Quiz 3: Diagnostic Research

when looking at diagnostic research, what can we look at to assess credibility?

the who (subjects), the what (methods), the quality, and the reproducibility

what do we think about when evaluating the who?

did the investigators include subjects with all levels or stages of the condition being evaluated by the index test (measure)? determines the usefulness of the test (measure) with respect to distinguish between important gradation in teh diagnosiss or impairement

what should we think about when looking at the what?

did the investigators eval the reliability of the index diagnostic test (measure)?, was there a comparison to the “gold” standard test? verifies the test does what its supposed to do, commonly used gold standards are radiographs, surgical exploration, and test or measure with previously demonstrated consistency and usefulness

what should we consider when thinking about the quality?

did all subject undergo the measure of interest and the gold standard test? address the issue of bias due to manipulation of th esample, were the individuals interpreting each test’s results unaware of the other test’s measures? was the time between application of the index test and the gold standard comparison diagnostic test short enough to minimize the opportunity for change in the subjects’ condition?

what should we think about when considering reproducibility?

was the study repeated on a second set of subjects?

what is reliability?

the stability of a measure, influence of random error, and sources of variability (observer, subject, instrument, and environment are possibilities)

what is validity?

the measure represents what its supposed to represent, could be an influence of systematic error, have the same sources of variability as reliability

what are additional evaluative criteria of diagnostic tests?

sensitity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio

what is the definition of specificity?

confirming where the proportion of false positives is low (SpPin: high specificity, positive test result, and rule in disorder)

what is the definition of sensitivity?

screening test where the proportion of false negatives is low (SnNout: hihg sensitivity, negative test result, rule out disorder)

what is the negative predictive value?

the probability that a person with a negative (normal) test result is truly free of disease. ability of the test to correctly determine the % of people without the disorder from all the people with a negative test result

what is positive predictive rule?

value is an answer to the question: if my patient’s test result is positive, what are the chances my patient does have the disease?. the ability of the test to correctly determine the percentage of people with the disorder from all the people with positive test results

what does is mean for a test to have a high positive predictive value?

It means that when a test result is positive, there is a high probability that the individual truly has the disease. This indicates the test's effectiveness in accurately identifying those with the disorder among those with positive results.

what does is mean for a test to have a lower negative predictive value?

It indicates that when a test result is negative, there is a higher probability that the individual may still have the disease. This suggests a reduced effectiveness of the test in correctly identifying those without the disorder among those with negative results.

what does it mean when a test has low sensitivity?

It means that the test is less likely to correctly identify individuals who have the disease, leading to a higher number of false negatives. A low sensitivity reduces the test's ability to detect the disorder accurately.

what does a test with high specificity mean?

It means that the test is effective in correctly identifying individuals who do not have the disease, resulting in a low number of false positives. High specificity indicates the test's ability to accurately confirm the absence of the disorder among those with negative results.

what are other operational definitions?

positive likelihood ratio (R+) and negative likelihood ratio. (R-)

what is the positive likelihood ratio?

the likelihood that a positive test result was observed in a person with the disorder vs in a person without the disorder of interest. It quantifies how much more likely a positive test result is in someone with the condition compared to someone without it, helping to assess test performance.

what is the negative likelihood ratio?

the likelihood that a negative test result observed in a person with the disorder vs in a person without the disorder. It helps assess the test's effectiveness in ruling out the disease.

what values of + LR and - LR correlate to a large and conclusive change?

• + LR greater than 10

• - LR less than 0.10

what values of + LR and - LR correlate to a moderate change?

• +LR = 5-10

• -LR = 0.10 - 0.20

what values of + LR and - LR correlate to a small but sometimes important change?

• +LR = 2-5

• -LR = 0.20-0.50

what values of + LR and - LR correlate to a negligible change?

• +LR = 1-2

• -LR = 0.50-1.0

what can CPRs be used for?

diagnostic categories, prognostic estimates, and response to treatment (interventional)