Evaluating psychological distress data

Area of Science:

• Psychiatry
• Psychometrics
• Public Health

Background:

• The Kessler k₆ scale is a widely used measure of psychological distress.
• Item response theory (IRT) offers an alternative method for analyzing psychological distress data.

Purpose of the Study:

• To compare the Kessler k₆ scale with an IRT-derived score (θ*) for measuring psychological distress.
• To evaluate the utility of both measures in predicting treatment program participation.
• To explore alternative methods for analyzing the individual items comprising the Kessler k₆ scale.

Main Methods:

• Analysis of the 2014 National Survey of Drug Use and Health data.
• Application of count models for the Kessler k₆ scores.
• Application of item response theory (IRT) models to generate an alternative distress score, θ*.
• Examination of ordered probability models for individual item analysis.

Main Results:

• The Kessler k₆ score and the IRT-derived score (θ*) showed low correlation and different distributions, with k₆ indicating better mental health.
• The k₆ score was a significantly better predictor of treatment program participation than θ*.
• Analyzing individual items using ordered probability models provided superior results for identifying determinants of psychological distress compared to using the aggregated k₆ score.

Conclusions:

• The Kessler k₆ scale is a more effective and simpler diagnostic tool for psychological distress than IRT-derived scores.
• IRT models have a limited role in analyzing psychological distress data, particularly for predicting treatment seeking.
• Researchers should consider analyzing individual items of psychological distress scales using appropriate statistical models for deeper insights into contributing factors.