Citation

The Lancet Psychiatry. Lancet Psychiatry 2020; 7(1): 1.

Copyright

(Copyright © 2020, Elsevier Publishing)

DOI

10.1016/S2215-0366(19)30483-3

PMID

31860445

Abstract

In clinical science, what we choose to measure says a lot about our priorities and the underlying power-structures that shape research. Nowhere is this better reflected than in the choice of primary outcomes for randomised controlled trials (RCTs), which are still considered by many as the gold standard of evidence. Good confirmatory RCTs demand a strong hypothesis, and a strong hypothesis requires a specific endpoint to be chosen by the investigators. For some clinical fields and disease states, this choice is substantially harder to make than for others. Take depression for example, one of the most common mental illnesses and one of the largest contributors to morbidity and poor health in the world. With no reliable biomarkers for diagnosis, mental health clinicians are left to measure symptoms in the form of questions and answers covering topics like mood, appetite, and sleep, with different weights placed on different categories of cognitive and somatic symptoms depending on the assessment scale. Ultimately, the task with each scale is the same—to sum up the scores and reduce the mental health state of the person seeking care to a single number. For a clinical trial, the choice then becomes which scale to use.

As in any clinical field, there may be disagreement about the choice of outcomes, and the clinical experience and personal preferences of investigators no doubt play a role in the selection of a primary outcome. Studies also need to be comparable with previous work. Using the same scales facilitates these direct comparisons ...

Language: en