When comparing the diagnostic sensitivity and specificity of two commercial tests, it is essential to go beyond the percentages given in their instructions for use. These parameters, although essential, may have been calculated under very different conditions, making a direct and reliable comparison difficult. Variables such as the number and type of sample, the reference method used or even the sample group of the study can significantly influence the results.
In this article we will discuss what aspects should be taken into account to correctly interpret these data and make purchasing decisions based on solid evidence, thus minimising the risk of choosing a less validated or clinically less reliable test.
What do analytical sensitivity and specificity mean in diagnostic tests?
Sensitivity and specificity are statistical measures that indicate the probability that a test correctly identifies whether or not a person has a disease. They are intrinsic characteristics of the test and do not depend on the prevalence of the disease.
Diagnostic sensitivity: what it measures and why it is key
Sensitivity indicates the ability of a test to correctly detect diseased individuals. That is, it measures how many true positives it identifies. A test with low sensitivity can give false negatives, which means missing patients who are actually infected.
Diagnostic specificity: how to avoid false positives
Specificity measures the ability of the test to correctly identify healthy people, avoiding false positives. This is especially important in order not to generate unnecessary alarms or unjustified treatment.
Factors affecting the sensitivity and specificity of diagnostic tests
- Reference technique used: The choice of reference technique directly affects the sensitivity and specificity obtained. According to CLSI EP12-A2, a test can be compared to another qualitative method or to one that has diagnostic criteria.
For example, comparing a rapid test to a molecular test (which is much more sensitive) will show a lower correlation in terms of sensitivity and specificity than if compared to another rapid test. However, if the result of the rapid test is close to that obtained by the molecular test, it would be a good indication of higher reliability. - Type of sample used: The type of sample used (nasal swab, nasopharyngeal swab, oropharyngeal swab, nasopharyngeal aspirate or wash, etc.) also impacts sensitivity and specificity. This is because certain analytes are more readily detectable in certain body sites.
Therefore, the sensitivity and specificity of the same test may vary from one sample type to another. - Number of samples: The reliability of sensitivity and specificity data improves significantly as the number of samples in the validation increases.
According to CLSI EP12-A2, it is recommended to test at least:- 50 positive specimens (both with the test method and with the comparative method) to assess sensitivity
- 50 negative specimens (with the comparative method) to assess specificity.
In practice, it is considered appropriate to use at least 100 specimens, adjusting the ratio of positives to negatives according to the estimated prevalence of the disease.
For example, if the prevalence is 5% in a population of 100 samples, 5 positive and 95 negative samples are expected.
- Sample group and study conditions: Two different sample groups may yield different results even if the same reference techniques and sample types are used. Factors such as the clinical status of the patients or where they come from may change the test performance.
The most reliable way to compare two tests is by parallel evaluation with the same samples. If this is not possible, comparisons with a common, validated reference technique should be used.
Specific considerations for molecular diagnostic products
In the case of molecular diagnostic tests, the technique allows the detection of a specific nucleotide sequence. This implies that for those products that detect both a micro-organism and possible mutations in its DNA, the following should be taken into account:
- The specificity and sensitivity shown in detecting the micro-organism.
- The concordance with genotyping to detect possible mutations.
For example, to compare two products that detect the tuberculosis bacterium and its resistance, we must first consider the results of sensitivity and specificity to detect the bacterium and, subsequently, the concordance of the genotyping to find out if the product is capable of correctly detecting the mutations that make it resistant to certain families of antibiotics.
In the case of two products that detect mutations in the human gene, for example to identify the risk of celiac disease, only the concordance with the genotyping with respect to the reference technique will be taken into account.
On the other hand, if the test works with direct sample, it is necessary to validate its specific detection capacity in that type of sample. A direct comparison between a test using extracted nucleic acid and a test using a direct sample would not be correct, as the former will always have a higher sensitivity.
It is also important that both products are validated for the same sample type for the comparison to be valid (e.g. DNA extracted from nasopharyngeal aspirates in both cases).
Which diagnostic test to choose? Key recommendations
Beyond price, choosing an appropriate diagnostic test involves assessing:
- Its technical validation according to recognised standards (such as CLSI EP12-A2).
- The type of sample it has been validated for
- The reference technique used in the study
- The size of the sample group
- The reliability of the manufacturer and its regulatory compliance
The most reliable option is always to compare two tests with the same samples and under the same conditions, or at least to verify that they have been validated against the same reference technique.
Understanding how sensitivity and specificity are measured and what factors affect these parameters allows purchasing decisions to be made based on robust and comparable data. This is especially relevant for both healthcare professionals and distributors who want to offer reliable products with high clinical value.
If you are still in doubt, you can contact our team for personalised technical advice. We will help you choose the most suitable option according to your needs and regulatory requirements.
