Intro to Lab Medicine Flashcards
What is a reference range?
Population mean +/- 2 std deviations
Total cholesterol
Good: Below 200
Bad: >240
HDL
Good: >60
Bad: <40
TG
Good: <150
Bad: >500
LDL
Good: <130
Bad: >160
Normal blod glucose/A1c
70-99
<5.7
Diabetes
> 126
>6.5
Tells what fraction of people with disease have a positive test result
Diagnostic sensitivity
TP/ (TP+FN)
Tells what fraction of healthy people have a negative test
diagnostic specificty
TN/ (TN +FP)
Increasing sensitivity leads to _______ sensitivity
decreasing
What is a receiver operator curve?
Plot of sensitivity vs. 1‐specificity
Graphical representation of separation
Measure area under the ROC curve (AUC)
Higher values for AUC indicate better ability
to SEPARATE the groups of interest
If a patient has a positive test result, how liekly is he to have the disease?
PPV =
TP/ (TP+FP)
A patient’s test result for a disease is positive. The test is 80% sensitive and 95% specific. How likely is it she has the disease?
TP=80, FP =5, so TP/(TP+FP) = 80/85 or 94%
If your patient has a negative test result, how likely is she to not have disease?
TN/(TN+FN)
How does prevalence influence predictive values? Higher prevalence? Lower prevalence?
Disease prevalence in a population has a significant impact on predictive values
Higher prevalence = fewer FP (more likely to have a TP or disease… AIDS in africa)
Lower prevalence = fewer FN (More likely to have a true negative)
How does prevalence effect PPV?
as prevalence increases PPV increases
What are examples of high prevalence?
HIV‐1 infection prevalence in many sub‐ Saharan countries is in the 20‐30% range of the entire population
The prevalence of Alzheimer disease increases dramatically with age: 1.5% of people aged 60‐69 have the disease vs. ~40% of those aged ≥90
Specialist’s practice vs. primary care
Newborn screening programs have a high FP which in turn affects PPV…
LOW PPV
What are examples of pre-analytical errors?
• Test selection and ordering
– Wrong test, order missed or mis‐transcribed
• Specimen collection
– Wrong type of specimen or container, wrong time
of collection, contamination, insufficient sample • Specimen labeling (identification)
– MOST COMMON
• Transportation to laboratory
– Delays and time‐dependent changes in constituents, loss of sample, hemolysis
What are analytical error examples?
Clerical errors in the laboratory
Sample identification mix‐ups
Wrong test performed
Technical errors (Assay failure, interferences)
Results not communicated
Results communicated to wrong physician Wrong results communicated
Results not interpreted correctly
Are examples of…
Post-Analytical error
