Clinical Laboratory Testing

Question 1

Sequence of events between ordering lab tests and getting results.

Answer 1

1. order made
2. transcribed into system
3. specimen collected
4. specimen labeled AT BEDSIDE and sent to lab
5. accessioned in lab, compare name, check quality of specimen
6. distribute to appropriate sections for testing
7. tests performed
8. results generated, reviewed by tech, reported to patient care area; critical values repeated for verification before being called to nurse/physician
9. hard copy of report sent to physic/nursing unit; entered in EMR
10. physician uses results to make clinical decisions

Question 2

Define and give 3 examples of pre-analytical variabels

Answer 2

affect test results BEFORE they are analyzed:

1. sample collection - misidentification, poor phlebotomy techniques, expired tubes
2. sample handing: transport conditions, centrifugation
3. patient factors: physiologic, pathologic

Question 3

Define and give 2 examples of analytical variables

Answer 3

anything that can interfere with the actual analysis of the test by the lab including:

1. interfering substances in the sample that affect the methodology of the test (drugs, anticoags, antibodies, hemolysis)
2. instrument issues (mechanical problems, calibration problems, climate control problems)

Question 4

Define and give 3 examples of post-analytical variables

Answer 4

anything that can happen after the results are reported including:

1. interpretation of the result in light of clinical findings
2. physician does not receive or does not review report
3. error in manual data transfer in medical record
4. verbal/written miscommunication
5. physician treating numbers and not patient (does not consider false-positives, etc)

Question 5

What are the parts of the two-step post analytical review of laboratory results?

Answer 5

1. analytical correctness (by tech)

2. clinical significance (by physician)

Question 6

Define “normal range”

Answer 6

the range of values into which 95% of people without disease will fall (5% of individuals with no dz can have lab realist outside of nml range)

Question 7

What is the difference between true lab normal range and a “normal range printed in a table in a book”?

Answer 7

ones printed in a book are educated estimates; lab normals vary based on each hospital’s patient population

Question 8

Define false positive

Answer 8

person who does not have the dz but tests positive

Question 9

Define false negative

Answer 9

person who has the dz but tests negative

Question 10

Describe the 2x2 Truth table.

Answer 10

total for positive results = TP + FP
total for negative results = FN + TN
Total with disease = TP + FN
Total with no disease = FP + TN

where tp is true positive, fp is false positive etc

Question 11

Define sensitivity. What is the formula?

Answer 11

the probability of a positive result in a person with the disease (true positive rate); the fewer the false negatives, the higher the sensitivity. greater the sensitivity means greater the efficacy to correctly identify diseased patients.

**focus is on groups with disease

sensitivity = ((TP)/(TP + FN)) x 100

Question 12

Define specificity. What is the formula?

Answer 12

the probability of a negative result in a person without disease (true negative rate); fewer the false positives, the higher the specificity.

**focus is on groups without disease

specificity = ((TN)/(TN+FP)) x 100

Question 13

Define predictive value.

Answer 13

the probability of disease/no disease for a positive/negative result, respectively.

positive predictive value - focus on pop. with positive test, regardless of dz status. the better the positive predictive value the more likely a positive test means the pt has the dz. greatly influenced by prevalence of the dz. (low prevalence means many more FP occur than TP)
positive predictive value = (TP/(TP+FP))x100

negative predictive value - focus on pop. with negative test, regardless of dz status. the better the negative predictive value, the more likely a negative test means the pt does not have the dz. not greatly influenced by prevalence
negative predictive value = (TN/(TN+FN))x100

Question 14

Define prevalence.

Answer 14

of existing cases in population AT THAT TIME- often expressed as a percentage of the population (positive tests more valid with high prevalence)

Question 15

Define incidence.

Answer 15

of NEW cases in population/unit time (OVER A PERIOD OF TIME); usually expressed as a number

Question 16

Define accuracy

Answer 16

how close the measured value is to the true value of the analyte

Question 17

Define precision

Answer 17

how reproducible is the result with repeated testing, whether the result is close to the true value or not