Yesterday quite a few people were puzzled as to why my doctors wanted me to wait four more weeks before going in for an MRI. The reason is that they are worried about false positives. Something far too few doctors seem to worry about. A false positive occurs when you have a test and it says you have some problem when in fact you do not. False positives are why testing seemingly healthy people for random illnesses is a bad idea. More than likely the test will claim there is problem when the person is fine than the other way around. If the treatment involves any risks you can easily do harm far more often than good. Note this does not apply to all tests at all ages. For example, it is a good idea for women to get breast cancer checks and men prostate cancer checks as they age. My doctors fear that the MRI will show "something." Given I had the test and it came up positive the next thing is to go in and find out what the something is. If it turns out the something is either nothing (a false positive) or a something that would have healed on its own, I am now a lot worse off. The surgery will have accomplished nothing other than cut up my leg and expose me to whatever risks the operation entails. The doctors look at it this way: If after another four weeks there is no progress and the MRI turns up something, then that vastly increases the odds the something is indeed something they should go in and try to repair. For those still reading and curious as to how this works here is an example. There is a test for a disease. If you have it, then the test will say you have it 90% of the time. Of course, this means 10% of the time it will not (a false negative). If you do not have the problem the test will still come up positive but only 5% of the time (a false positive). On the other hand it will be right 95% of the time. The way I have set this up it appears that if the test says you have the problem you should get treated. But should you? Suppose that if you have the disease and you do not get treated you would score the misery it imposes at 100. But if you are treated the misery index goes to 0 on average. On the other hand if you get treated and you did not have the disease there is some chance things will go wrong and your misery score will climb to 150 on average. Here we have 100 misery points from the untreated disease and another 50 from the problems imposed by the treatment's side effects. Now should you have the test and treatment if it says you are ill? Sorry you still cannot answer that question! The final bit of information you need is how likely are you to have the disease before you took the test. If you are a seemingly healthy suppose there is only a 1% chance you have it. Should you get tested? NO! To see why imagine you run this test on 100,000 people. Of those 99% do not have the disease or 99,000 of them. Of this group, 5% will generate a false positive or 4,950 of them. Now 1,000 of the people being tested are ill. This group will generate positive results 90% of the time. That means we will have 900 true positives. Given you have a positive result there are 900 chances out of 5,850 that you have the disease or about a 15% chance. From the figures above getting treated if you have the disease is worth 100 and getting treated when you do not costs you 50. In this example getting treated if you get a positive result therefore yields .15*100-.85*50 for an expected loss of 27.5. Here we have a pretty reliable test and yet getting treated given it tells you that you have a disease is a bad idea. Now suppose that you have some symptoms for the disease. (Like your leg is not functioning after about 7 weeks off from running ). Of those who have these symptoms suppose 40% have the disease. Now should you get tested? Yes! If you test 100,000 people with the symptoms 60,000 will be healthy and they will generate 3000 false positives. On the other side of the ledger, 40,000 will be ill and they will generate 36,000 true positives. Now given you get a positive test result the odds you have the disease are 36,000 out of 39,000 or a whopping 92% chance! Getting treated in this case is worth a LOT: .92*100-.08*50 or 88. To end this very long post this is the reason my doctors want to wait another four weeks before having me get an MRI. Given my symptoms (how long my leg has not worked properly, and how long I have been off from running) they think the chances that I am in fact healthy or will get better without intervention are still too high. That is I still fall into the first example. If things do not improve after four weeks then that increases the chances I do indeed have a problem they should treat if the MRI comes up positive. In this case the symptom that changes the odds I have something needing surgery, prior to having the MRI, is having a leg that does not improve with extended rest.

Can you get a myofascial massage in the meantime on the scarred area to see if it helps?

I am glad you posted this. Everyone is following your issues, and everyone here has some medical issue to deal with. The idea of sites like ours, in addition to the cameraderie and sharing of our day and our feelings, is also to share knowledge. Good for you to take the time to write this. Hope it all works out well for you. Spareribs

Ok, ok, I get it. Let me see if this very short version helps reduce the amount of . How to calculate the odds you are ill given a test says you are. 1. What is the chance you are ill absent the test? Call it x. So with probability 1-x you are healthy. 2. If you are ill what is the chance the test will say you are? Call this y. 3. If you are healthy what is the chance the test will say you are ill? Call this z. Chance you are ill given the test says you are: x*y/[x*y+(1-x)*z]. Call this value X. Should you get the test? Only if the following is positive: X*(expected value to you of the treatment given you are ill) - (1-X)*(expected cost to you of the treatment given you are not ill) <= if="" this="" value="" is="" positive="" get="" the="" test="" and="" treated="" if="" it="" says="" you="" are="" ill.="" if="" it="" is="" negative="" do="" not="" even="" bother="" getting="" tested.="" before="" you="" put="" in="" some="" crazy="" big="" number="" for="" the="" value="" of="" the="" treatment="" given="" you="" are="" ill="" remember="" many="" illnesses="" get="" better="" without="" treatement.="" in="" some="" cases="" you="" may="" not="" respond="" to="" the="" treatement.="" some="" treatments="" are="" themselves="" dibiliating="" in="" one="" way="" or="" another.="" consider="" the="" cost="" of="" the="" cure="" against="" what="" the="" disease="" will="" do.="" before="" you="" put="" in="" zero="" for="" the="" cost="" of="" the="" treatment="" given="" you="" are="" healthy="" consider="" the="" following:="" hospitals="" are="" a="" really="" good="" place="" to="" pick="" up="" a="" "super="" bug."="" if="" you="" have="" surgery="" the="" operation="" can="" kill="" you="" or="" cause="" permanent="" injury.="" operations,="" even="" if="" they="" go="" well,="" are="" no="" fun.="" so="" if="" the="" surgery="" is="" for="" nothing="" .="" .="" .="" drugs,="" no="" matter="" how="" well="" tested,="" typically="" carry="" some="" risks.="" some="" carry="" a="" lot="" of="" risks.="" so="" if="" the="" drug="" is="" for="" something="" you="" do="" not="" have="" .="" .="" .=""></=>evanflein hmm not sure if that is supposed to make me happy I cannot run or just jealous that I cannot run. lol!

Wishing you the best, Twocat. Even though you have logic behind the 4 week wait, it's got to frustrating to wait that much longer. Hang in there. TomS

Just bumped into this here far after the fact.  What happened?

And the point made in the OP is excellent.  There are loads of problems with false positives with testing.  A false positive test almost always leads to more testing, procedures and lots of anxiety, and occasionally unnecessary treatments.  And there is a huge cost burden that results.  And occasionally the definitive testing causes harm (like a biopsy that gets infected or collapses a lung or some such thing).

Some tests however are appropriate for screening large numbers of people despite low prevalence of disease.  Cervical cancer and breast cancer are two examples (tested respectively by pap smears and mammograms in relatively higher risk populations).  Another couple examples are screening for hypertension and high cholesterol.  The criteria for a test being a good screening test include that:

• it is inexpensive
• definitive testing is easily available
• detecting an abnormality can lead to an actual treatment
• it is easy to administer

Pap smears, mammograms, blood pressure checks, blood lipid checks all fit these criteria.  MRI tests do not.

Crud....That's worse than waiting four weeks till the next Batman movie....

 

... It's more like waiting four weeks to see if you're going to be kicked in the shins...

 

Maybe more like waiting four weeks to see if your blind date to the prom would be your cousin or somethin....

 

Thanks for the up date Two Kat...