Compute the answer on your own before looking at the answer. One problem is that a decision must be made about what test value will be used to distinguish normal versus abnormal results. Unfortunately, when we compare the distributions of screening measurements in subjects with and without disease, we find that there is almost always some overlap, as shown in the figure to the right. Deciding the criterion for "normal " versus abnormal can be difficult.
There may be a very low range of test results e. However, where the distributions overlap, there is a "gray zone" in which there is much less certainly about the results. If we move the cut-off to the left, we can increase the sensitivity, but the specificity will be worse. If we move the cut-off to the right, the specificity will improve, but the sensitivity will be worse. Altering the criterion for a positive test "abnormality" will always influence both the sensitivity and specificity of the test.
ROC curves provide a means of defining the criterion of positivity that maximizes test accuracy when the test values in diseased and non-diseased subjects overlap. As the previous figure demonstrates, one could select several different criteria of positivity and compute the sensitivity and specificity that would result from each cut point.
In the example above, suppose I computed the sensitivity and specificity that would result if I used cut points of 2, 4, or 6. If I were to do this for the example above, by table would look something like this:. I could then plot the true positive rate the sensitivity as a function of the false positive rate 1-specificity , and the plot would look like the figure below. Note that the true positive and false positive rates obtained with the three different cut points criteria are are shown by the three blue points representing true positive and false positive rates using the three different criteria of positivity.
This is a receiver-operator characteristic curve that assesses test accuracy by looking at how true positive and false positive rates change when different criteria of positivity are used. If the diseased people had test values that were always greater than the test values in non-diseased people, i. The closer the ROC curve hugs the left axis and the top border, the more accurate the test, i.
The diagonal blue line illustrates the ROC curve for a useless test for which the true positive rate and the false positive rate are equal regardless of the criterion of positivity that is used - in other words the distribution of test values for disease and non-diseased people overlap entirely.
So, the closer the ROC curve is to the blue star, the better it is, and the closer it is to the diagonally blue line, the worse it is. This provides a standard way of assessing test accuracy, but perhaps another approach might be to consider the seriousness of the consequences of a false negative test. For example, failing to identify diabetes right away from a dip stick test of urine would not necessarily have any serious consequences in the long run, but failing to identify a condition that was more rapidly fatal or had serious disabling consequences would be much worse.
Consequently, a common sense approach might be to select a criterion that maximizes sensitivity and accept the if the higher false positive rate that goes with that if the condition is very serious and would benefit the patient if diagnosed early. Here is a link to a journal article describing a study looking at sensitivity and specificity of PSA testing for prostate cancer.
BMC Family Practice , ]. In the video below Dr. David Felson from the Boston University School of Medicine discusses sensitivity and specificity of screening tests and diagnostic tests.
When evaluating the feasibility or the success of a screening program, one should also consider the positive and negative predictive values. These are also computed from the same 2 x 2 contingency table, but the perspective is entirely different.
One way to avoid confusing this with sensitivity and specificity is to imagine that you are a patient and you have just received the results of your screening test or imagine you are the physician telling a patient about their screening test results.
If the test was positive, the patient will want to know the probability that they really have the disease, i. Conversely, if it is good news, and the screening test was negative, how reassured should the patient be? What is the probability that they are disease free?
Another way that helps me keep this straight is to always orient my contingency table with the gold standard at the top and the true disease status listed in the columns. The illustrations used earlier for sensitivity and specificity emphasized a focus on the numbers in the left column for sensitivity and the right column for specificity. If this orientation is used consistently, the focus for predictive value is on what is going on within each row in the 2 x 2 table, as you will see below.
If a test subject has an abnormal screening test i. In the example we have been using there were 1, subjects whose screening test was positive, but only of these actually had the disease, according to the gold standard diagnosis. Interpretation: Among those who had a positive screening test, the probability of disease was Negative predictive value: If a test subject has a negative screening test, what is the probability that the subject really does not have the disease?
What are sensitivity and specificity? What is predictive value? What criteria should be considered for an effective screening program? Life-threatening diseases, such as breast cancer, and those known to have serious and irreversible consequences if not treated early, such as congenital hypothyroidism, are appropriate for screening.
Treatment of diseases at their earlier stages should be more effective than treatment begun after the development of symptoms. For example, cancer of the uterine cervix develops slowly, taking more than a decade for the cancer cells to progress to a phase of invasiveness.
During this preinvasive stage, the cancer is usually asymptomatic but can be detected by screening using the Pap smear. Treatment is more effective during this stage than when the cancer has become invasive. On the other hand, lung cancer has a poor prognosis regardless of the stage at which treatment is initiated. Early diagnosis and treatment appear to prolong life little more than therapy after symptoms have developed.
Screening to detect early stage lung cancer using currently available techniques would not be beneficial. The prevalence of the detectable preclinical phase of disease has to be high among the population screened.
This relates to the relative costs of the screening program in relation to the number of cases detected and to positive predictive value. It is important to understand that a Pap smear may be referred to as "abnormal," but may not mean that a person has cervical cancer.
Some organizations also recommend HPV human papilloma virus screening in certain populations during the Pap smear. This blood test measures the prostate specific antigen PSA levels in the blood. Antigens are any substances that evoke responses from a person's immune system.
The prostate specific antigen levels can be elevated in the presence of prostate cancer. However, it is important to understand that other benign prostate conditions may also elevate PSA, such as benign prostatic hyperplasia BPH , which is noncancerous swelling of the prostate. The pros and cons of PSA screening should always be discussed with your healthcare provider before testing. Some of the cons include unnecessary testing and procedures, unnecessary costs, and significantly increased anxiety.
Many organizations, including the USPSTF, recommend mammography screening for breast cancer every 1 year to 2 years after age This test is done in conjunction with a clinical breast exam. Many organizations, including the USPSTF, recommend screening for colon cancer or colon polyps at age 50, earlier if you have a family history or other risk factors. Characteristics that make a disease amenable to screening include a significant negative impact on health, an identifiable asymptomatic period, and improved outcomes with early intervention.
A useful screening test must have sensitivity and specificity for the disease being screened. It also must be cost effective and acceptable to patients. Sensitivity, specificity, and disease prevalence all interact to determine a test's positive predictive value--the likelihood that a positive test result indicates that the disease is present.
0コメント