need help with your account or subscription? click here to email us (or see the contact page)
join telegramNEW! discord
jump to exam page:
search for anything ⋅ score predictor (โ€œpredict me!โ€)

Retired NBME 20 Answers

nbme20/Block 3/Question#20 (reveal difficulty score)
A 67-year-old woman comes to the physician ...
Low sensitivity ๐Ÿ” / ๐Ÿ“บ / ๐ŸŒณ / ๐Ÿ“–
tags:

 Login (or register) to see more


 +18  upvote downvote
submitted by โˆ—hayayah(1212)
get full access to all contentpick a username

Sensitivity tests are used for screening. Specificity tests are used for confirmation after positive screenings.

Sensitivity tests are used for seeing how many people truly have the disease. Specificity tests are for those who do not have the disease.

A highly sensitive test, when negative, rules OUT disease. A highly specific test, when positive, rules IN disease. So, a test with with low sensitivity cannot rule out a disease. A test with low specificity can't rule in disease.

The doctor and patient want to screen for colon cancer and rule it out. The doctor would want a test with high sensitivity to be able to do that. He knows that testing her stool for blood will not rule out the possibility of colon CA.

get full access to all contentpick a username
sympathetikey  SeN Out (Snout) --> sensitive test; - test rules out SPec In (Specin) --> specific test; + test rules in +27
usmlecrasher  can anyone pls explain why it is not << potential false- positive results >> ??? +1
almondbreeze  correct me if I'm wrong, but 'high FP (choice C)=low specificity (choice B)'. Whereas high specificity is required to rule in dz +2
almondbreeze  picked positive predictive value myself. can anyone explain why not PPV? +1
williamfreakingosler  The principle @hayayah is talking about (a negative test being relied upon to reliably rule out) is negative predictive value ("NPV"). I don't see why "uncertain NPV" isn't the correct answer, particularly because NPV is predicated on the disease having the same base rate in the person(s) being tested as in the population that was characterized for the test statistic. Given that the patient has a strong family history of colon cancer, the NPV of FOBT is uncertain. Said another way, the sensitivity of a test does not change with the population, but the NPV does. The whole reason the doctor is denying FOBT is because of bayesian thinking (a priori information related to family history), and from my point of view bayesian logic is more relevant to PPV/NPV than to sensitivity, hence my confusion over why NPV isn't the right answer. +4
ibestalkinyo  I thought negative predictive value for the same reasoning +
raga7  AFTER THE RESULT OF TEST WE CAN USED PPV OR PPN, BUT FOR TEH FIRST TIME LOOKING ANY DESEASE USE SENSITIVITY OR SPECIFICITY. +2



 +5  upvote downvote
submitted by aazib05(5)
get full access to all contentpick a username

because they have asked about THE TEST, and sensitivity/specificity are the properties of the test. whereas PPV & NPV are dependant upon the population being tested, it's not the intrinsic property of the test.

get full access to all contentpick a username
lovebug  @aazib05 simple and clear. Thank you! +1



 +3  upvote downvote
submitted by โˆ—bbr(58)
get full access to all contentpick a username

Another way to think of this: She has a strong family history, so we are thinking she probably does indeed have this mutation (probably a True Positive). Our fear, would be we do the wrong test and aberrantly tell her that she is in the clear (False Negative). Having a high False Negative would be deleterious to this patient, and plugging this into a 2x2 table gives a low sensitivity (TP/ TP +FN).

get full access to all contentpick a username
trazobone  What a beautiful explanation THANK YOU +



 +2  upvote downvote
submitted by โˆ—l0ud_minority(15)
get full access to all contentpick a username

I picked the wrong one because Fecal Occult Blood Testing (FOBT) is high-sensitivity. My thought process is even if it is positive it won't tell use anything specific will need to scope anyways.

https://nccrt.org/wp-content/uploads/FOBTCliniciansReferenceFinal.pdf

get full access to all contentpick a username
divinedomain  this is exactly my reasoning +
abhishek021196  Even i got it wrong, then searched google and got this from Harrison (FOBT has low sensitivity and better methods are now available for screening) : https://harrisons.unboundmedicine.com/harrisons/view/Harrisons-Manual-of-Medicine/623689/all/COLORECTAL_CANCER PREVENTION Early detection of colon carcinoma may be facilitated by routine screening of stool for occult blood (Hemoccult II, ColonCare, Hemosure); however, sensitivity only โˆผ50% for carcinoma; specificity for tumor or polyp โˆผ25โ€“40%. Newer tests (e.g., Cologard) incorporating detection of blood and mutated genes are more sensitive and specific. False positives for occult blood: ingestion of red meat, iron, aspirin; upper GI bleeding. False negatives: vitamin C ingestion, intermittent bleeding. Genetic testing is unaffected by these factors. +



 +0  upvote downvote
submitted by โˆ—abhishek021196(119)
get full access to all contentpick a username

Even i got it wrong, then searched google and got this from Harrison (FOBT has low sensitivity and better methods are now available for screening) : https://harrisons.unboundmedicine.com/harrisons/view/Harrisons-Manual-of-Medicine/623689/all/COLORECTAL_CANCER - PREVENTION : Early detection of colon carcinoma may be facilitated by routine screening of stool for occult blood (Hemoccult II, ColonCare, Hemosure); however, sensitivity only โˆผ50% for carcinoma; specificity for tumor or polyp โˆผ25โ€“40%. Newer tests (e.g., Cologard) incorporating detection of blood and mutated genes are more sensitive and specific. False positives for occult blood: ingestion of red meat, iron, aspirin; upper GI bleeding. False negatives: vitamin C ingestion, intermittent bleeding. Genetic testing is unaffected by these factors.

So, to rule OUT a disease (Sen-Out/snout), screening test should have high sensitivity. Vice versa, to rule IN a disease (Spec-In/specin), confirmatory test should have high specificity.

I still don't understand how we rule out NPV and PPV from the answers since we're specifically talking about the 'at-risk' population group only and these values are used for that itself.

get full access to all contentpick a username



Must-See Comments from nbme20

amorah on Cytomegalovirus infection
masonkingcobra on Contact with parakeets
hayayah on Capillary hydrostatic: increased; ...
hayayah on X chromosome-linked isoenzymes
medbitch94 on Mannose 6-phosphate
hello on Capillary hydrostatic: increased; ...
hayayah on Osteoblasts
imgdoc on Hypophosphatemia
hayayah on Missense
celeste on 50%
justgettinby on Omeprazole
andrewk1 on Cold, dry air
yotsubato on Jugular venous pressure of 12 mm Hg
strugglebus on Drug effect

search for anything NEW!