Presentations by pediatric patients for mental health concerns have been increasing in both Canada and the US.1,2 The emergency department has been considered a mental health safety net—one that has been stretched to its limit.3,4 Consistent and expedient assessment of pediatric patients who present to the emergency department with a mental health crisis is challenging. A standardized clinical assessment tool was needed because physicians had indicated that there was no such screening tool.3,5,6
A new screening tool based on the previously developed mnemonic designed to assist physicians with obtaining a psychosocial history from adolescents as part of a routine visit was developed by researchers at the Children’s Hospital of Eastern Ontario.7 The HEADS mnemonic has different variations, such as the HEADDS or HEEADSSS, but with commonality among all of them.8-10 The mnemonic generally stands for key areas, such as home, education, activities/ambition, drugs and drinking, sexuality, suicide and depression, and safety.
The HEADS-ED is based on previous research on longer screening tools, such as the Childhood Acuity of Psychiatric Illness (CAPI) scale and the Child and Adolescent Needs and Strengths-Mental Health (CANS-MH 3.0) scale.11 HEADS-ED stands for Home, Education, Activities and peers, Drugs and alcohol(Drug information on alcohol), Suicidality, Emotions and behaviours, and Discharge resources. It uses an embedded scoring system with points for each variable (0 = no action needed; 1= needs action but not immediately; and 2 = needs immediate action).
Initial findings indicate that HEADS-ED is psychometrically sound with evidence of criterion, concurrent and predictive validity, and interrater reliability.7 The HEADS-ED was correlated with a comprehensive clinician rating of mental health strengths and needs (CANS-MH 3.0) as well as ratings of depression by youths using the Children’s Depression Inventory (CDI).12,13 The study also supported the predictive validity of the tool. The total score from the HEADS-ED indicated meaningfully and statistically different mean scores for patients who were referred for admission to an inpatient psychiatric unit (above the 75th percentile) and those who were referred for consultation (above the 50th percentile).
Using an algorithm of a total HEADS-ED score of greater than 7 and a suicidal risk factor of 2, we determined a sensitivity of 81.8 and a specificity of 87 for predicting admission. Furthermore, a receiver operating characteristic (ROC) curve procedure was used to analyze the tool for admission decisions. Results indicated area under the ROC curve of 0.817, P < .01, demonstrating that the tool had good detection of indicators of admission to inpatient psychiatry.7
In the following fictitious vignette, we can see how the tool can be useful in screening a patient who presents in mental health crisis.
CASE VIGNETTE
Sara’s parents called the police because of her violent outburst. When Sara’s parents refused to allow her to meet friends on a school night, she went into a rage, started screaming obscenities, locked herself in the bathroom, and said she wished she had never been born. When the police arrived, Sara seemed to be withdrawn, did not make eye contact, and was generally nonresponsive; she was taken to the emergency department at a local hospital.
Sara is 14 years old and is currently living with her parents and 2 younger siblings. Over the past 6 to 8 months, she has become increasingly withdrawn and sullen. Sara reports occasional weekend use of marijuana. Sara also has been having academic problems. She is failing all her classes; she routinely cuts school and when she does go, she does not participate.
Sara has recently made new friends at school and stopped seeing her old friends. She has changed the way she dresses, the way she speaks, and the music she listens to. Problematic behaviors have escalated. She ignores her curfew and is “sneaking” out of the house.
She has no history of medication or mental health treatment; nor is there a history of suicidal ideation or behavior.
Completed HEADS-ED
The Figure presents the emergency department assessment for Sara using the HEADS-ED. The HEADS-ED gives a concise picture of the main concerns for this patient. It also gives us a total score that can indicate overall severity of symptoms. The tool reminds physicians of the key elements in gaining a broad picture of the patient. On the basis of this score, the emergency department physician can make determinations as to disposition and follow-up recommendations. In this case, although the patient scored a “9,” there was no indication of suicidality so the patient was not referred for inpatient psychiatric admission. However, since the patient’s profile indicates severe symptoms, the emergency department physician targeted appropriate outpatient follow-up.
Conclusion
A Web site (www.heads-ed.com) that uses the HEADS-ED tool in an interactive way that will summarize the patient’s main symptoms is currently under development. This Web site will provide information on useful resources within the community that match to various levels of the patient’s symptoms. The future vision for this media would be linking the symptom profile to disposition recommendations, which may include:
• Referral for psychiatric consultation for possible admittance to inpatient services, or
• Outpatient follow-up recommendations tailored to available community resources
While the HEADS-ED has strong predictive validity for consultation and admission decisions, more research is needed to support this as a tool on which to base clinical decisions. Therefore, it should be considered an informative measure that helps guide clinical decision making, assists with communication, and aids in directing the interview.
