Blocked Bowel

The Press headlines this morning claim the Ministry of Health is blocking bowel screening tests. The article itself states the MoH does “not recommend” use of do-it-yourself bowel screening kits.  I can find no MoH press release relating to these kits, so I am a little confused as to whether or not the MoH is blocking anything at all.  This uncertainty aside, the issue has arisen because of the availability of bowel cancer testing kits which are self administered in the home and made available by the charity and advocacy group Bowel Cancer New Zealand/Beat Bowel Cancer Aotearoa (BBCA). There is evidently frustration within this group with respect to the government’s progress towards a bowel cancer screening program in New Zealand (there is currently a pilot underway).  My interest is not the politics, but merely to inform about the test.

The test being offered as best as I can make out is from Clinical Genomics and is their “ColVantage Home” test.  In the information on their website they give two numbers derived from trials which I believe are critical that any potential user of the test understand before they decide to use the test (indeed, not just for this test).  Those numbers are the sensitivity of the test, given as 86%, and the specificity given at 96-98%.  I expect to most those numbers are meaningless, and even to the scientist they are “minimal” in the sense no indication of how accurate these numbers are given (think “margin of errors” given in political polls).  However, let me take the numbers as gospel and let you know what they mean.

The False Positive.  I'd love to acknowledge the source, but I can't locate them.

The False Negative. I’d love to acknowledge the source, but I can’t locate it.

First, a sensitivity of 86%.  This simply means that for every 100 people who actually have bowel cancer, 86 will test positive with the home kit.  The very very important fact to note here is that a negative test does not mean you definitely do not have bowel cancer.  The 14 missed cases are called False Negatives, or Type II errors and are beautifully illustrated by this classic infographic of the doctor failing to diagnose pregnancy.  My concern with False Negatives and a “home kit” is that some people may feel falsely reassured with a negative test.  On the other hand, a test administered by a medical doctor is not simply the test, but exposure to the doctor who may see other signs or symptoms that the test doesn’t pick up.




False positive

False positive

On the other hand a 96-98% specificity means that for every 100 people who do not have bowel cancer 2 to 4 will test positive.  That is, a positive test does not mean you definitely have bowel cancer. The missed 2 to 4 cases are called False Positives, or Type I errors as illustrated by the diagnosis of the man in the image.

Another way of thinking about this is that all of us before a test have a certain probability of having bowel cancer.  We may get an idea of what that probability is by knowing the rates of bowel cancer amongst people like us.  In New Zealand the rates are very high – some of the highest in the world at about 0.3% or 300 per 100,000 adults.  However, risk is higher as people age, hence screening programs are targeted at older adults (over 50s in this case).  There are other risk factors.  Check out the risk calculator here.  A positive test essentially increases your risk, but not to 100%, and a negative test, decreases it, but not to 0%.  If it was me, and I was worried enough to want to have a test because of increased risk factors, I’d go and ask my GP for one.  On the other hand, if I didn’t want my wife to know (duh!) and got a kit, and it was positive, I’d also head down to the GP.

As I stated before – beware of being falsely reassured by a negative test.  However, there is also a major issue with false positives which bedevil any screening program, and even more so one in which the most at risk population are not being targeted (in this case there is nothing stopping 30 year olds having tests).  This requires a bit of maths.  Let’s pretend for a minute that the tests are only being done by people in a population with a very high prevalence of bowel cancer – say 1%.  That means that for every 10,000 people tested 100 will have bowel cancer for whom 86 the test will be positive.   Of the 9,900 who do not have bowel cancer, 198 to 396 will also test positive – ie falsely positive.  ie, in total, 284 to 480 people will test positive.  ie 22% to 30% of those who test positive actually have bowel cancer.  For the individual who started knowing they had (say) a 1% risk, they now know they have a 22 to 30% risk.  Certainly worth checking out more.

Of course, if people have these tests regularly, the proportion of people who end up having a false positive test at some time will increase (along with the proportion of people with bowel cancer who have a positive test).  ie repeat testing will increase sensitivity and decrease specificity.

Finally, if you have a history of bowel cancer in your family then don’t hesitate to ask your GP for tests and advice on what you can do to minimise your risk through improving your life style.


Got something to say? Don't be shy

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s