Capstone Abstract--Joel A. Fein

Subsequent Psychiatric Diagnoses in Adolescents withSomatic Diagnoses in the Emergency Department

Joel A. Fein, M.D., M.P.H., Capstone Advisor: David Mandell, Sc.D.

My capstone experience was designed to introduce me to health services research and large database analysis.  For my specific project, I sought to determine the association between somatic diagnoses and subsequently diagnosed psychiatric disorders in adolescent ED patients. 

Many adolescents who present to the Emergency Department (ED) without overt psychiatric signs and symptoms are nonetheless at risk for psychiatric disorder.  It has been suggested that somatic complaints, such as chest pain, abdominal pain, headache, fatigue, and weakness, may be indicators of emotional or behavioral disorders. I hypothesized that adolescents who receive somatic diagnoses in the ED, but were previously unidentified as having a psychiatric disorder, would be more likely than those without somatic diagnoses to have a subsequent mental health diagnosis within one year. I used a retrospective cohort design to measure the association between somatic ED visits and discharge diagnoses with psychiatric treatment.  The sample included all Medicaid recipients in Pennsylvania, who were between 11 years and 16 years of age on January 1, 1995.

Of the 6165 subjects who had an ED visit in 1995, 609 (9.9%) had a somatic discharge diagnosis. Similar proportions of somatic subjects (14%, 95% CI 11, 17) and non-somatic subjects (15%, 95% CI 14, 16) had at least one psychiatric diagnosis in the year after their ED visit.  Logistic regression analysis, controlling for demographic differences, found no independent effect of a somatic ED diagnosis on subsequent psychiatric diagnosis.  Younger adolescents (11-13 years old) were less likely to have a somatic diagnosis in the ED (8.4% vs. 12.9%, p=0.001). Latino adolescents were 2.3 times (95% CI 1.8, 2.9) more likely than Non-Latinos to receive a somatic diagnosis in the ED.  Of the 115 subjects diagnosed with a psychiatric disorder during the 1995 ED visit, 103 (89.6%) of these subjects received at least one psychiatric diagnosis in the year following that visit.

We concluded that adolescents with somatic diagnoses in the ED, and who have not had prior mental health diagnoses, have similar subsequent psychiatric diagnoses as other ED patients. In contrast, the vast majority of adolescents who were newly diagnosed with psychiatric disorders in the ED had subsequent mental health diagnoses in the subsequent year.

My exposure to the inner workings of this health services research project placed me in an environment that was foreign to me but commonplace to many of my public health and medical colleagues.  I built a vocabulary from which to draw when reviewing health services literature and grants, and finally comprehend the labor involved in refining a research question - from the outside this had seemed simple and mundane.  As this was a major gap in my prior public health experiences, I am grateful for this opportunity and believe that I will use these skills in the future to collaborate on similar projects.  Finally, I believe that the project’s focus on how various demographic and situational factors impact the outcomes of the subjects alters my view of the adolescent patients in my roles as an attending in the CHOP Emergency Department.  For example, I currently consider how their Medicaid eligibility may be more transient than I previously believed, and also look more carefully at how chief complaints relate to eventual diagnoses. I am sure that more aspects of the research will impact my clinical care as time passes.