RISK OF SUICIDE ATTEMPT IN NEWLY DIAGNOSED ADHD WITH AND WITHOUT STIMULANT TREATMENT WITHIN THREE MONTHS OF DIAGNOSIS

Introduction

ADHD is a common comorbidity of adolescent psychiatric inpatients who present with suicide attempts. Further elucidation of what factors may be related to or protective against suicide attempts would be greatly beneficial in this population. Recently reported research using population level data in Sweden indicates that drug treatment in ADHD improved suicidal behaviors. Multiple other studies agree that stimulant treatment may have a possible protective effect. However, these studies used a multivariate analysis approach and did not consistently account for confounding variables such as use of antidepressants and previous suicide attempts. The purpose of this study is to reexamine previous findings using a US based dataset with a specific focus on the child and adolescent psychiatric population. Further, we aim to use an alternative statistical approach to better clarify the impact of stimulant treatment on suicidality while controlling for confounding variables.

Methods

A retrospective cohort study was conducted using the US collaborative Network dataset of the TriNetX database, aggregating EHR data from 43 health care organizations. Two cohorts were created. Cohort one (232,662 subjects) consisted of Subjects with new encounter diagnosis of ADHD (ICD code F90) between the ages of 10 and 21 who had a first instance of stimulant medication prescription (VA CN800, CN801, CN802, CN809) within 3 months of ADHD encounter diagnosis. Cohort 2 (544,276 subjects) consisted of those with a first encounter diagnosis of ADHD who never had a stimulant prescription, or the prescription was written more than 3 years after diagnosis. Propensity score matching was used to created two balanced cohorts based on characteristics of age at first encounter for diagnosis of ADHD, White race, Black race, Hispanic or Latino ethnicity, Male sex, Prior suicide attempt (ICD T14.91), prior depressive episodes (ICD F32), prior major depression disorder (MDD) recurrent diagnosis (ICD F33), prior bipolar diagnosis (ICD F31), or prior antidepressant prescription (VA CN 600). Balanced cohorts were compared using measure of association analysis with the primary outcome of suicide attempts between 3 months and 3 years post first encounter with ADHD diagnosis. Analysis was performed using the built in analytics tools of TriNetX.

Results

For cohort 1 and 2 respectively, before (after) balancing cohorts using propensity score matching, average age at first encounter diagnosis was 14.7 (14.6) and 14.4 (14.6), percent of subjects with white race was 71.2 (71.2) and 64.3 (71.3), with black race was 12.4 (12.4) and 16.3 (12.4), with Hispanic or Latino ethnicity was 8.4 (8.4) and 10.7 (8.4), with female sex was 40.7 (40.6) and 38.1 (40.5), with a previous suicide attempt was 0.3 (0.3) and 0.3 (0.2), with a previous depressive encounter diagnosis was 9.0 (8.9) and 7.3 (8.8), with previous encounter diagnosis of MDD recurrent was 3.4 (3.3) and 2.3 (3.3), with a previous encounter diagnosis of bipolar disorder was 0.8 (0.8) and 1.5 (0.8), and who were prescribed antidepressants previously was 12.6 (12.6) and 8.5 (12.6). The number of patients with suicide attempts between 3 months and 3 years after first encounter diagnosis with ADHD was 732 and 743 for cohort 1 and 2 respectively. This was an absolute risk of 0.003 in both cases with a risk difference of 0.000, and a z score of -0.287. P value for this z score was 0.774. There does not appear to be a protective effect against suicide attempts of prescribing stimulants within 3 months of initial encounter with ADHD diagnosis.