Clinical Summary
Clinical Summary: Characteristics and Predictors of Fluctuating Attention-Deficit/Hyperactivity Disorder in the Multimodal Treatment of ADHD (MTA) Study
ADHD is often treated as either persistent or remitted, but many patients do not follow a stable course. In this long-term MTA follow-up, fluctuation was common, meaning a single visit can misclassify trajectory and clinicians need a more longitudinal view of symptoms, impairment, and treatment needs.
Design
The MTA continued for 14 additional years with prospective follow-ups approximately biennially (8 assessments) until 16 years after baseline.
N
The current subsample (N = 483; 83.4% of original sample)
Population
participants with ADHD who had at least 1 follow-up assessment in adulthood (age 18 or older)
Duration
until 16 years after baseline
Key Findings
- Fluctuating ADHD occurred in 63.8% of the sample; within this group (N = 335), participants had a mean = 3.58, SD = 1.36 ADHD classification changes over time and average first remission in early adolescence (mean = 12.52, SD = 3.63).
- Endpoint status poorly reflected longitudinal course: the endpoint symptom persistent subgroup consisted of 80.1% fluctuating, 15.5% stable persistence, 4.4% sustained partial remission, and 0.0% recovery, while the endpoint symptom remission subgroup consisted of 62.3% fluctuating, 22.0% recovery, 15.7% sustained partial remission, and 0.0% stable persistence.
- Among longitudinal fluctuators, similar proportions met criteria for ADHD symptom persistence (56.6%) and remission (43.4%) at MTA endpoint.
- Within the fluctuating group, each added point in average environmental demands score across time was associated with a 1.58 higher odds of experiencing a full remission period than a persistent period and a 1.36 higher odds of experiencing a partial remission period than a persistent period at any given time point.
- For fluctuators, each point higher in environmental demands at a given time point, compared to the individual's own average level, was associated with being 1.28 times more likely to be experiencing an episode of full remission vs an episode of persistence at that time point; each participant on average contributed 5.04/6 possible data points (83.5% complete data).
Clinical Bottom Line
Fluctuation is the most common long-term ADHD course, so one-time endpoint classification is not a reliable proxy for prognosis. Ongoing monitoring is essential, and environmental context appears clinically relevant, especially when patients are temporarily doing well under higher demands.
Practice Implications
- Do not equate current remission or persistence at a single follow-up with long-term course; in this cohort, 80.1% of endpoint persisters and 62.3% of endpoint remitters were actually fluctuating over time.
- Monitor symptoms, impairment, and treatment use longitudinally, because fluctuating cases showed mean = 3.58, SD = 1.36 classification changes over 16 years and first remission typically occurred at mean = 12.52, SD = 3.63.
- Include family psychiatric burden and internalizing history in prognostic assessment, since parent SCID diagnoses, childhood mood disorder, childhood depression severity, and 36-month MTA treatment response predicted longitudinal subgroups, whereas baseline ADHD severity did not.
- Discuss environmental fit with patients and families; among fluctuators, higher environmental demands were linked to better concurrent status, with 1.58 higher odds of full remission and 1.36 higher odds of partial remission versus persistence for each added point in average demands score across time.
