Mean, Median, Mode, and Confidence Intervals Sensitivity, specificity, and Predictive Value Standard Curves

A Proficiency Testing Service (e.g., CAP) reported the following results for a Cholesterol test (unknown Control) reported by your laboratory. Your results were compared with 648 other labs using the same type instrumentation and methodology: Your Result = 168 mg/dlGroup Mean = 155 mg/dlGroup SD = 11 mg/dl (SD = One Standard Deviation)Group ± 2 SD Range = 22 mg/dl Calculate the value for your Standard Deviation Index and choose the correct answer below:

+ 1.18

Below are the creatinine levels in mg/dL for 21 healthy hospital employees: .87.98.931.04.86.901.051.08.84.971.12.95.961.021.01.93.91.98.99.941.04 What is the median of these data?

.97 mg/dL

Creatinine levels in mg/dL for 21 healthy hospital employees .87.98.931.04.86.901.051.08.84.971.12.95.961.021.01.93.91.98.99.941.04 What is the mean of the data?

.97 mg/dL

What is the standard deviation of this set of data? Creatinine (mg/dL)x-(x-)20.87-0.100.010.980.010.00011.040.070.00490.86-0.110.01210.90-0.070.00491.050.080.00641.080.110.01210.84-0.130.01690.97001.120.150.02250.95-0.020.00040.960.010.00011.020.050.00251.010.040.00160.93-0.040.00160.91-0.060.00360.980.010.00010.990.020.00040.94-0.030.00091.040.070.00490.93-0.040.0016mean= 0.97total0.1076

0.07 mg/dL

Using these values: Mean( )= 0.97 mg/dL; Standard deviation (SD) = .07 mg/dL, what is the "normal range" for creatinine levels? That is, in what range will 95% of healthy individuals' creatinine levels fall?

0.83 - 1.11 mg/dL

In a normal distribution, approximately what percent of data would be more than +/- 3 standard deviations (SD) from the mean?

1%

Using the formula on the right and the data below, what is the standard error of estimate (Se) for this regression line? PointReference Method (x)Test Method (y)x-y-(x-)(y-)(x-)2120-3-5159245-1001366111148934129TotalAverage= 5Average= 52820 b = 1.4 a (y-intercept) = -2

1.18

The mean for the normal hemoglobin control is 14.0 mg/dL. The standard deviation is 0.15 with an acceptable control range of +/- 2 standard deviations (SD). What are the acceptable limits of the control?

13.7 - 14.3

What is the mean of the following data set? 15, 16, 18, 16, 14, 10, 20, 12, 14

15

What is the standard deviation of the following data set, rounded to the nearest whole number? 15, 16, 18, 16, 14, 10, 20, 12, 14

3

How many standard deviations (SD) above and below the mean is accepted as being an appropriate control limit range on a control chart utilizing the Westgard rules:

3 SD

Based on the results for the laboratory test method that is shown below, what is the specificity of this method?

67%

What is the coefficient of variation for this creatinine testing method? Use (Mean = .97 mg/dL, s = .07 mg/dL)

7.22%

Calculate the coefficient of variation for a set of data where the mean = 89 mg/dL and two standard deviations = 14 mg/dL

7.9%

For a certain method, a control has a mean result of 12 with a standard deviation (SD) of 2. What is the acceptable 95% range for this control?

8 - 16

What is the Coefficient of Variation for a set of numbers when: ± 1 SD = 6.8, mean = 80, and the range for ± 1 SD = 73.2 - 86.8?

8.5%

Given the following examination scores; what is the mean? 788871898994618291767372818780

81

The statistics for a specific laboratory test method are illustrated to the right. What is the sensitivity of this test?

94%

As defined by a Gaussian distribution curve, what percentage of values would be expected to fall within two standard deviations of the mean?

95%

In a normal distribution of results, the mean value +/- 2 SDs will include what percentage of the population?

95.5%

A mean value of 100 mg/dL and a standard deviation of 1.8 mg/dL were obtained from a set of glucose measurements on a control solution. Based on a 95% confidence interval (in mg/dL), what would be the calculated 2 SD upper and lower limits for this control?

96.4 to 103.6

Calculate the specificity for the following medical research project completed by a manufacturer: Test ResultsPositiveNegativeDisease9847No Disease16873

98.2%

Which of the following best defines "specificity"

A test's ability to correctly exclude individuals who do not have the given disease or disorder.

An assayed glucose control produces the following results in 10 runs: 99, 99, 99, 98, 98, 98, 98, 97, 97, 97. The established range for this control is 90-92. Where does the problem lie with this control?

Accuracy

Which of the following terms may be defined as the extent to which a measurement agrees with the true value of the analyte being measured?

Accuracy

The closeness of the results obtained to the true and actual value of an analyte is a reflection of:

Accuracy of the assay

The accuracy of an immunoassay is its ability to discriminate between results that are true positive and results that are true negative. Two parameters of test accuracy are specificity and sensitivity. Which of these statements apply to an immunoassay with high specificity?

Accurately identifies the absence of disease

The accuracy of an immunoassay is its ability to discriminate between results that are true positives and results that are true negatives. Two parameters of test accuracy are specificity and sensitivity. Which of these statements apply to an immunoassay with high specificity, but low sensitivity?

Accurately identifies the absence of disease and has few false-positives

Which of the following statements regarding a biomarker with high sensitivity is TRUE?

Accurately identifies the presence of disease and has few false-negatives.

Which of the following statements best describes the mean and median as they relate to a normal "gaussian" distribution curve?

Both are equal

Which of the following statistical methods would be employed to determine how closely two different methods compare with each other?

Coefficient of variation

What is the purpose of a t-test?

Comparison of two means from matched groups

Which of the following is true regarding a normal distribution curve?

Data are more likely to fall closer to the mean.

What is defined as the ability of the test to detect a given disease or condition?

Diagnostic sensitivity

The number of patients correctly diagnosed for the disease or not having the disease is defined as:

Efficiency

The accuracy of an immunoassay is its ability to discriminate between results that are true positive and results that are true negative. Two parameters of test accuracy are specificity and sensitivity. Which of these statements apply to an immunoassay with low sensitivity?

Has many false-negative results

A laboratory developing a new testing procedure runs the test on 1,000 patients. Later confirmation shows that the experimental procedure produced remarkably few false-negative results, but many false-positive results. Which of the following statements is true regarding the experimental procedure?

It has high sensitivity

You are working to validate a new method for the detection of C-reactive protein (CRP) in serum. To assess precision, you run a medium level control 30 times and determine the mean to be 4.0 mg/dL with a standard deviation of 0.04 mg/dL. What could be said about the precision of this run assuming a CV less than 3% is acceptable?

It is acceptable, with a coefficient of variation of 1%

Analytical sensitivity of a method generally refers to:

Its ability to measure low concentrations of analytes.

Which of the following can be defined as the average of the data points or the sum of all the data points divided by the number of total points?

Mean

Which of the following terms is used to express the average of a series of numbers or values:

Mean

If a laboratory 's control range (using a 99.7% confidence interval) for a given assay is 20.0 to 50.0, what would its mean and one standard deviation be:

Mean = 35.0 SD = 5.0

An analytical method with a low detection limit would:

Measure low concentrations of analyte

Which of the following terms would be used to express the middle value in a series of results?

Median

If a test is said to have a sensitivity of 95%, statistically, this means that it will:

Miss 5 out of 100 true positives

The numeric value that occurs most frequently in a sample distribution is called the:

Mode

The value that occurs most frequently in a set of results or values would be termed

Mode

All of the terms below are a measure of central tendency, EXCEPT

Molarity

You are running quality control for Hemoglobin A1C by high-pressure liquid chromatography (HPLC). The high control was 2.5 standard deviations above the mean. The low control was 2.2 standard deviations below the mean. Should you continue to run patient samples?

No. These QC values indicate a violation of the Westgard R4S rule and are therefore cause for rejection.

The term that identifies how close a set of results are to each other when running a quality control validation or similar study is called:

Precision

Based on the following specificity and sensitivity information, which of the answer choices is true? Procedure #1 detected 70/100 true positives and 100/100 true negatives Procedure #2 detected 90/100 true positives and 80/100 true negatives

Procedure 2 is more sensitive and less specific

Precision is a measure of:

Random variability

The 95% confidence limit refers to values that fall within:

Range equal to the mean plus or minus two standard deviations

The degree in which a measure is consistent in producing the same results when measuring the same analyte is called:

Reliability

Which of the following best describes random error?

Results may be either above or below the mean.

The research team at the hospital selected 16 employees at random and tested their BUN levels. They found an average of 16 mg/dL with a standard deviation of 6.5 mg/dL. They used this data to construct a range of normal values for the whole healthy population. The underlined phrase is an example of what portion of the research study?

Sample

The quality of a test to show a positive result in the presence of the disease it was designed to detect is known as:

Sensitivity

You are evaluating the effectiveness of a new, office-based rapid test for Influenza B. Among 1000 patients tested for Influenza B, 20 tested positive using this rapid test whereas the remaining 980 patients tested negative. In addition to the 20 patients who tested positive, 10 more patients were infected with influenza B, but tested negative when using this particular rapid test. What is the sensitivity and specificity of this test?

Sensitivity = 66% and Specificity = 100%

When evaluating the throughput of a particular method you should consider all of the following except:

Sensitivity of technique

Ideally, what testing should be done prior to starting a new lot of control material in order to establish its mean and standard deviation?

Should be tested in 20 separate runs

The ability of an instrument or testing method to provide a negative result for patients who do not have the disease or condition is known as:

Specificity

What does the equation in parenthesis measure? (TN/TN+FP)

Specificity

The positive square root of the variance of a set of values is called:

Standard deviation

The term, analytical specificity, refers to:

The ability to measure only that substance being assayed.

A direct antigen test for detecting the presence of an infection has the statistics in the table below: Test Sensitivity Specificity 5% Prevalence Positive Predictive Value 5% Prevalence Negative Predictive Value 30% Prevalence Positive Predictive Value 30% Prevalence Positive Predictive Value Direct Antigen 90%96%54%90%91%89%

The higher the prevalence, the greater the chance a positive test result is correct

Which of the following best defines "sensitivity"?

The percentage of patients with a disease who will have a positive test result

Based on the data illustrated on the right, which of the following statements is true? For the experimental and tried-and-true procedures, which test would produce fewer false-negatives?

The tried-and-true method will give 7 false positives out of every 100 samples.

All of the following decrease medical errors in the clinical laboratory, EXCEPT:

Variance