-
pre-analytical factors of error in lab tests (6)
patient variables (age, breed, etc), lipemia, hemolysis, icterus, collection procedures, and shipping procedures
-
lipemia is caused by...
circulating trigycerides, causing visible latescence (fat-containing meal prior to sample collect)
-
hemolysis is caused by...(5)
rupture of red blood cells due to pathology, difficult venipuncture, wrong needle used, excessive suction on syringe, rough transfer of blood
-
icterus is caused by...
increased concentration of bilirubin (yellow discoloration)
-
analytical factors of error in lab tests (4)
analyzer and equipment used to process the sample, test methodology and components, quality control procedures, and operating and maintenance procedures
-
post-analytical factors of error in lab tests (2)
data review by the laboratory and clinician, data entry and reporting
-
basic principals of a quality control system (3)
standard operating procedure, calibration, daily controls prior to analyzing patient samples
-
how do you determine if an isolated laboratory abnormality is indicative of disease or lab error/ biological variation?
the more tests you run, the more likely you are to get normal animals with test results outside of the RI; consider the animal's age, breed, history, diet, excitement, and medications; examine possible sources of error; for some analytes, a small deviation from RI is significant, for others only large deviations are medically significant (know the difference); note possibility of interassay variation; draw a new sample and retest if results don't make sense
-
define sensitivity
animals with disease that have positive test results; TP/ (TP+FN)
-
define specificity
animals without disease that have negative test results; TN/ (TN +FP)
-
Negative predictive value
% of animals with negative test that are actually disease-free; TN/ (TN+FN)
-
positive predictive value
% of animals with positive test that actually have the disease; TP/ (TP+FP)
-
as prevalence of a disease goes down, ____ of a test goes up.
NPV; thus, more animals that test negative are actually healthy
-
as prevalence of a disease goes up, _____ of a test goes up.
PPV; more animals that test positive actually have the disease
-
Confirmatory tests should have high ________ in order to ....
specificity; so if it is positive, you know the animal is not healthy (low FP); used when treatment may be very harmful; DON'T WANT TO TREAT NEGATIVE ANIMALS
-
screening tests have high ______ in order to...
sensitivity; negative animals can be ruled out (Low FN); used when cost of FN is high-such as in a herd; DON'T WANT TO MISS POSITIVE ANIMALS
-
reference intervals are established by sampling animals from a ________; then,.....
healthy population; for gaussian curve, RI is + or - 2SD from the mean (therefore, ~5% of normal animals will fall outside of RI
|
|
