Clinical Chemistry Exam 1

What is clinical judgment?

Making observations without official lab results

What is quality?

Accuracy and precision

What two things ensure accurate and reliable test results?

Quality assurance and quality control

What is pre-analytical testing phase?

Happens before experiment

What is analytical testing phase?

Happens during the experiment

What is post-analytical testing phase?

Happens when results are sent out

What is quality assurance?

The process that guarantees results is accurate

What testing phases does quality assurance involve?

Pre-analytical, analytical, and post-analytical

What is quality control?

The process of monitoring and accessing the analytical phase of testing

What testing phase does quality control involve?

Analytical phase

What is the most critical testing phase?

Pre-analytical

Name some common pre-analytical errors

Wrong patient sample, incorrect test ordered/performed, sample not properly collected, & sample not preserved properly

What is the most common pre-analytical error?

Patient sample not properly collected

In which testing phase do most errors occur?

Pre-analytical phase

Name some common analytical errors

Instrument malfunction, lack of maintenance, incorrect calibration, defective reagents, & improper training

Name some common post-analytical errors

Wrong results reported, computer error, & incorrect interpretation by physician

What are Levey-Jennings control charts?

Charts used to determine quality control over time

What are the two types of errors in the analytical phase?

Systemic error and random error

What are the two subcategories of systemic error?

Constant error & proportional error

What is the difference between constant and proportional error?

Constant: No change occurs

Proportional: Change occurs based on other factors

When is quality control considered to be performing adequately?

The control falls within its set values

What is a shift?

Abrupt change (quick change)

What is a trend?

Gradual change (occurs over time)

What are the 6 main Westgard rules?

1 2s, 1 3s, 2 2s, R4s, 4 1s, & 10x

What is 1 2s?

One of the two controls falls outside ± 2s

What is so special about 1 2s?

It is considered a warning rule & you don’t reject the sample run

What type of error is 1 2s?

Random error

When does 1 2s usually occur?

1/20 times

What is 1 3s?

One of the two controls falls outside ± 3s

What type of error is 1 3s?

Random error

Do you reject or accept 1 3s?

Reject

What is 2 2s?

Two consecutive measurements exceed the same ± 2s

What type of error is 2 2s?

Systemic error

Do you reject or accept 2 2s?

Reject

What is 4 1s?

Four consecutive measurements exceed the same ± 1s

What type of error is 4 1s?

Systemic error

Do you reject or accept 4 1s?

Reject

What is R4s?

One control measurement exceeds the ± 2s, and another exceeds on the other side

What type of error is R4s?

Random error

Do you reject or accept R4s?

Reject

What is 10x?

Ten consecutive measurements fall on one side of the mean

What type of error is 10x?

Systemic error

Do you reject or accept 10x?

Reject

What is the purpose of proficiency samples?

Allows every lab to be on the same page because every lab should get the same results

What is the purpose of Lean Six Sigma?

Eliminate waste, promote productivity, and help with patient bed turnover rate

What is allowable analytical error?

Determines if the total error is less than the total allowable analytical error

Who has to certify every lab?

Clinical Laboratory Improvement Amendments (CLIA)

What is diagnostic sensitivity?

Proves a person DOES have a disease/condition

What is analytical sensitivity?

Detects small quantities/small changes

What is clinical sensitivity?

The proportion of individuals with a disease/condition who test positive

How do you calculate diagnostic sensitivity?

# positive test/ # of people tested

What is diagnostic specificity?

Proves a person DOES NOT have a disease/condition

What is analytical specificity?

Detects only the desired analytes

How do you calculate diagnostic specificity?

# of negative tests/ # of people tested

What are positive predictive values?

Probability of people who test positive

How do you calculate positive predictive values?

# of people who tested positive/ # of positive tests

What are negative predictive values?

Probability of people who test negative

How do you calculate negative predictive values?

# of people who tested negative/ # of negative tests

The mean value for a control is 80 mg/dL and one SD is 5 mg/dL. The control is ran twice in the same run, and the values are 68 and 67 mg/dL. What do these results show?

A) Poor precision and good accuracy

B) Good precision & poor accuracy

C) Good precision and good accuracy

D) Poor precision and poor accuracy

B) Good precision & poor accuracy

Which of the following is a false positive?

A) Patient with Hepatitis C and his liver enzyme test results are elevated

B) Patient is not pregnant and has a positive test result
C) Patient has prostate cancer, but his PSA is normal
D) Patient does not have diabetes and has a normal fasting blood glucose

B) Patient is not pregnant and has a positive test result

What are descriptive statistics?

Numerical values that summarize clinical data

What is the center in descriptive statistics?

Mean, median, or mode

What is the spread in descriptive statistics?

Standard deviation

What is the shape in descriptive statistics?

Distribution of clinical data

When is it allowed to use the mean for the center in descriptive statistics?

When the clinical data makes sense

Example: data numbers are similar in values

What is spread?

Relationship between the data points and the mean

How do you calculate coefficient of variation (CV)?

SD/ mean

What is the shape of a normal distribution of clinical data?

Gaussian distribution

What percent of clinical data falls within ± 1 SD?

68%

What percent of clinical data falls within ± 2 SD?

95%

What percent of clinical data falls within ± 3 SD?

99%

What are the two ways to measure precision?

Within run (repeatability) & within laboratory (across run)

What is within-run precision?

Level of closeness between results under the same condition

EX: 1 sample ran 20 times

What is within laboratory precision?

Closeness between results over multiple days

Ex: 1 sample ran 20 times over the course of 10 days

What type of error is indicated when the y-intercept changes?

Constant error

What type of error is indicated when the slope changes?

Proportional error

What type of error is indicated when the distance of each point to the regression line (S y/x) changes?

Random error

What are the three parametric tests?

F-test, t-test, & ANOVA

When do you use parametric tests?

When the data has a normal distribution (Gaussian data)

What does an F-test use?

The entire population

What does an F-test compare?

The difference between the variances of two groups

What a t-test use?

The sample population

What does a t-test compare?

The means between two groups

What does an ANOVA test use?

Variance

What does an ANOVA test compare?

If three or more groups have the same mean

What is inferential statistics?

Statistics used to conclude clinical data regarding the mean or SD

When do you use nonparametric analysis

When clinical data is not normally distributed (non-Gaussian)

Is nonparametric analysis or parametric analysis more commonly used in laboratories?

Nonparametric analysis

How do you test for nonparametric analysis?

Reference range studies (shows normal and abnormal ranges)

What are the categories of applications of reference range?

Diagnosis of a disease, monitoring of a physiologic condition, & monitoring therapeutic drugs

What do you need to know in order to determine reference range?

Parameters

If you need to test 20 samples, is that parametric or nonparametric?

Parametric

If you need to test 120 samples, is that parametric or nonparametric?

Nonparametric

What is the most common error when establishing reference ranges?

Using old samples (always use fresh samples)

Reference ranges are usually designed to include ———-% of the clinically normal and healthy population.

A) 50%

B) 100%

C) 95%

D) 99%

C) 95%

What percent of reference ranges fall within the normal and healthy range?

95%

What percent of reference ranges fall within the abnormal range?

5%

What does spectrometry measure?

Light intensity

What are the four basic quantitative measurements?

Fluorescence, spectrometry, electroanalytical methods, & chromatography

What happens to radiant energy during spectrophotometry?

Partially reflected, absorbed, & transmitted