Modified Adolescent Outpatient Screening Improves Suicide Risk Prediction

Primary care visits provide an opportune time to screen adolescent patients for depression and suicide risk.

Editor's Note: Where Discovery Leads is a multimedia storytelling project that delves into key research themes at Children's Hospital of Philadelphia Research Institute. This is part four of a five-part series that focuses on the scientific studies aimed at addressing behavioral and mental health.

Complementing the series is a brief video featuring Tami Benton, MD, Psychiatrist-in-Chief, as she highlights CHOP's behavioral health lifespan research, which is moving treatment closer to personalized medicine.

Researchers at Children's Hospital of Philadelphia are leveraging the power of the electronic health record (EHR) to help predict suicide risk among young patients.

As part of the Enhancing Triage and Utilization for Depression and Emergent Suicidality (ETUDES) Center, investigators and clinicians at CHOP are partnering with the University of Pittsburgh to enhance the identification and management of adolescent depression and suicide risk in pediatric primary care.

"Our Primary Care Network does an exceptional job screening patients for depression, screening up to 70,000 adolescents each year," said Jami Young, PhD, Endowed Chair and the Associate Chair of Research in the Department of Child and Adolescent Psychiatry and Behavioral Sciences and PolicyLab faculty member, who co-leads the ETUDES Center.

Primary care is a key access point in identifying young patients at risk and triaging them to receive risk-level appropriate interventions. While patients aren't typically coming to their regular well visit intending to discuss depressive feelings, the visit provides an opportune time to screen for depression.

On its Screening for Suicide Risk in Clinical Practice page, the American Academy of Pediatrics (AAP) cites a study that found most people who die by suicide visit a healthcare provider in the weeks or months leading up to their death. The AAP recommends universal screening for depression in patients 12 years of age or older.

The Patient Health Questionnaire-9 (PHQ-9) is widely used as a self-report screening tool to identify depression in adolescents and adults during primary care visits. Item 9 on the PHQ-9, which assesses suicidal ideation and thoughts of harming oneself, correlates with increased risk of suicide attempts and deaths in adults and adolescents. However, more than one-third of individuals who made suicide attempts or died by suicide within 30 days of completing the screening reported no suicidal ideation.

Given the increasing rates of suicide and suicidal adolescents in the United States, researchers at the University of Pittsburgh led by David Brent, MD, Endowed Chair in Suicide Studies, are partnering with CHOP researchers including Dr. Young and Fuchiang (Rich) Tsui, PhD, Endowed Chair in Biomedical Informatics and Entrepreneurial Science in the Department of Anesthesiology and Critical Care Medicine and the Department of Biomedical and Health Informatics, on the National Institute of Mental Health-funded ETUDES center to help address youth suicide.

Dr. Tsui specializes in working with the large volume of longitudinal data housed within CHOP's EHR. He taps into this valuable resource to find solutions for the challenges facing clinicians and researchers, to improve care and outcomes for patients of all ages.

The collaboration resulted in the first study comparing the conventional PHQ-9 with the PHQ-9: Modified for Teens (PHQ-9M) to determine the value of the additional items included on the PHQ-9M for predicting suicide attempts. While the PHQ-9 and PHQ-9M are similar, the PHQ-9M keeps the original nine core items and includes four supplemental items that specifically ask about depression during the past year, difficulty in completing everyday tasks and interactions, serious suicidal thoughts within the past month, and suicide attempts throughout the patient's lifetime.

"I think some clinicians are uncomfortable asking about suicide, but using the expanded screening tool, which the patient often completes on a tablet, we're seeing a real benefit," said Dr. Young, a study co-author who has been analyzing PHQ-9M screening at CHOP for several years.

Based on their study findings, Dr. Tsui, Dr. Young, and their collaborators suggest that the supplemental items on the PHQ-9M screening allow for better prediction of youth suicide attempts compared with screening using the PHQ-9 in primary care.

"In conjunction with the historical data for the patient, we hope this work provides the clinicians with more comprehensive information about their patients and informs clinical decision-making," Dr. Tsui said.

Predictive Modeling for Suicidal Behavior

The current CHOP-led study evaluated the effectiveness of the PHQ-9 compared to the PHQ-9M to predict the risk of suicide attempts in a retrospective cohort of 130,028 CHOP patients, 12 to 17 years old, who were screened during primary care visits between 2016 and 2022.

"We used multivariate analysis of electronic health record models to determine which items on the screening tool correlated with the likelihood of suicide risk," Dr. Tsui said. "Results indicated that the modified version improved the prediction of suicide attempts when screening adolescents."

Study findings were published in JAMA Network Open.

This cohort study of adolescent screenings demonstrated that the supplemental items on the PHQ-9M screening — in particular, an item that asks about lifetime suicide attempts and another that assesses suicidal ideation within the past month — improved the prediction of youth suicide attempts compared with screening using the PHQ-9 alone. Additionally, an item that asks about persistent depression over the past year was also predictive of subsequent suicide attempts, emphasizing the importance of providing mental health services to adolescents who report ongoing symptoms.

"Most of the items on the PHQ-9 ask about how the adolescent is feeling in the past two weeks, and we know that adolescents' moods change over time," Dr. Young said. "These results suggest that persistent depression, which we know clinically is when people start to experience hopelessness, puts youth at increased risk for suicide."

Over 80% of adolescents attending a primary care well visit at CHOP are screened for depression.

"This is higher than rates reported in other healthcare systems and likely considerably higher than screening rates in primary care offices that do not have screening embedded in electronic health records," Dr. Young said.

In choosing to use the PHQ-9M as its screening measure, CHOP has long been at the forefront of not only screening for depression but also identifying adolescents at risk for suicide.

Dr. Tsui expressed appreciation for the proactive use of screening and documentation in the CHOP Primary Care Network, without which this work would not be possible. Arcus at CHOP provided data support. The Pediatric Research Consortium, the primary care practice-based research network at CHOP, has also been instrumental in supporting this work and the ongoing efforts in the ETUDES Center.

Behavioral Health Lifespan Research: Moving Closer to Personalized Medicine. Access the video transcript.