This was an interesting week, first we had a department grand rounds on Personalized Healthcare including a demonstration of the right way to take a Family history and then we had a Medical Informatics grand rounds on Practical Proteomics.
Some take home points and highlights from this week:
A routine patient interview could go something like this:
Dr. Jomes: Ms Smith, lets talk about any family history of medical problems. Did anyone in your family have any cancers?
Ms. Smith: Oh yes, my dad had colon cancer and my aunt had breast cancer.
Dr. Jones: Is that your paternal aunt?
Ms. Smith: No that was my mom's sister.
Dr. Jones: I see. Ok any other cancers like that of the ovary or uterus or lung?
Ms. Smith: Not that I know of.
And then Dr. Jones would go on to cover other conditions like Diabetes, CAD etc.
The summary note would something like this:
Pt with family h/o colon ca in father and breast ca in aunt.
You can see the process is quite haphazard and focuses on conditions rather than the family tree and could easily miss early deaths, critical conditions and does not give an overall picture that can help give us clues to heritable conditions.
So what should the ideal process look like?
Dr. Jones: So Ms. Smith, lets talk about your family history. Do you have any brothers or sisters?
Ms Smith: Yes, I have a brother who is 45
Dr. Jones: And does he have any medical concerns?
Ms. Smith: No he is doing well.
Dr. Jones: Do you have any children?
Ms. Smith: Yes, a son, he is 14 and a daughter she is 12 and they are well.
Dr. Jones: Good. Does your brother have any children?
Ms. Smith: Yes he has 2 sons and they are well.
Dr. Jones: Now tell me about your parents
Ms. Smith: My dad is 65 and he had colon cancer and my mom is 62 and she has high blood pressure and high cholesterol.
Dr. Jones: Tell me about your father's side of the family. Does he have brothers or sisters?
After completing the father's side, Dr. Jones would probe into the mother's side of the family. The key difference is asking about sibs, children and parents and then asking about their health; and probing on one side of the family and then moving to the other side.
Ideally, Dr. Jones would be creating a family tree as he goes along.
This is an example of what it could look like in a family with Lynch syndrome (Hereditary non-polyposis colorectal cancer):
What should lead one to suspect this syndrome?
There are well defined Amsterdam criteria which require early onset of colon cancer, successive generations with colon cancer and multiple family members with colon cancer. A simple intervention like creating a family tree can help lead to diagnosing this with genetic testing and could impact the lives of multiple members of the family.
So what are the barriers?
Patients should create family trees before their office visit using some simple genealogy tools like Geni. They could then print out the family tree and write in what they know of the medical histories of their family.
A tool that could be used for this on a handheld device like an Android tablet or an iPad would also be very helpful. Patients could use this in the waiting room. There should be a way of importing this into the EHR.
Some take home points and highlights from this week:
- Lynch Syndrome is relatively common and very much underdiagnosed.
- "Old" docs can still learn new tricks on how to take a good family history!
- Doing this the right way (see below) could significantly impact our patients' lives
- The lab mouse and humans have the same number of genes. The complexity of the phenotype is determined not by the genes but the proteins these genes code for.
- Thus our Genome is just a script but the proteome is the actor that makes or breaks the show.
A routine patient interview could go something like this:
Dr. Jomes: Ms Smith, lets talk about any family history of medical problems. Did anyone in your family have any cancers?
Ms. Smith: Oh yes, my dad had colon cancer and my aunt had breast cancer.
Dr. Jones: Is that your paternal aunt?
Ms. Smith: No that was my mom's sister.
Dr. Jones: I see. Ok any other cancers like that of the ovary or uterus or lung?
Ms. Smith: Not that I know of.
And then Dr. Jones would go on to cover other conditions like Diabetes, CAD etc.
The summary note would something like this:
Pt with family h/o colon ca in father and breast ca in aunt.
You can see the process is quite haphazard and focuses on conditions rather than the family tree and could easily miss early deaths, critical conditions and does not give an overall picture that can help give us clues to heritable conditions.
So what should the ideal process look like?
Dr. Jones: So Ms. Smith, lets talk about your family history. Do you have any brothers or sisters?
Ms Smith: Yes, I have a brother who is 45
Dr. Jones: And does he have any medical concerns?
Ms. Smith: No he is doing well.
Dr. Jones: Do you have any children?
Ms. Smith: Yes, a son, he is 14 and a daughter she is 12 and they are well.
Dr. Jones: Good. Does your brother have any children?
Ms. Smith: Yes he has 2 sons and they are well.
Dr. Jones: Now tell me about your parents
Ms. Smith: My dad is 65 and he had colon cancer and my mom is 62 and she has high blood pressure and high cholesterol.
Dr. Jones: Tell me about your father's side of the family. Does he have brothers or sisters?
After completing the father's side, Dr. Jones would probe into the mother's side of the family. The key difference is asking about sibs, children and parents and then asking about their health; and probing on one side of the family and then moving to the other side.
Ideally, Dr. Jones would be creating a family tree as he goes along.
This is an example of what it could look like in a family with Lynch syndrome (Hereditary non-polyposis colorectal cancer):
 | |||
| (From David G. Jagelman Inherited Colo-rectal Cancer Registries at the Cleveland Clinic) |
There are well defined Amsterdam criteria which require early onset of colon cancer, successive generations with colon cancer and multiple family members with colon cancer. A simple intervention like creating a family tree can help lead to diagnosing this with genetic testing and could impact the lives of multiple members of the family.
So what are the barriers?
- Time: It could take a good 10-15 minutes to create a simple family tree and longer for large families with multiple conditions. In the context of the 15-20 minute office visits this is challenging.
- Lack of specific electronic tools that support doing this within the EHR
- Patients often may not know details of their family members' medical history.
- Family members may not want to divulge their medical details.
Patients should create family trees before their office visit using some simple genealogy tools like Geni. They could then print out the family tree and write in what they know of the medical histories of their family.
A tool that could be used for this on a handheld device like an Android tablet or an iPad would also be very helpful. Patients could use this in the waiting room. There should be a way of importing this into the EHR.