age the person was when the cancer was diagnosed is important. The age of diagnosis is less important for some other diseases.

Accuracy of family histories of cancer

One of the central questions that the researchers from McMaster University asked was: “What is the evidence that patients or members of the public accurately know and are able to report their family history?” They found 19 studies that met their requirements for being an adequate study of the accuracy of this knowledge. They looked at a specific group of cancers, all believed to have an increased familial risk: breast cancer, bowel cancer, ovarian cancer and prostate cancer.

The family history information was collected from individuals by telephone or personal interview, or the person in the study had completed a form. Most of these people had cancer themselves. Accuracy was tested against medical records, other family informants and/or cancer registry data. One of the problems the researchers identified was the difficulty of being absolutely certain what “the truth” was because medical records and even death registries can also be inaccurate.

However, they were able to draw some cautious conclusions. One was that family medical histories on these types of cancers were often quite inaccurate when family members were not pressed with further questioning. Also, the likelihood that family members know about someone in the family having cancer depends on the type of cancer. Reporting of breast cancer was especially accurate. But when a family member had bowel cancer, other relatives often knew nothing about it.

The researchers also identified 40 studies that examined paper- or computer-based tools for collecting family cancer histories. These trials also looked at how these different instruments assigned people a low, moderate or high risk of cancer.

They were looking specifically for tools such as questionnaires or family tree programs that might be relevant to general or family practice, that is, people gathering information to help them make decisions about screening tests, for example. Most of these tools were forms or programs to be used by the patients themselves however.

The researchers concluded that the accuracy of patients doing these is similar to the accuracy of history-taking by a doctor: that is, a doctor is not necessarily more accurate than a patient. But there are many unanswered questions about the effects of these tools. Until more research is done, we cannot know if the cancer family history tools assessed are empowering or harmful to patients and family members, or whether they improve or worsen health outcomes.

© IQWiG (Institute for Quality and Efficiency in Health Care)