Citation

Haddad M, Tylee A. J. Sch. Health 2013; 83(1): 36-44.

Affiliation

Senior Lecturer in Mental Health, (mark.haddad.1@city.ac.uk), School of Health Sciences, City University London, Northampton Square, London EC1V 0HB, UK. Head, Section of Primary Care Mental Health, (a.tylee@iop.kcl.ac.uk), Chair, Kings Health Partners Primary Care and Public Health Research Group, Health Service & Population Research Department, PO 28 Institute of Psychiatry at King's College London, De Crespigny Park, London SE5 8AF, UK.

Copyright

(Copyright © 2013, American School Health Association, Publisher John Wiley and Sons)

DOI

10.1111/j.1746-1561.2012.00745.x

PMID

23253289

Abstract

BACKGROUND: Depression affects around 5% of adolescents and its identification and management is an important part of front-line professionals' roles. There are few validated measures of knowledge and skills in this area. We describe a multiple-choice question set to test nurses' depression knowledge and vignettes to examine case recognition skills. METHODS: A 24-item knowledge test and 12 vignettes were developed based on relevant literature and expert panel review. Three rounds of panel review assessed face and content validity and expert agreement of vignette depression status. The measures were piloted with 26 school nurses. Following amendments, administered to 146 school nurses. A depression attitude scale was used concurrently so that associations among knowledge, attitudes, and condition recognition could be explored. RESULTS: Readability for the knowledge test and vignettes was satisfactory. Item difficulty and discrimination indices for most knowledge questions were acceptable; overall, participants scored 50% correctly, with less than 5% unanswered. The panel reached 89% agreement about vignette depression status, and nurse participants' judgments of the vignettes achieved 65% sensitivity and 47% specificity. CONCLUSION: The study produced psychometrically tested instruments for measuring depression recognition and knowledge. There was evidence for content validity, and limited evidence of convergent validity from associations among measures. Some of the items may be modified, and a smaller set of vignettes having the best expert agreement may be useful in future research.

Language: en