Quiz 3 Flashcards
390 students reported no influenza symptoms on the survey and tested negative for influenza via PCR.
This describes cell D of the 2 x 2 table – they are survey negative and PCR (gold standard) negative. This represents the true negatives identified by the screening test.
92% of students with PCR-defined influenza reported influenza symptoms on survey.
This measure is sensitivity or A/(A+C). The numerator describes those who were identified as positive on the survey (screening test) and by the gold standard PCR test (A cell). The denominator described here includes all people who are PCR positive (A cell + C cell).
Of the 397 students that reported no influenza symptoms on survey, 98% of them did not have PCR defined influenza.
This measure is negative predictive value = D/(C+D). The numerator described here includes the individuals who were identified as negative on both the survey and PCR (D cell). The denominator described here includes all individuals who were identified as negative only on the survey (C cell + D cell).
20 students reported influenza symptoms on the survey but did not have PCR-defined influenza.
These 20 students represent those who were survey positive, but PCR negative. In other words they reported having symptoms of influenza disease, but did not actually have the disease. These are the individuals in cell B, and they are our false positives.
Of the 500 students in the study, 18% have PCR-defined influenza.
This is the definition of prevalence: (Total # of people identified by the gold standard as diseased )/(# of people in population).
What is the sensitivity of the urine test? (Report as a percentage. For example, if the answer is 19.5%, enter “19.5” as your answer. Please do not round.)
Sensitivity: (True Positive)/(True Positive+False Negative)= A/(A+C) = 50/(50+50)
=50/100 = 0.50
0.50*100 = 50%
What is the specificity of the urine test? (Report as a percentage. For example, if the answer is 19.5%, enter “19.5” as your answer. Please do not round.)
Specificity: (True Negative)/(True Negative+False Positive)= D/(B + D)
= 350/(50+350)=350/400 = 0.875
0.875*100 = 87.5%
In public health practice, developing a highly sensitive screening test that comes at the expense of specificity for a given population will result in the following (select all that apply):
The correct answers are: increase the number of true positives and increase the number of false positives. Increasing the sensitivity of a test would increase the number of people with disease who correctly screen positive for having the disease, so there would be more true positives and fewer false negatives. Decreasing the specificity of a test would decrease the number of people without disease who correctly screen negative for disease, so there would be fewer true negatives and more false positives.
A test that is reliable will also necessarily be
A test that is reliable gets the same or similar answers most of the time, so a reliable test will necessarily be precise. However, a reliable test does not necessarily give us the correct answer, so we do not know if it is a valid / accurate test. We also do not know whether this test is the best test available for identifying the disease, so it is not necessarily the gold standard test.
When a serious disease can be treated if it is caught early, it is important to have a test with high sensitivity.
This statement is true. When a serious disease can be treated if it is caught early, we want a test with high sensitivity, because we want to correctly identify as many people as possible who truly have disease so that they can get treatment. A test with high specificity would allow us to correctly identify most people who truly do not have the disease, but these are not the people who we need to identify for treatment, so high sensitivity is more important than high specificity in this scenario.
