The Dunn Worry Questionnaire and the Paranoia Worries Questionnaire: new assessments of worry
Freeman D., Bird JC., Loe BS., Kingdon D., Startup H., Clark DM., Ehlers A., Černis E., Wingham G., Evans N., Lister R., Pugh K., Cordwell J., Dunn G.
<jats:title>Abstract</jats:title><jats:sec id="S0033291719000588_sec_a1" sec-type="other"><jats:title>Background</jats:title><jats:p>The cognitive process of worry, which keeps negative thoughts in mind and elaborates the content, contributes to the occurrence of many mental health disorders. Our principal aim was to develop a straightforward measure of general problematic worry suitable for research and clinical treatment. Our secondary aim was to develop a measure of problematic worry specifically concerning paranoid fears.</jats:p></jats:sec><jats:sec id="S0033291719000588_sec_a2" sec-type="methods"><jats:title>Methods</jats:title><jats:p>An item pool concerning worry in the past month was evaluated in 250 non-clinical individuals and 50 patients with psychosis in a worry treatment trial. Exploratory factor analysis and item response theory (IRT) informed the selection of scale items. IRT analyses were repeated with the scales administered to 273 non-clinical individuals, 79 patients with psychosis and 93 patients with social anxiety disorder. Other clinical measures were administered to assess concurrent validity. Test-retest reliability was assessed with 75 participants. Sensitivity to change was assessed with 43 patients with psychosis.</jats:p></jats:sec><jats:sec id="S0033291719000588_sec_a3" sec-type="results"><jats:title>Results</jats:title><jats:p>A 10-item general worry scale (Dunn Worry Questionnaire; DWQ) and a five-item paranoia worry scale (Paranoia Worries Questionnaire; PWQ) were developed. All items were highly discriminative (DWQ <jats:italic>a</jats:italic> = 1.98–5.03; PWQ <jats:italic>a</jats:italic> = 4.10–10.7), indicating small increases in latent worry lead to a high probability of item endorsement. The DWQ was highly informative across a wide range of the worry distribution, whilst the PWQ had greatest precision at clinical levels of paranoia worry. The scales demonstrated excellent internal reliability, test-retest reliability, concurrent validity and sensitivity to change.</jats:p></jats:sec><jats:sec id="S0033291719000588_sec_a4" sec-type="conclusions"><jats:title>Conclusions</jats:title><jats:p>The new measures of general problematic worry and worry about paranoid fears have excellent psychometric properties.</jats:p></jats:sec>