Web-based screening for Panic Disorder: Validity of a single-item instrument

Abstract

Background: Panic Disorder (PD) is a common mental disorder with an important social and economic cost. Web-based screening tools for early detection of PD are useful for clinical and research purposes. However, there is a paucity of instruments that specifically measure PD online. The aim of this study is to analyze the validity of one item from the Web Screening Questionnaire designed to detect PD symptoms (WSQ-Panic).

Methods: A total of 171 participants completed the WSQ-Panic online and were assessed by telephone using the Structured Clinical Interview for DSM Disorders (SCID-I). The sensitivity, the specificity, predictive values (PPV, NPV), and area under the ROC curve (AUC) were calculated, and the optimal cut-off point was determined.

Results: The WSQ-Panic showed a sensitivity of 0.83 and a specificity of 0.74. The PPV was 0.46 and NPV was 0.94. The AUC was 0.82 (95% CI: 0.74-0.90), which indicates a moderate accuracy. The optimal cut-off point is ≥2.

Limitations: The representativeness of the sample is limited. All the interviews were conducted by phone. Six-month prevalence according to SCID-I criteria was considered, whereas the WSQ-Panic assesses current symptoms.

Conclusion: The WSQ-Panic accuracy is acceptable as an Internet screening tool, comparable to longer instruments for PD detection. This instrument is valid to quickly identify patients who suffer from panic symptoms, which can cause important distress and possibly lead to PD. It can also be very useful for screening participants in online self-help treatments and for research purposes.

Keywords: Common mental disorder; Panic Disorder; Single-item instrument; Web-based screening.