Table 2.
The Impact of Pharmacotherapy of Childhood-Onset Attention-Deficit/Hyperactivity Disorder on the Development of Substance Use/Use Disorders
| Disorder | Study | Follow-up (years) | Subjects | Age and sex | Measure | Medication | Findings | Comments |
|---|---|---|---|---|---|---|---|---|
| ADHD treatment reduced subsequent substance use/use disorders | ||||||||
| ADHD | Paternite et al. (1999) (Prospective longitudinal follow-up), fall-back) | At ages 21–23 years |
N (total) = 219 N (ADHD) = 219 N (no ADHD) = N/A N (Tx stimulant ADHD) = 121 N (Tx behavioral ADHD) = 98 |
4–12-Year-olds 100% male |
ADHD dx: psychiatric chart Substance use: SADS-L SUD dx: SADS-L |
MPH | Higher doses of MPH Tx associated with fewer AUD diagnoses. Magnitude of effect not reported. | Higher MPH dose with lower suicide attempts Better response to medication was associated with lower MMPI D scores and better social functioning Longer MPH duration was associated with multiple academic, social, and psychiatric outcomes over time |
| ADHD | Biederman et al. (1999) (Prospective longitudinal follow-up) Biederman (2003) (Prospective longitudinal follow-up) Biederman et al. (2008) (Prospective longitudinal follow-up) |
4 Years 10 Years |
N = 212 N = 56 Tx ADHD N = 19 UnTx ADHD N = 137 non-ADHD controls Initial N (total) = 260 N (ADHD) = 140 N (no ADHD) = 120 N (Tx ADHD) = 92 N (UnTx ADHD) = 39 follow-up N (ADHD) = 112a N (Tx ADHD) = 82 N (UnTx ADHD) = 30 |
15–21-Year-olds 100% male 6–17 Years old (mean @ follow-up = 22 years old) 100% male |
K-SADS | Stimulants and non-stimulants for ADHD Tx | Tx ADHD at baseline were at a significantly reduced risk for a SUD at follow-up in adolescence versus UnTx ADHD (aOR: 0.15 [0.04–0.6]) Follow-up of smaller sample into adulthood did not show stimulant/nonstimulant Tx impact on SUDs |
Findings adjusted for baseline severity between groups The direction of the Tx effect similar for each SUD Only small group of adults persistently on stimulants in follow-up |
| ADHD | Katusic et al. (2005) (Retrospective cohort) |
17.2 Years |
N (total) = 1137 N (ADHD) = 379 N (controls) = 758 N (Tx ADHD) = 295 N (UnTx ADHD) = 84 |
5+ Years old 74.9% male |
ADHD dx, SUD dx: School and medical records (teacher/parent questionnaires, dx ADHD, Tx hx) |
Stimulants | Tx ADHD were at significantly decreased risk for developing SUD compared UnTx ADHD (20% vs. 27%) | Findings largely accounted for by the significant decrease in substance use in Tx versus UnTx boys with ADHD (22% vs. 36%) |
| ADHD | Wilens et al. (2008) (Prospective longitudinal follow-up) | 5 Years |
N (total) = 262 N (ADHD) = 140 N (no ADHD) = 122 follow-up N (ADHD) = 114 N (Tx ADHD) = 94 N (UnTx ADHD) = 20 |
6–18-Year-olds with ADHD 100% female |
SCID/K-SADS | Stimulants | Stimulant exposure decreased likelihood of manifesting a SUD compared to no stimulant exposure (N = 113; HR = 0.27 [0.125–0.60], χ2 = 10.57, p = 0.001) | Significant protective effect of stimulant exposure on the age-adjusted rate of development of drug use disorders |
| ADHD | Chang et al. (2014) (Retrospective cohort, registry) | 3 Years |
N (controls) = N (ADHD) = 38,753 N (Tx ADHD) = 19,410b N (UnTx ADHD) = 19,343b |
8–46-Year-olds 49.3% male |
ADHD dx: ICD diagnoses from the National Patient Register Substance use: ICD codes and convictions for substance related crime |
Stimulants and nonstimulants | Tx ADHD youth aged ≤15 years had a 31% significantly decreased rate of SUD compared to UnTx youth There was a 13% decrease in SUD risk for every year of stimulant Tx |
No indication that stimulant ADHD medication increased the rate of SUD in this cohort but rather, there was a negative association (HR = 0.38, 95% CI [0.23–0.64]) |
| ADHD | Hammerness et al. (2017) (Prospective open-label clinical trial) | 24 Months |
N (total) = 211 N (no ADHD) = 52 N (Tx ADHD) = 115 N (comparators) = 44c |
12–17-year-olds Tx 76% male, controls 73% male, comparators 68% male |
ADHD dx: clinical interview SUD dx Substance use: DUSI, urine toxicology screensd |
Extended-release MPH | Tx ADHD had significantly lower rates of alcohol [χ2 (3) = 12.56, p = 0.006] and drug use [χ2 (3) = 18.68, p = 0.001] when compared to UnTx and comparators No significant differences in rates of alcohol and drug use between Tx ADHD and non-ADHD controls |
Prospective clinical trial with main outcomes smoking, substance use, and SUD |
| ADHD | Quinn et al. (2017) (Prospective longitudinal observational) | Up to 9 years |
N (total) = 2,993, 887 N (no ADHD) = N/A N (ADHD) = 2,993,887 N (Tx ADHD) = 2,552,774 N (UnTx ADHD) = 441,113 |
13 Years old or older 52.8% male |
ADHD dx: ICD codes and/or ADHD Tx history in health care claims Substance-related events: substance use Tx history (emergency department claims, Tx visits) |
Stimulants and/or atomoxetine | Lower risk (male 35%, female 31%) of concurrent substance-related events during Tx episodes versus UnTx episodes Lower risk for SUD between Tx and UnTx ADHD; and ∼50% reduction in new onset of SUD in Tx versus UnTx groups Lower odds of substance-related events at 2-year follow-up in male patients (OR = 0.81, 95% CI [0.78–0.85]) and female patients (OR = 0.86, 95% CI [0.82–0.91]) |
In adjusted within-individual models, ADHD medication predicted a 19% reduction in the odds of substance-related events 2 years later among male patients (OR = 0.81, 95% CI [0.78–0.85]) and a 14% reduction among female patients (OR = 0.86, 95% CI [0.82–0.91]) |
| ADHD | Upadhyay et al. (2017) (Retrospective cohort from survey) | Not reported |
N (total) = 6483 N (no ADHD) = 5498 N (ADHD) = 985 N (Tx ADHD) = 333 N (UnTx ADHD) = 652 |
13–18-Year-olds 65.4% male |
ADHD dx: WHO CIDI Substance use: NCS-A SUD dx: NCS-A |
Not specified | Tx ADHD had significantly lower rates of substance use when compared to UnTx ADHD (OR = 0.53, 95% CI [0.31–0.90]) | ADHD Tx is negatively associated with substance use in adolescents with ADHD-C (OR = 0.53, 95% CI [0.24–0.97]) and those with ADHD-H (OR = 0.23, 95% CI [0.07–0.78]), but not ADHD-I subtype (OR = 0.49, 95% CI [0.17–1.39]) |
| ADHD and CD | Rasmussen et al. (2019) (Case control, retrospective survey) |
Retrospective |
N (total) = 472 N (no ADHD) = N/A N (ADHD) = 472 N (Tx ADHD) = 136 N (UnTx ADHD) = 336 |
17–56-Year-olds 72.7% male |
ADHD dx: SCL-90 R; The hyperkinetic checklist Substance use: SCL-90 R |
Stimulants | Tx ADHD had significantly lower rates of SUD compared to UnTx ADHD: Alcohol (13% vs. 34%) and drug (20% vs. 37%) | Previously CD-treated persons had a substantially lower frequency of problems (alcohol/substance abuse, criminality), and of certain psychiatric disorders (depressive, anxiety, and personality ones) |
| ADHD treatment increased subsequent substance use/use disorders | ||||||||
| ADHD | Lambert and Hartsough (1998) Lambert (2002) |
24–28 Years |
N = 492 N = 175 stimulant-Tx ADHD N = 39 with Tx ADHD and seizures N = 68 with ADHD without stimulant Tx N = 51 without ADHD with behavior problems N = 159 non-ADHD age-matched controls |
Kindergarten to 5th grade 83% male controls: 82% male |
CSBS and QDIS-III-R | Stimulant Tx for >6 weeks | ADHD > controls: tobacco, alcohol, marijuana, “stimulants,” and cocaine dependence Tx >1 year with stimulants versus Tx <6 months or no TX had higher rates: tobacco (42% vs. 26%) and cocaine dependency (21% vs. 10%). Follow-up: Tx ADHD with stimulants >1 year were twice as likely to become dependent on cocaine as UnTx |
CD, a major independent contributor to SUD, was not accounted for and was overrepresented in the group Tx with stimulants |
| ADHD treatment had no effect on subsequent substance use/use disorders | ||||||||
| Reading disorder and ADHD | Mannuzza et al. (2003) | 16 Years |
N = 231 N (reading disorder Tx) = 39 N (reading disorder UnTxe) = 63 N (controls) = 129 |
7–13-Year-olds 73.6% male |
Reading disorder dx: WISC K-SADS and DSM III-R |
MPH | No significant differences among Tx versus UnTx groups on rates of SUD | Normal controls had a significantly higher rate of cocaine use (60%; p = 0.01) compared to ADHD groups |
| ADHD | Barkley et al. (2003) (Prospective longitudinal follow-up) | 13 Years |
N (total) = 239 N (no ADHD) = 81 N (ADHD) = 158 N (Tx children ADHD) = 98 N (UnTx children ADHD) = 21 N (Tx high schoolers ADHD) = 32 N (UnTx high schoolers ADHD) = 115 |
4–12 Years | ADHD dx: CPRS-R, WWPARS, SCID Substance use: SCID |
Stimulants (MPH, d-amphetamine, pemoline, or combination of MPH and d-amphetamine or pemoline and d-amphetamine) |
No significant differences among Tx versus UnTx groups on rates of SUD, except for cocaine use. Higher risk for cocaine use disorders in Tx versus untreated groups (26% vs. 5%; p = 0.037) Duration of Tx not associated with any SUD risk |
Significant effect of Tx on subsequent SUD lost after correcting for ADHD severity and comorbid CD |
| ADHD | Faraone et al. (2007) (Case control, retrospective survey) | N/A |
N (total) = 206 N (full ADHD) = 127 N (late onset ADHD) = 38 N (ADHD UnTx) = 108 N (ADHD past Tx) = 33 N (ADHD past and current Tx) = 65 |
18–55-Year-olds 48% male in late-onset ADHD 53% male in full ADHD |
SCID/K-SADS DUSI |
Stimulant/nonstimulant Tx for ADHD | No significant differences among Tx versus UnTx groups on rates of SUD | Retrospective case control survey |
| ADHD | Winters et al. (2011) (Prospective longitudinal follow-up) | 15 Years |
N (total) = 242 N (ADHD) = 149 N (no ADHD) = 93 |
8–10-Year-olds 81% male |
ADHD Dx: DICA-R Substance use: ADI, CUQ |
Stimulants | No effect of stimulant medication on risk of SUDs and tobacco use | |
| ADHD | Molina et al. (2013) (Randomized controlled trial) | 8 Years |
N (total) = 697 N (ADHD) = 436 N (no ADHD) = 261 N (Tx ADHD) = not reported N (UnTx ADHD) = note reported |
7–9.9-Year-olds 81.3% male |
ADHD Dx: DISC-IV; teacher reported ratings of ADHD symptoms Substance use: child/adolescent-reported MTA questionnaire SUD Dx: DISC-IV |
Predominantly stimulants, but occasionally atomoxetine, guanfacine, clonidine, and amitriptyline | No significant differences among Tx versus UnTx groups on rates of SUD or substance use | Neither randomized Tx assignment nor proportion of days medicated in the past year was a predictor of SUD at the 6- and 8-year assessments |
| ADHD | Dalsgaard et al. (2014) (Retrospective cohort, registry) | 31 Years old |
N (total) = 2,500,208 N (ADHD) = 208 N (no ADHD) = 2,500,000 N (Tx ADHD) = 208 N (UnTx ADHD) = 2,499,792 |
4–15-Year-olds 87.9% male |
ADHD dx: DSM-IV and ICD-10 diagnostic checklists SUD dx and substance use: ICD diagnoses from national register |
MPH or dexamphetamine | No effect of stimulant Tx in childhood on adult SUD Early initiation of stimulant Tx for ADHD was associated with a reduced risk of later alcohol and other SUDs |
For every year older at initiation of stimulants, the risk of SUD in adulthood increased by a factor of 1.46 |
| ADHD | Levy et al. (2014) (Retrospective survey) | Not reported |
N (total) = N (ADHD) = 232 N (no ADHD) = 335 N (Tx ADHDf) = 139 N (UnTx ADHD) = 93 |
Birth cohort; at least 5 years of age ADHD 72% male; controls 62.7% male |
ADHD dx: school and medical records, MINI SUD dx: school and medical records, MINI |
Stimulants | No significant differences among Tx versus UnTx groups on rates of SUD Higher risk for SUD in those Tx after age 13 years |
Confounds not accounted for in analyses |
| ADHD | McCabe et al. (2016) (Retrospective cohort, multi-cohort national registry) | Not reported |
N (total) = 40,358 N (controls) = 35,447 N (ADHD) = 4911 N (Tx stimulant ADHD) = 3539 N (Tx nonstimulant ADHD) = 1332 N (UnTx ADHD) = not reported |
High school seniors 48% male |
Substance use: alcohol single question, cigarette questionnaire, marijuana and other drug use questionnaire. ADHD dx: not reported |
Stimulant and nonstimulant medication therapy for ADHD. | No significant differences among Tx versus UnTx groups on rates of SUD Individuals who initiated Tx earlier (<9 years old) and for a longer duration (6 years or more) had significantly lower risk for substance use compared to individuals who initiated Tx later (>9 years old) and for a shorter duration (<6 years) |
In general, the prevalence of substance use was highest among individuals who reported the latest onset and shortest duration of prescription stimulant medication therapy for ADHD (aOR for Controls 1.8 (95% CI [1.3–2.6]) |
| ADHD | McCabe et al. (2016) (Retrospective cohort, Multi-cohort National Registry) | 4 Years |
N (total) = 4755 N (ADHD) = 556 N (Tx medical ADHD) = 322 N (nonmedical use) = 124 |
Secondary school students 47.1% male |
SSLS | Stimulants | Individuals who initiated Tx earlier (≤9 years of age) had significantly lower odds of marijuana and other substance use compared to those who initiated Tx later (>9 years of age) | The odds of substance use and substance-related problems were significantly greater among those who initiated earlier nonmedical use of stimulant medications relative to later nonmedical initiation |
| ADHD | Mannuzza et al. (2008) (Prospective longitudinal follow-up) | Late adolescence (M = 18.4) and adulthood (M = 15.3) |
N (total) = 354 N (no ADHD) = 178 N (ADHD) = 176 N (Tx ADHD) = 176 N (UnTx ADHD) = N/A |
6–12-Year-olds 100% male |
ADHD, substance use and SUD: DISC, CHAMPS | MPH | Earlier onset ADHD Tx (initiation ages <7 years) had significantly lower rates of SUD compared to later onset ADHD Tx (initiation ages 8–12) (27% vs. 44% Wald χ2 = 5.38, p < 0.02) | Even when controlling for SUD, age at stimulant Tx initiation was significantly and positively related to the subsequent antisocial personality disorder Age at first MPH Tx was unrelated to mood and anxiety disorders |
| ADHD and CD | Harty et al. (2011) (Prospective longitudinal follow-up) | 9.3 Years |
N (total) = 182 N (no ADHD) = 85 N (ADHD) = 66 N (ADHD+CD) = 31 N (Tx ADHD) = 69 N (UnTx ADHD) = 28 |
7–11-Year-olds 87.8% male |
ADHD dx: DISC; CBCL; IOWA; CPRS CD dx: CBCL; IOWA; CPRS Substance use: K-SADS; RAPI SUD dx: K-SADS. |
Stimulants | No significant differences among Tx versus UnTx groups on rates of SUD | Within an ethnically diverse urban sample, the increased rate of substance use associated with ADHD was fully accounted for by comorbid CD |
a
Statistical analyses conducted on this group only.
b
Not reported specifically for 15 and under, however, 48.7% of male patients and 53% of female patients in total sample were exposed to ADHD (Tx).
c
Eighteen currently on medication, 26 not currently on medication (historical Tx history not specified).
d
Tx participants only.
e
Placebo.
f
Tx defined as 6 months of longer.
ADHD, attention-deficit/hyperactivity disorder; ADI, Adolescent Diagnostic Interview; aOR, adjusted odds ratio; CBCL, Child Behavior Checklist; CD, conduct disorder; CDRS-R, Children's Depression Rating Scale-Revised; CGAS, Children's Global Assessment Scale; CHAMPS, Schedule for the Assessment of Conduct, Hyperactivity, Anxiety, Mood, and Psychoactive Substances; CI, confidence interval; CPRS-R, Conners Parent Rating Scale-Revised; CSBS, California Smoking Baseline Survey; CUQ, Cigarette Use Questionnaire; DICA-R, Diagnostic Interview for Children and Adolescents-Revised; DISC-IV, Diagnostic Interview Schedule-IV; DSM III-R, Lifetime Semi Structured Psychiatric Interview; DUSI, Drug Use Screening Inventory; dx, diagnosis; hx, history; ICD, International Classification of Disease; IOWA, Inattention/Overactivity with Aggression; K-SADS, Schedule for the Assessment of Conduct, Hyperactivity, Anxiety, Mood; MDD, major depressive disorder; MINI, Mini International Neuropsychiatric Interview; MPH, methylphenidate; MTA, Multimodal Treatment Study of Children with ADHD; NCS-A, National Comorbidity Survey-Adolescent; NMUPPR, Non-Medical Use of Prescription Pain Relievers; OR, odds ratio; QDIS-III-R, Quick Diagnostic Interview Schedule 3rd edition; RADS, Reynolds Adolescent Depression Scale; RAPI, Rutgers's Alcohol and Drug Use Problem Index Questionnaire.; SADS-L, Schedule for Affective Disorders and Schizophrenia-Lifetime; SCID, Structured Clinical Interview for DSM Disorders; SIQ-Jr, Suicidal Ideation Questionnaire-Jr. High Version; SSLS, Secondary Student Life Survey; SUD, substance use disorder; TAU, treatment as usual; Tx, treatment; UnTx, untreated; WHO CIDI, World Health Organization Composite International Diagnostic Interview 3.0; WISC, Wechsler Intelligence Scale for Children; WWPARS, Werry-Weiss-Peters Activity Rating Scale.