Premium
The Digital Choice Anxiety Survey (DCAS): Initial development and validation
Author(s) -
Hall Aidan,
MakMcCully Rachel,
Glenn Jordan M.,
Madero Erica N.,
Fuseya Nami,
Gills Joshua L.,
Gray Michelle,
Bott Nick
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.044177
Subject(s) - anxiety , worry , clinical psychology , convergent validity , psychology , cronbach's alpha , generalized anxiety disorder , construct validity , psychiatry , psychometrics , prodrome , depression (economics) , internal consistency , psychosis , economics , macroeconomics
Background Psychiatric symptomatology, including anxiety, partially comprise the clinical prodrome of dementia. Moreover, increased anxiety may represent a risk factor associated with neurodegenerative disease. Efficient, longitudinal measurement of anxiety remains difficult with traditional paper‐pencil measures. As few digital measures of anxiety symptomatology exist, we aimed to develop and validate a digitally native, efficient and repeatable, anxiety survey to measure self‐reported anxiety in adults: the Digital Choice Anxiety Survey (DCAS). Method Development of the DCAS included three studies: Study one included adult participants (n=407) who completed a larger set of anxiety questions, alongside the Penn State Worry Questionnaire (PSWQ) and General Anxiety Disorder 7‐item (GAD‐7) survey. Factor analysis then reduced the larger set of DCAS questions and investigated convergent validity and accuracy of the DCAS. Study two recruited a sub‐sample of the original cohort (n=70) who completed the reduced DCAS survey to investigate convergent validity with the Geriatric Depression Scale (GDS). Study three recruited a separate sample of adults (n =58), who were administered the DCAS at baseline and again approximately two weeks after initial administration to examine test‐retest reliability. Result Study one: Mean participant age was 42.92±12.86 years. After removing questions with a Spearman’s correlation greater than or equal to 0.6, DCAS was shortened from 20 to 7 items. The 7 items showed strong internal reliability (Cronbach’s α=0.86), with a one factor solution found utilizing factor analysis (all factors r>0.56). There were positive associations with the PSWQ and GAD‐7 (r=0.82 & r=0.85), with overall accuracy of 0.93 for moderate anxiety on both PSWQ & GAD‐7. Study two: Mean participant age was 48.94±14.63 years. A positive association was found between the 7‐item DCAS and the GDS‐15 (r=0.58, p<0.01). Study three: Mean participant age was 54.79±12.53 years. Participants completed the DCAS an average of 11.7 days apart (SD=1.2) and showed robust test‐retest reliability (Pearson’s r=0.81). Conclusion This study provides evidence for psychometric validity and reliability of the DCAS. Follow‐up studies will investigate performance of the DCAS on clinically characterized populations and examine participant‐associated meta‐data.