2017 brainfood stillwell aquickertestfordepression 883x432 1

A quicker test for depression

16 October 2017

The article at a glance

When it comes to assessing depression, the usual paper-based questionnaires are long and frustrating for patients. University of Cambridge researchers found that …

When it comes to assessing depression, the usual paper-based questionnaires are long and frustrating for patients. University of Cambridge researchers found that computerised adaptive tests (CAT) are quicker to complete and more accurate.

The silhouette of a woman sitting in a dark tunnel with a light at the end.

Computerised adaptive testing (CAT) can make depression screening more accurate by selecting only the most informative questions and thus shortening the assessment procedure, University of Cambridge researchers found in a study involving more than 2,000 people from the United States.

A shorter and more informative assessment can be more acceptable to patients, because many find responding to the exact same questions time and time again a tedious exercise. CAT uses people’s responses to previous questions to select the most useful question to ask next, using a specially calibrated algorithm. Using this technique to test depression “can be more accurate and efficient” than fixed-length tests, while reducing the length of assessments by as much as 82 per cent, found the study just published in the Journal of Medical Internet Research.

The study is co-authored by three researchers at the Psychometrics Centre in Cambridge Judge Business School – Dr Chris Gibbons, a Research Fellow at University of Cambridge Institute of Public Health; Aiden Loe, a PhD student in psychology at the University of Cambridge; and Dr David Stillwell, University Lecturer in Big Data Analytics & Quantitative Social Science at Cambridge Judge Business School and Deputy Director of the Psychometrics Centre.

“By making depression assessments more relevant and less burdensome, we hope that patients will be happier to share this information with the healthcare professionals who can offer help,” said co-author Dr Chris Gibbons.

The Centers for Disease Control and Prevention in the United States says that “depression is the most common type of mental illness, affecting more than 26 per cent of the US adult population. It has been estimated that by the year 2020, depression will be the second leading cause of disability throughout the world, trailing only ischemic heart disease.”

The Cambridge study examines how CAT can shorten the Center for Epidemiologic Studies Depression Scale (CES-D), the most widely used global instrument to test for depression. That test uses a 20-item scale that asks people how often in the past week (from rarely or none to most or all of the time) they experienced symptoms such as feeling lonely, poor appetite and restless sleep. It’s commonly accepted that people who score 16 or above on the test’s 0 to 60 scale are likely to be clinically depressed.

Dr David Stillwell
Dr David Stillwell

The Cambridge study’s 2,060 participants, all from the United States and ranging from ages 19 to 77 (with an average age of 26), submitted responses to the CES-D test through myPersonality, a Facebook application developed by co-author Dr David Stillwell that allows users to take online psychometric tests.

The Cambridge study finds that while the current CES-D is well validated, the use of CAT can reduce the number of items without sacrificing accuracy. CAT can also adjust for demographic differences in interpreting certain items, as well as potential reliability issues arising from some items being too similar.

Questionnaires like the CES-D are commonly used in clinical research as well as clinical practice. In clinical research, depression questionnaires are often included alongside assessments for other mental health and psychological phenomena. This often results in a large “pack” of questionnaires, which can take over an hour to complete.

Aiden Loe, lead author of the study, notes the difficulties some people face when completing these intimate assessments, stating that “completing questionnaires of a personal nature can be difficult and burdensome for people who have mental health problems. Our computer adaptive testing algorithm can reduce redundancy and burden associated with fixed-length questionnaire assessment by selecting only the most relevant questions to ask.”

The study is entitled “Computerised adaptive testing provides reliable and efficient depression measurement using the CES-D scale”.

“The CES-D CAT is a precise and efficient tool for screening depressive symptoms,” the study concludes. “Furthermore, the measurements provided by CAT are more likely to result in more meaningful research conclusions than classical approaches.”