Table 1.
Randomized controlled trials that address discontinuation of psychostimulants in children with ADHD.
| Reference | Industry sponsored | Duration of medication discontinuation | Total N | Age (years) | Number of boys (% of total) | Medication(s) | Primary outcome measure(s) | Findings |
|---|---|---|---|---|---|---|---|---|
| Abikoff et al. (14) | No | 4 weeks | 50 | 6–12 | 45 (90%) | MPH, dextroamphetamine, pemoline | CTRS BRS Home Hyperactivity Scale Parent Attitude Test Standardized Tests of achievement and cognition |
The combination of a stimulant with cognitive training did not facilitate stimulant withdrawal |
| Abikoff et al. (15) | Yes | 1 year | 103 | 7–9.9 | (93%) | MPH | CPRS Home Situations Questionnaire CTRS School Situations Questionnaire | All children relapsed when switched to placebo (single-blind), mean 8.6 vs. 17.1 days for MPH alone vs. MPH with multimodal psychosocial treatments, none of the parent, teacher, or psychiatrist evaluations yielded significant group or interaction effects. |
| Arnold et al. (16) | Yes | 2 weeks | 75 | 6–16 | 61 (81%) | Dexmethylphenidate | CGI-I SNAP-ADHD |
Placebo group 3.6 × more likely to “fail” treatment (CGI-I 6 or 7); statistically significant deterioration on Teacher and Parent SNAP scales compared to continued medication. |
| Banaschewski et al. (17)* | Yes | 6 weeks | 153 | 6–17 | 119 (78%) | Lisdexamfetamine | CHIP-CE: PRF WFIRS-P |
The improvement in health-related quality of life and functional status during the lead-in phase was maintained in the lisdexamfetamine arm while those in the placebo arm had statistically significant deterioration for both. |
| Brown et al. (18) | No | 24 h to 1 week | 35 | 5–14 | 28 (80%) | MPH | ACRS CPRS Various tests of attentional deployment and cognitive style, academic achievement |
Neither the combination of a stimulant with cognitive therapy nor a stimulant with attention control therapy for 3 months facilitated stimulant withdrawal |
| Coghill et al. (17)* | Yes | 6 weeks | 153 | 6–17 | 119 (78%) | Lisdexamfetamine | ADHD-RS | Rates of treatment failure were 15.8% in the lisdexamfetamine group and 67.5% in the placebo group. Median time to treatment failure was 17 days for the placebo group. |
| Gillberg et al. (19) | No | 12 months double-blind treatment followed by 3 months of single-blind placebo in both groups. Placebo was tapered over 2 weeks. | 62 | 6–11 | 51 (82%) | Amphetamine | CTRS | The improvement in Conners Teacher and Parent Rating Scale scores during the lead-in phase was maintained in the amphetamine arm while those in the placebo arm had significant deterioration for both; 71% of those in the placebo group withdrew or went into open treatment compared to 29% of those in the amphetamine group. |
| Klein et al. (20) | No | Stimulant holiday for 3 months over the summer for two consecutive summers | 58 | 6–12 | 53 (91%) | MPH | Height Weight |
At the end of the first summer, the group that had been discontinued from stimulants weighed on average 0.9 kg more than the treatment group; there was no statistically significant difference in height. At the end of the second summer, the group that had been discontinued from stimulants was on average 1.5 cm taller than the treatment group; there was no longer a statistically significant difference in weight. |
| Martins et al. (21) | No (medication supplied by industry) | Stimulant holiday for four weekends | 40 | Mean age 9.0 and 9.6 for MPH and placebo group, respectively | 40 (100%) | MPH | ACRS SERS |
There was no difference between groups, suggesting a lack of rebound ADHD symptoms during a short-term MPH weekend discontinuation. The weekend placebo group had significant reduction in insomnia reported, and there was a trend toward reduction in decreased appetite. |
| Matthijssen et al. (22) | No | 3-week taper, 4-week discontinuation | 94 | 8–18 | 73 (78%) | MPH | ADHD-RS CGI-I |
Mean scores favored the group that continued MPH treatment on the ADHD-RS. 40% of those who discontinued medication worsened on the CGI-I compared to 16% of those who continued medication. |
| Nolan et al. (23) | No | 2-week crossover | 19 | 6–17 | 18 (95%) | MPH, dextroamphetamine | Various measures of tics and ADHD symptoms | No significant withdrawal effect on tics with placebo. Significant increase in some parent-reported behavioral symptoms, hyperactivity, and aggression while on placebo. |
| Waxmonsky et al. (24) | No | Weekend drug holidays | 71 | 5–12 | MPH | Weight, height, CGI-S, growth trajectories | Medication use was associated with reductions in height and weight, caloric supplement and drug holidays increase weight velocity more than monitoring | |
| Zeiner et al. (25) | No | 3 weeks | 21 | 7–12 | 21 (100%) | MPH |
PACS CTRS Neuropsychological testing |
76% of boys had a significant worsening in behavioral problems either at home or at school while on placebo |
*
One study with two resultant papers.
ACRS, Abbreviated Conners Rating Scale; ADHD-RS, ADHD Rating Scale-IV; BRS, Hillside Behavior Rating Scale; CGI, Clinical Global Impression (Severity/Improvement); CHIP-CE: PRF, Child Health and Illness Profile—Child Edition: Parent Report Form; CTRS, Conners Teacher Rating Scale; MPH, methylphenidate; PACS, Parental Account of Childhood Symptoms; WFIRS-P, Weiss Functional Impairment Rating Scale-Parent Report.