Validity of Psychiatric Rating Scales

False Positives & Negatives

Whether it is a psychiatric rating scale or a laboratory test in medicine, tests need to be valid, i.e. they need to measure what they are attempting to measure. However, all tests have a certain amount of false positives (the test measures positive for a disorder when a person does not have the disorder) and all tests have a certain number of false negatives (the test measures negative for a disorder when the person actually has the disorder).

FALSE POSITIVE = TEST MEASURES POSITIVE WHEN A PERSON DOES NOT HAVE THE DISORDER

FALSE NEGATIVE = TEST MEASURES NEGATIVE WHEN A PERSON HAS THE DISORDER

An ideal test is one which false positives and false negatives are minimized.

Sensitivity & Specificity

A statistical analysis is performed on tests to evaluate their validity, reported as the test’s sensitivity and specificity:

SENSITIVITY = THE PERCENT OF A GROUP OF PEOPLE WITH THE DISORDER CORRECTLY IDENTIFIED BY THE TEST AS HAVING THE DISORDER.

SPECIFICITY = THE PERCENT OF A GROUP OF PEOPLE WITHOUT THE DISORDER CORRECTLY IDENTIFIED BY THE TEST AS NOT HAVING THE DISORDER.

In the rating scale section for parents, children, adolescents, adults, teachers, and clinicians, the sensitivity and specificity of the rating instruments will be reported (as available) so that you can evaluate the validity of the rating scales you are using.

For some disorders, one rating scale may have the best sensitivity and specificity if done by a particular rater, e.g. the parent rater with the Mood Disorder Questionnaire—Adolescent Version, and therefore the results of that rating scale would have the best reliability for the disorder it is screening for.

Having multiple observers provide ratings on a person for a disorder, i.e. the ratings of a parent, teacher, clinician and person) may also improve the overall confidence in the ratings, e.g. childhood depression.