Statistical Information and Courses from MediaLab, Inc.

Rule-out (also referred to as exclusion or cross-out) is a process by which antibodies are identified as being unlikely in a given sample because of the absence of an expected antigen-antibody reaction. In other words, the absence of a reaction is noted with a cell that is positive for the corresponding antigen. Rule-out, while very useful, can lead to error. Ruling out an antibody should be combined with other supporting data to increase confidence in the solution; the more data collected, the higher the probability that the final solution is correct.Non-reactive cells are selected for rule-out. To be classified as non-reactive, a cell must NOT have reacted in any phase of testing in a given panel or screen. In the case of cold antibodies: if reactions are only occurring at immediate spin and are negative in the AHG phase, then that panel cell can be used as a rule out cell for IgG reactive antibodies but not for antibodies that react at immediate spin (IgM).If there is no reaction with a panel cell then it is possible that antibodies to the antigens on that cell are not present in the sample being tested.

The core of the Six Sigma quality management approach is measurement of defects in order to get as close to zero defects as possible. Sigma is a statistical term that measures how far a process deviates from total accuracy or perfection. The process sigma, which is also known as the sigma level, is a measure of process capability. The higher the process sigma, the more capable the process is. A Six Sigma process has a short-term (DPMO) process sigma of 6. When determining the long-term process sigma, 1.5 is subtracted from the short-term metric, so that the long-term process sigma for a Six Sigma process is 4.5. Six Sigma is often wrongly defined as "3.4 defects per million products," when in fact, Six Sigma is actually defined as 3.4 defects per million opportunities (DPMO). Six Sigma's goal is to improve all processes to that level or better.To determine the number of opportunities a process contains, one should think of the number of opportunities in which a defect may occur. For example, if you are measuring emergency department (ED) stat turnaround times from order to completion, a defect would be any result not reported within the specified turnaround time. Opportunities for defects (delays) can occur in the three phases of laboratory testing (pre-analytical, analytical, and post-analytical phases). An example of DPMO and process sigma (sigma level) measurement is given on the following page.

Since standard deviation, mean, median, and mode are all absolute data on statistical samples, they do not permit a direct comparison of variation between samples with different means or different units of measurement.One way to obtain a measure of variation that has no units is to divide the standard deviation (s) by the mean (), and multiply by 100 to give a percent. This quantity is called the coefficient of variation (CV), and can be used to compare methods that give different units. For example, the coefficient of variation for two different glucose methods would be calculated as shown below after the mean and standard deviation for each method has been established. The hexokinase method has = 99 mg/dL, and s = 8.0 mg/dL. The orthotoluidine method has = 105 mg/dL, and s = 12.5 mg/dL. From these CV's we would conclude that the hexokinase method is relatively more precise because it has a lower CV.

We have listed the 'classic' cardiovascular risk markers as LDL-C, HDL-C and triglycerides. But there are many more cardiovascular risk markers as well as cardiovascular risk factors. A cardiovascular risk factor is a condition (not a laboratory analyte) that is associated with an increased risk of developing cardiovascular disease. Examples include: Age Gender (males are at increased risk) Heredity Hypertension Cigarette Smoking Obesity Diabetes StressThere are also negative risk factors, factors which decrease a person's risk of cardiovascular disease. Examples include: Optimal HDL-C concentration Exercise Estrogen Moderate alcohol intakeThis course will not focus on cardiovascular risk factors. Instead we will focus on newer, emerging cardiovascular risk markers. There are well over twenty well-studied cardiovascular risk markers; in this course we will focus on some of the more established markers and the ones which are becoming more commonly measured in the clinical laboratory. These include apolipoprotein A1/apolipoprotein B100, Lp(a), oxidized LDL, LpPLA2, hsCRP and lipoprotein particle size and concentration.It is important to remember that the association between a cardiovascular risk marker and actually having or developing cardiovascular disease is a statistical one. The fact that a patient has a particular risk marker which is abnormal simply increases the probability of developing cardiovascular disease, it does not mean that he or she is certain to develop cardiovascular disease. Conversely, if an individual does not have a particular cardiovascular risk marker present it does not guarantee protection against cardiovascular disease. We must always remember that some percentage of individuals who have heart attacks or strokes will not have abnormal risk markers present.

Mean can be defined as the average of the data points. Standard deviation (SD) is a measure of imprecision. It indicates the variability or dispersion around the mean. Together, mean and SD determine acceptable ranges for a lot of control material. New control values must be calculated and acceptable ranges established for each new lot of control materials. Ideally, at least 20 samples should be tested over time for good statistical data.The mean is calculated by adding all of the values, and dividing by the number of values. The formula is: For example, suppose you wanted to find the mean of the values 4, 6, 2, 8, and 5. The mean is: The standard deviation (abbreviated s or SD) is calculated according to the following formula: That is, calculate the deviation from the mean for each point, square those results, sum them, divide by the number of points minus one, and finally take the square root. For example, the deviations from the mean in the above example are -1, 1, -3, 3, and 0. The squared deviations are 1, 1, 9, 9, and 0. The standard deviation is therefore:

Systematic error causes inaccurate results that are consistently low or high. Factors that contribute to systematic error include: Change in reagent lotChange in calibrationWrong calibrator values assignedImproperly prepared or deteriorating reagentsPipettor maintenance error (not adjusted correctly or misaligned)Deterioration of a photometric light source

A good QC program requires documenting control results and observing and assessing that documentation daily. This section is an overview of some of the ways to document and assess QC results. First, we'll cover some basic statistical terms:meanGaussian (or bell) curvestandard deviation

In the late 1950's, Dr. William Youden (1900-1971) developed what has now become known as the Youden Plots. This statistical technique involves both normal and abnormal controls and graphically helps to differentiate between systematic and random errors. The inner square of the plot (yellow) represents one standard deviation (1SD). The next larger square (green) represents 2SD, and the outer square (blue) represents 3SD. A horizontal median line is drawn parallel to the X-axis and a second median line is drawn parallel to the Y-axis. The intersection of the two median lines is called the Manhattan Median. One or two 45-degree lines are drawn through the Manhattan Median. The results of at least two different levels of controls (e.g. Level 1/Level 2 or Normal/Abnormal) are then plotted on the chart as X-axis versus Y-axis.

The p value is a statistical tool that increases the confidence that an antibody has been identified with a scientifically acceptable level of uncertainty (0.05). As applied to antibody identification, it is computed using Fisher's exact test. Tidbit: This is the same Fisher who helped developed the Fisher-Race theory of Rh inheritance.The p value is calculated using the number of cells that are positive and negative with the patient's plasma. Calculating p values is beyond the scope of this case study but basic understanding of p values at the conceptual level is covered.