Relationship of ADHD symptoms and global illness severity in adults treated with lisdexamfetamine dimesylate.

UNLABELLED The relationship between attention-deficit/hyperactivity disorder (ADHD) symptoms and global clinical assessment of functionality is complex. This post-hoc analysis explores this relationship and suggests implications for patient assessment in clinical practice. Adults with ADHD on a stable lisdexamfetamine dimesylate (LDX) dose for ≥ 6 months were enrolled in a double-blind, placebo-controlled, randomized withdrawal study. Participants entered a 3-week open-label phase continuing their prior LDX dose and were then randomized to placebo or the same LDX dose for a 6-week, double-blind, randomized withdrawal phase. ADHD symptom distribution was measured by the ADHD Rating Scale IV (ADHD-RS-IV) with Adult Prompts total score reflecting DSM-IV-TR ADHD symptom criteria and severity by Clinical Global Impressions-Severity (CGI-S) ratings at study entry and at end of study. Of 123 participants enrolled in the open-label phase, 116 were included in the randomized withdrawal phase (placebo, n = 60; LDX, n = 56). As reported in a prior publication, mean (standard deviation) ADHD-RS-IV total score change from baseline (week 3) to end of study (randomized-withdrawal phase) was 16.8 (11.80) for placebo and 1.6 (8.63) for LDX. At end of study, for placebo and LDX, 5.0% and 32.1% of participants, respectively, had a CGI-S = 1, 11.7% and 35.7% had a CGI-S = 2, 11.7% and 17.9% had a CGI-S = 3, 33.3% and 7.1% had a CGI-S = 4, 35.0% and 7.1% had a CGI-S = 5, and 3.3% and 0% had a CGI-S = 6; no participants had a CGI-S = 7 (P < 0.0001). The CGI-S ratings increased (worsened) as ADHD symptom scores worsened. Post-hoc regression analysis between ADHD-RS-IV scores and CGI-S demonstrated shared variance of 47% at week 3 and 69% for both placebo and LDX at end of study. Although ADHD symptom scores demonstrate a linear relationship with global illness severity, the variance suggests that other factors not captured by symptom scales are also important in assessing patient outcomes in clinical practice. ( TRIAL REGISTRATION ClinicalTrials.gov NCT00877487.).