Sensitivity and specificity are critical concepts in medical testing and diagnosis. Their application is vital to creating the most accurate tests possible, which lead to better diagnoses and more effective treatment. This article explores the difference between sensitivity and specificity and how to apply them in medical testing.
Sensitivity vs Specificity: Definitions
Sensitivity and specificity measure a test’s ability to accurately diagnose a patient as either diseased or disease-free. In other words, they do not analyze a patient’s health but rather the accuracy of the test itself.
- Sensitivity: Measures a test’s ability to accurately diagnose a patient as diseased.
- Specificity: Measures a test’s ability to accurately diagnose a patient as disease-free.
Both sensitivity and specificity are expressed in percentages. For example, a test that accurately diagnoses 76 out of 100 diseased people as having that disease would have a sensitivity of 76%. A highly sensitive test rarely returns a false negative result, which means it usually recognizes when a patient has a disease. A highly specific test rarely returns a false positive result, which means it usually recognizes when a patient does not have a disease.
The idea is to develop a test that is both highly sensitive and highly specific. However, this is not usually possible. Typically, sensitivity and specificity are inversely proportional, meaning that as one increases, the other decreases. They do not depend on the prevalence of disease in the target population.
True and False Positives and Negatives
Understanding true and false positives and negatives is vital to understanding sensitivity and specificity. In medical testing and diagnosis, a positive result indicates that a patient does have a disease. A negative result indicates that a patient does not have a disease. A “true” result is an accurate result while a “false” result is an inaccurate result.
True and false positives and negatives can be summarized as follows:
- True positive: An accurate positive result (the test is correct in diagnosing the patient as diseased)
- True negative: An accurate negative result (the test is correct in diagnosing the patient as disease-free)
- False positive: An inaccurate positive result (the test is incorrect in diagnosing the patient as diseased)
- False negative: An inaccurate negative result (the test is incorrect in diagnosing the patient as disease-free)
Gold Standard Test
A gold standard test is a test known to be highly accurate in diagnosing patients. It is the best test for the disease currently available. All new tests for the disease are measured against the gold standard to determine their sensitivity and/or specificity.
Positive and Negative Predictive Values
Positive and negative predictive values determine what percentage of test results are accurate.
- Positive predictive value (PPV): The percentage of positive test results that are true positives
- Negative predictive value (NPV): The percentage of negative test results that are true negatives
For both PPV and NPV, the closer the percentage gets to 100, the more accurate the test is. Disease prevalence affects both PPV and NPV. Generally, the more prevalent the disease, the higher the PPV and vice-versa.
Using Sensitivity and Specificity in Medical Testing and Diagnoses
Sensitivity and specificity are essential to ensuring the accuracy of test results. Without accuracy, diseased patients risk forgoing vital treatment while disease-free patients may undergo unnecessary treatment. Mistakes like these are costly in terms of time, funding, and – most importantly – human lives.
Sensitivity and Specificity Equations
Medical professionals use two equations to determine and express sensitivity and specificity in terms of percentages. The equations are as follows:
- Sensitivity = # of true positives / (# of true positives + # of false negatives)
= probability of the test returning a true positive
- Specificity = # of true negatives / (# of true negatives + # of false positives)
= probability of the test returning a true negative
Sensitivity vs Specificity in Testing
As a general rule, highly specific tests effectively rule out disease when positive, and highly sensitive tests effectively rule out disease when negative. Two mnemonics are useful for expressing and remembering this:
- SpPIN: A highly sPecific test, if Positive, rules IN the disease
- SnNOUT: A highly SeNsitive test, if Negative, rules OUT the disease
When to Use Highly Sensitive vs Highly Specific Tests
Although the ideal is to have a test that is both highly sensitive and highly specific, this is rarely possible. Therefore, medical professionals must decide which kind of test will provide the best results. Pretest probability is useful in making this decision. Pretest probability is the likelihood before testing that a patient or group has the disease in question. This often depends upon the prevalence of the disease in the larger population. This is where the acronyms SpPIN and SnNOUT will come in handy for clinicians.
Below are more detailed discussions of these principles. However, in general:
- Lower pretest probability = a higher chance of false positives and a lower PPV
- Higher pretest probability = a higher chance of false negatives and a higher PPV
Sensitivity vs Specificity: Low Pretest Probability
If a person or population is unlikely to have a certain disease (low pretest probability), then the most efficient approach is to try to rule out the disease. Generally, low pretest probability means greater accuracy in ruling out than ruling in disease. This calls for the principle of SnNOUT, which relies on a highly sensitive test. Highly sensitive tests are good at determining if a patient has a disease and rarely return a false negative.
Because most people do not have the disease in question, only a few will turn out to be positive. Statistically, in this case, a highly sensitive test will only miss the disease in a few people (false negatives), whereas the high number of true negatives will result in a greater number of false positives with a highly specific test. Therefore, a highly sensitive test is the more appropriate option in cases of low pretest probability.
Sensitivity vs Specificity: High Pretest Probability
High pretest probability takes the principles of low pretest probability and inverts them. With high pretest probability, the goal is to “rule in” or identify those who do have the disease. Generally, high pretest probability means greater accuracy in ruling in than ruling out disease. This calls for the principle of SpPIN, which relies on a highly specific test. Highly specific tests are good at determining if a patient does not have a disease and rarely return a false positive.
Because the likelihood of the patient having the disease is high, a highly specific test that returns a positive result is likely to be accurate. However, in this case, a highly sensitive test poses a greater chance of a false negative than a true negative, which may necessitate more testing. Therefore, a highly specific test is the more appropriate option in cases of high pretest probability.
Improving Sensitivity and Specificity in Testing
It is important not to rely on sensitivity and specificity on their own. When considering which diagnostic test to use, it is important to factor in pretest probability as well as positive and negative predictive values. Combined, these will help give a more balanced picture and avoid unnecessary repeat testing to verify results. Additionally, improving attenuation correction and early event detection (EED) will increase the effectiveness of sensitivity and specificity.
Sensitivity and specificity are vital measures of a medical test’s accuracy. With more accurate tests, medical professionals can save time and money as well as improve outcomes for patients. They can also avoid unnecessary and often traumatic treatment for disease-free patients.
Thank you for reading! Have some feedback for us? Contact the AZ Animals editorial team.