76 The predictive validity of MyHEARTSMAP for psychosocial screening in the emergency department. (19th August 2020)

• Record Type: Journal Article
• Authors: Atwal, Amanbir; Doan, Quynh; Wright, Bruce; Hankinson, Elizabeth; Virk, Punit; Azizi, Hawmid; Stenstrom, Rob; Black, Tyler; Gokiert, Rebecca; Newton, Amanda
• Is Part Of: Paediatrics & Child Health. Volume 25(2020)Supplement 2
• Page Start: e31
• Page End: e32
• Publication Date: 2020-08-19
• Subjects: Pediatrics -- Periodicals; Children -- Health and hygiene -- Periodicals; 618.92
• DOI: 10.1093/pch/pxaa068.075
• Languages: English
• ISSNs: 1205-7088

Abstract

Abstract: Background: Mental health concerns in childhood and adolescence are prevalent, affecting nearly one million Canadian youth. In the absence of screening, up to 98% of these concerns can go undiagnosed, leading to significant health, educational, and social consequences. Consequently, the American Academy of Pediatrics recommends the development of screening tools to facilitate early identification and access to treatment. The Emergency Department (ED) represents a unique environment to implement such universal screening, as it is immediately accessible and may be the only point of contact for some vulnerable youth with undiagnosed illness. However, there are few existing instruments which take into account commonly cited barriers such as time constraints, disruption of ED flow, limited resources, and patient privacy. Objectives: To facilitate efficient screening with minimal impact on ED flow, our team developed MyHEARTSMAP, an electronic self-administered screening tool. The tool is adapted from HEARTSMAP, a previously validated computerized assessment and management tool used by ED clinicians. MyHEARTSMAP has previously been evaluated for face validity and inter-rater reliability. Here, we measured the sensitivity and specificity of MyHEARTSMAP in identifying mental health concerns in youth. Design/Methods: A prospective cohort study was conducted at two tertiary care pediatric EDs. Eligible youth aged 10-17 years presenting for a non-mental health complaint were invited to self-screen using MyHEARTSMAP. An accompanying parent/guardian could also complete an assessment of their child. The sensitivity and specificity was measured as the proportion of screened youth with mental health concerns identified through self-assessment by MyHEARTSMAP compared to assessment performed by a clinician (criteron standard). Results: 760 youth and/or parents completed the study intervention. The sensitivity at identifying any psychiatric concerns was comparable between youth and guardian assessments: 92.7% (95%CI: 89.1, 95.4%) and 93.1% (95%CI: 89.5, 95.8%) respectively. The specificity at identifying youth without any psychiatric issues was also comparable between youth and their guardians: 42.2% (95%CI: 37.3, 47.3) and 37.0% (95%CI: 32.2, 42.1), respectively. Conclusion: MyHEARTSMAP is sensitive for identifying youth with mental health concerns. While it showed only modest specificity, false positives were almost entirely (98%) mild issues identified by youth and deemed to be normal by clinicians. This would not place a burdensome demand on mental health services and could be effectively assessed without specialized psychiatric training. Thus, MyHEARTSMAP may be an effective tool for early identification and management of mental health concerns.