Treatment Outcomes With Licensed and Unlicensed Stimulant Doses for Adults With Attention-Deficit/Hyperactivity Disorder: A Systematic Review and Meta-Analysis

Luis C Farhat; José M Flores; Victor J Avila-Quintero; Guilherme V Polanczyk; Andrea Cipriani; Toshi A Furukawa; Michael H Bloch; Samuele Cortese

doi:10.1001/jamapsychiatry.2023.3985

. 2023 Oct 25;81(2):157–166. doi: 10.1001/jamapsychiatry.2023.3985

Treatment Outcomes With Licensed and Unlicensed Stimulant Doses for Adults With Attention-Deficit/Hyperactivity Disorder

A Systematic Review and Meta-Analysis

Luis C Farhat ^1,^✉, José M Flores ^2,³, Victor J Avila-Quintero ³, Guilherme V Polanczyk ¹, Andrea Cipriani ^4,^5,⁶, Toshi A Furukawa ⁷, Michael H Bloch ^3,⁸, Samuele Cortese ^9,^10,^11,^12,^13,^✉

¹Department of Psychiatry, Faculdade de Medicina FMUSP, Universidade de São Paulo, São Paulo, Brazil

²Division of Child & Adolescent Psychiatry, Department of Psychiatry and Biobehavioral Sciences, University of California, Los Angeles (UCLA), Los Angeles, California

³Child Study Center, Yale University School of Medicine, New Haven, Connecticut

⁴Department of Psychiatry, University of Oxford, Oxford, United Kingdom

⁵Oxford Health NHS Foundation Trust, Warneford Hospital, Oxford, United Kingdom

⁶Oxford Precision Psychiatry Lab, NIHR Oxford Health Biomedical Research Centre, Oxford, United Kingdom

⁷Department of Health Promotion and Human Behavior, Kyoto University Graduate School of Medicine/ School of Public Health, Kyoto, Japan

⁸Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut

⁹Center for Innovation in Mental Health, School of Psychology, Faculty of Environmental and Life Sciences, University of Southampton, Southampton, United Kingdom

¹⁰Clinical and Experimental Sciences (CNS and Psychiatry), Faculty of Medicine, University of Southampton, Southampton, United Kingdom

¹¹Solent NHS Trust, Southampton, United Kingdom

¹²Department of Child and Adolescent Psychiatry, NYU Grossman School of Medicine, New York, New York

¹³Division of Psychiatry and Applied Psychology, School of Medicine, University of Nottingham, Nottingham, United Kingdom

Accepted for Publication: August 18, 2023.

Published Online: October 25, 2023. doi:10.1001/jamapsychiatry.2023.3985

^✉

Corresponding Author: Luis C. Farhat, MD, Instituto de Psiquiatria do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (IPq-HCFMUSP), Rua Dr. Ovídio Pires de Campos, 785, CEAPESQ, 3° andar (ala norte), sala 9, São Paulo, SP 05403-903, Brazil (luis.farhat@usp.br); Samuele Cortese, MD, PhD, School of Psychology, Centre for Innovation in Mental Health (CIMH), Faculty of Environmental and Life Sciences, University of Southampton, Highfield Campus, Building 44, Southampton SO17 1BJ, United Kingdom (samuele.cortese@soton.ac.uk).

Author Contributions: Dr Farhat had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Farhat and Flores contributed equally. Drs Bloch and Cortese were cosenior authors.

Concept and design: Farhat, Cipriani, Furukawa, Bloch, Cortese.

Acquisition, analysis, or interpretation of data: Farhat, Flores, Avila-Quintero, Polanczyk, Cipriani, Bloch, Cortese.

Drafting of the manuscript: Farhat, Flores, Furukawa, Bloch.

Critical review of the manuscript for important intellectual content: Flores, Avila-Quintero, Polanczyk, Cipriani, Bloch, Cortese.

Statistical analysis: Farhat, Flores, Cipriani, Bloch.

Obtained funding: Bloch.

Administrative, technical, or material support: Bloch, Cortese.

Supervision: Polanczyk, Cipriani, Bloch, Cortese.

Conflict of Interest Disclosures: Dr Polanczyk reported personal fees from Ache, Apsten, Abbott, Medice, and Takeda outside the submitted work. Dr Cipriani reported grants from Angelini Pharma and personal fees from Lundbeck Pharma and Angelini Pharma outside the submitted work. Dr Furukawa reported personal fees from Boehringer-Ingelheim, DT Axis, Kyoto University Original, Sony, and grants from Shionogi and personal fees from Shionogi outside the submitted work; in addition, Dr Furukawa had a patents pending (2018-177688 and 2022-082495) and intellectual properties licensed to Mitsubishi-Tanabe. Dr Bloch reported grant or research support from Therapix Biosciences, Emalex, Biosciences, Janssen Pharmaceuticals, Biohaven Pharmaceuticals, the National Institutes of Health, Lesbian Health Fund, the Yale Foundation for Lesbian and Gay Studies, and Patterson Foundation; he has served on the advisory board/data monitoring and safety board of Therapix Biosciences. Dr Bloch has also received royalties from Wolters Kluwer for Lewis’s Child and Adolescent Psychiatry: A Comprehensive Textbook, Fifth Edition and moonlighting pay from the Veteran’s Administration. Dr Cortese reported personal fees from Association for Child and Adolescent Mental Health, Canadian ADHD Alliance Resource, British Association of Pharmacology, and Healthcare Convention outside the submitted work. No other disclosures were reported.

Funding/Support: Dr Cipriani is supported by the National Institute for Health Research (NIHR) Oxford Cognitive Health Clinical Research Facility, by an NIHR Research Professorship (RP-2017-08-ST2-006), by the NIHR Oxford and Thames Valley Applied Research Collaboration, by the NIHR Oxford Health Biomedical Research Centre (NIHR203316), and by Wellcome (GALENOS project). Dr Cortese is currently supported by funding from the National Institute for Health and Care Research (NIHR203684, NIHR203035, NIHR130077, NIHR128472, RP-PG-0618-20003), Solent National Health Service Trust (Research Capability Funding, 2022), University of Southampton (Knowledge and Exchange Enterprise Fund 2023, Enterprise Development Fund 2022), and European Research Agency (Horizon: Project 101095568).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed are those of the authors and not necessarily those of the UK National Health Service, the National Institute for Health Research, or the UK Department of Health.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC10600727 PMID: 37878348

This study investigates the impact, based on averages, of stimulant doses on treatment outcomes in adults with attention-deficit/hyperactivity disorder.

Key Points

Question

Are unlicensed doses of stimulants associated with positive treatment outcomes in adults with attention-deficit/hyperactivity disorder (ADHD)?

Findings

In this systematic review and meta-analysis of group averages, unlicensed doses of stimulants were associated with small, possibly nonclinically meaningful, additional reductions of ADHD symptoms and increased risk of treatment discontinuation due to adverse events in comparison with licensed doses of these medications.

Meaning

Based on averages from available placebo-controlled randomized clinical trials, unlicensed doses of stimulants may not have positive risk benefits for adults with ADHD and practitioners should generally consider unlicensed doses cautiously.

Abstract

Importance

Stimulants (methylphenidate and amphetamines) are often prescribed at unlicensed doses for adults with attention-deficit/hyperactivity disorder (ADHD). Whether dose escalation beyond US Food and Drug Administration recommendations is associated with positive risk benefits is unclear.

Objective

To investigate the impact, based on averages, of stimulant doses on treatment outcomes in adults with ADHD and to determine, based on averages, whether unlicensed doses are associated with positive risk benefits compared with licensed doses.

Data Sources

Twelve databases, including published (PubMed, Cochrane Library, Embase, Web of Sciences) and unpublished (ClinicalTrials.gov) literature, up to February 22, 2023, without language restrictions.

Study Selection

Two researchers independently screened records to identify double-blinded randomized clinical trials of stimulants against placebo in adults (18 years and older) with ADHD.

Data Extraction and Synthesis

Aggregate data were extracted and synthesized in random-effects dose-response meta-analyses and network meta-analyses.

Main Outcome Measures

Change in ADHD symptoms and discontinuations due to adverse events.

Results

A total of 47 randomized clinical trials (7714 participants; mean age, 35 (SD, 11) years; 4204 male [56%]) were included. For methylphenidate, dose-response curves indicated additional reductions of symptoms with increments in doses, but the gains were progressively smaller and accompanied by continued additional risk of adverse events dropouts. Network meta-analyses showed that unlicensed doses were associated with greater reductions of symptoms compared with licensed doses (standardized mean difference [SMD], −0.23; 95% CI, −0.44 to −0.02; very low certainty of evidence), but the additional gain was small and accompanied by increased risk of adverse event dropouts (odds ratio, 2.02; 95% CI, 1.19-3.43; moderate certainty of evidence). For amphetamines, the dose-response curve approached a plateau and increments in doses did not indicate additional reductions of symptoms, but there were continued increments in the risk of adverse event dropouts. Network meta-analysis did not identify differences between unlicensed and licensed doses for reductions of symptoms (SMD, −0.08; 95% CI, −0.24 to 0.08; very low certainty of evidence).

Conclusions and Relevance

Based on group averages, unlicensed doses of stimulants may not have positive risk benefits compared with licensed doses for adults with ADHD. In general, practitioners should consider unlicensed doses cautiously. Practitioners may trial unlicensed doses if needed and tolerated but should be aware that there may not be large gains in the response to the medication with those further increments in dose. However, the findings are averages and will not generalize to every patient.

Introduction

Stimulants (methylphenidate and amphetamines) are recommended in clinical practice guidelines (CPGs)^1,2,3 for the treatment of adults with attention-deficit/hyperactivity disorder (ADHD), a common psychiatric condition^4,5 that has been associated with criminality,⁶ increased mortality,⁷ and substantial financial costs.⁸ Because there is considerable variability in how patients respond to stimulants, practitioners are advised to adopt escalating-dose stepwise titration to identify optimal doses that ensure maximal reductions of symptoms, improvement in functioning, and acceptable tolerability.^9,10,11,12

Based on currently available evidence, practitioners are unable to predict the optimal dose for a given patient a priori without trialing doses, as needed and tolerated, sequentially. There is consensus that stimulants should be started at low doses, but there is little agreement on the maximum doses at which escalation should stop, with CPGs generally recommending levels that are higher than those licensed by regulatory agencies such as the US Food and Drug Administration (FDA). For instance, while the FDA recommends a maximum of 60 mg per day for most formulations of methylphenidate, CPGs^1,2,3 generally recommend doses up to 100 mg per day.

Whether stimulant dose escalation beyond FDA recommendations is associated with positive risk benefits is unclear. Although an analysis at the level of the individual is required for the identification of personalized trade-offs, an examination based on averages may be useful to inform expectations about such risk benefits in the absence of evidence-based personalized information. Indeed, recommendations from CPGs^1,2,3 are based on average associations¹³ to inform practitioners about expected outcomes, which may or not apply to individuals in routine clinical practice. Current recommendations from CPGs about stimulant dosing for adults are based on the observation that flexible-dose randomized clinical trials (RCTs) allowing titration to higher doses^14,15 showed the largest effect sizes across RCTs,¹⁶ despite the fact that more recent large fixed-dose RCTs have not confirmed a dose-response relationship between stimulants and reductions of symptoms.^17,18,19 In this context, rigorous systematic reviews and meta-analyses are needed to inform guideline development, evidence-based communications with patients, and decision making.²⁰

This study aimed at filling this gap. We identified RCTs of stimulants against placebo in adults with ADHD and conducted dose-response meta-analyses with fixed-dose RCTs to evaluate the associations, based on averages, between doses and treatment outcomes. Fixed-dose RCTs enable strict examination of dose dependency.²¹ However, they may underestimate reductions of symptoms and inflate dropout rates at the higher doses, as participants who are unable to tolerate a given dosage are removed rather than given a lower dose.²² Flexible-dose RCTs are more closely related to real-world practices and may be particularly appropriate to investigate the benefits of escalation to higher doses.²³ Treatment decisions, such as the prescription of stimulants at licensed or unlicensed doses, are dichotomous, which is not captured by dose-response meta-analyses that assess doses continuously. Therefore, we also conducted network meta-analyses (NMAs) with fixed-dose and flexible-dose RCTs to estimate the comparative outcomes of stimulants at unlicensed doses in association with licensed doses. NMAs allowed us to extend beyond the direct evidence (from fixed-dose RCTs with stimulant arms at licensed and unlicensed doses) and incorporate indirect evidence (from RCTs of any dosing design with stimulant arms at any dose against placebo) in network estimates, ensuring increased statistical power and precision of the resulting estimates.²⁴

Methods

We followed the Cochrane Handbook²⁵ and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines²⁶ (eMethods 1 in Supplement 1). We also preregistered the protocol (PROSPERO CRD42020161804) (eMethods 2 in Supplement 1).

Data Sources

We drew on a database of RCTs of ADHD medications,¹³ collated through searches in 12 electronic databases, including published (eg, PubMed, the Cochrane Central Register of Controlled Trials, Embase, Web of Science Core Collection) and unpublished (eg, ClinicalTrials.gov) literature, from inception to February 22, 2023, without language restrictions (eMethods 3 in Supplement 1).

Study Selection

Two researchers independently screened titles/abstracts and then conducted a full-text assessment of selected records to confirm eligibility. Any discrepancies were discussed with a third senior researcher. We included double-blinded placebo-controlled RCTs that examined methylphenidate or amphetamines as monotherapy for at least 1 week in adults (18 years or older) with ADHD according to the Diagnostic and Statistical Manual of Mental Disorders (eMethods 4 in Supplement 1).

Data Extraction

Aggregate data were extracted independently by 2 investigators (eMethods 5 in Supplement 1). For fixed-dose RCTs, we extracted the group’s assigned dose in milligrams per day. For flexible-dose RCTs, we extracted the maximum dose in the titration schedule in milligrams per day. When doses were determined by weight (eg, 1 mg/kg per day), we calculated the doses in milligrams per day considering the sample’s average weight. We also coded whether doses were within (licensed) or beyond (unlicensed) the FDA recommended range for each medication (eTable 1 in Supplement 1). Lastly, we converted doses of methylphenidate and amphetamine formulations in equivalent doses considering immediate-release methylphenidate hydrochloride and mixed amphetamine salts as the reference medications, respectively (eTable 1 and eMethods 6 in Supplement 1).

Outcomes

The primary outcome was change in ADHD symptoms from baseline to end point. When data from multiple informants were available, we preferred observer/clinician reports. When data from multiple rating scales were available, we preferred the ADHD rating scale (ADHD-RS),²⁷ the Conners Adult ADHD Rating Scale,²⁸ or any other rating scale, in that order. We considered results from intention-to-treat analyses, using the method adopted by each study to handle missing data. However, we also considered data from participants who completed the study (modified intention-to-treat) if those were the only data reported. Secondary outcomes were discontinuation rates due to adverse events (tolerability) and any reason (acceptability).

Data Synthesis

We calculated standardized mean difference (SMD) (change in ADHD symptoms) and odds ratio (OR) (tolerability, acceptability) with 95% CIs. We conducted analyses for methylphenidate and amphetamines separately.

We conducted 1-stage random-effects dose-response meta-analyses²⁹ using restricted cubic splines with knots at the 10th, 50th, and 90th percentiles³⁰ or log-linear/linear regression to model the dose-response associations. The log-linear/linear function was only chosen if the second coefficient of the spline was not significantly different from 0.³¹ We converted SMD and OR to change in ADHD-RS values and number needed to harm, respectively, because those are clinically interpretable measures (eMethods 7 in Supplement 1). For change in ADHD symptoms, we also estimated the 50% and 95% effective doses (ED50, ED95), which represent the mean dose that is associated with 50% and 95% of the maximum change in ADHD symptoms compared with placebo, respectively.³²

We conducted frequentist random-effects NMAs³³ to compare placebo, licensed, and unlicensed doses of stimulants. We evaluated the plausibility of the transitivity assumption by comparing the distribution of clinical and methodological characteristics across comparisons (eMethods 8 in Supplement 1). We assumed common heterogeneity variance τ² across comparisons.

We quantified heterogeneity in dose-response meta-analyses with the variance partition coefficient (VPC)²⁹ and in NMAs by comparing τ² with its empirical distribution.^34,35 In NMAs, we assessed incoherence with global and local tests.^36,37

We assessed risk of bias of RCTs for each outcome with the Cochrane risk of bias tool 2 (eMethods 9 in Supplement 1).³⁸ We plotted contour-enhanced funnel plots³⁹ for the placebo-controlled comparisons to evaluate publication bias if more than 10 studies were available. We assessed the certainty of findings from the NMAs with the Confidence in Network Meta-Analysis (CINeMA) framework^40,41 (eMethods 9 in Supplement 1). The robustness of our findings was examined in numerous sensitivity analyses for both dose-response meta-analyses and NMAs (eMethods 10 in the Supplement). All analyses were conducted in R version 4.2.2 (The R Project) with packages dosresmeta³¹ (version 2.0.1), meta⁴² (version 6.2-1), and netmeta⁴³ (version 2.8-0).

Results

Study Selection and Characteristics

From the initial 14 331 records retrieved through literature searches and screened, 47 RCTs (29 methylphenidate; 18 amphetamines) were included (Figure 1) (eResults 1 in Supplement 1). These trials comprised 7714 participants (5125 methylphenidate; 2589 amphetamines) with a mean age of 35 (SD, 11) years, of whom 4204 (56% of those with available data) were males and 5153 (87% of those with available data) self-identified as White (eTable 2 in Supplement 1). Most RCTs recruited participants diagnosed with DSM-IV criteria (40 [85%]) and were exclusively conducted in the US (32 [68%]) (eTable 2 in Supplement 1). Additionally, most RCTs adopted a flexible-dose design (27 [57%]) (eTable 3 in Supplement 1), had parallel arms (29 [62%]), and were conducted in the short term (median duration, 4.5 weeks; IQR, 3-6) (eTable 2 in Supplement 1). A total of 8 RCTs were rated at high risk of bias (18.2%), 14 at low risk of bias (31.8%), and 22 at some concerns of risk of bias for the primary outcome (50%) (eResults 2 in Supplement 1).

Figure 1. — The list of excluded records, with exclusion reasons, was provided in Open Science Framework.⁴⁴ ADHD indicates attention-deficit/hyperactivity disorder; RCT, randomized clinical trial.

Methylphenidate

Change in ADHD Symptoms

The dose-response curve based on 8 fixed-dose RCTs and 18 stimulant arms (dose range 15 to 82.5 mg per day) indicated that increments in doses were associated with additional reductions of symptoms throughout the entire dose range for which data were available. However, improvements were incrementally smaller beyond 35 to 40 mg per day (Figure 2A; Table, which also reports change in ADHD-RS values). Consistently, the ED50 was estimated at 25 mg per day while the ED95 was estimated at 72.5 mg per day. NMA corroborated that unlicensed doses were associated with larger reductions of symptoms in comparison with licensed doses, but the incremental benefit was small (26 studies; SMD, −0.23; 95% CI, −0.44 to −0.02; P = .03; very low certainty of evidence) (Figure 3A; eFigure 1, eFigure 2, and eTable 4 in Supplement 1).

Figure 2. — Dose-response curve for change in attention-deficit/hyperactivity disorder symptom severity (A) and tolerability (B). The curves are presented until the maximum dose for which data were available for equivalent doses of methylphenidate. The shaded areas indicate 95% CIs. The dotted line indicates the US Food and Drug Administration maximum recommended dose for immediate-release methylphenidate hydrochloride.

Table. Results Per Dose After Conversion of Standardized Mean Difference (SMD) and Odds Ratio (OR) to Clinically Interpretable Values.

Dose, mg/d^a	Change in ADHD symptom severity (95% CI)				Tolerability (95% CI)
	Methylphenidate		Amphetamines		Methylphenidate			Amphetamines
	SMD	ADHD-RS values, mean	SMD	ADHD-RS values, mean	OR	RR	NNH	OR	RR	NNH
0	Reference	−9.7	Reference	−9.7	Reference	Reference	NA	Reference	Reference	NA
5	−0.06 (−0.08 to −0.04)	−9.76 (−9.99 to −9.53)	−0.14 (−0.18 to −0.10)	−10.68 (−11.14 to −10.22)	1.12 (1.06-1.18)	1.12 (1.06-1.18)	412 (275-823)	1.07 (1.02-1.13)	1.07 (1.02-1.13)	706 (380-2468)
10	−0.13 (−0.17 to −0.09)	−10.57 (−11.03 to −10.11)	−0.27 (−0.36 to −0.19)	−12.18 (−13.21 to −11.26)	1.25 (1.12-1.39)	1.24 (1.12-1.38)	198 (128-412)	1.15 (1.04-1.28)	1.15 (1.04-1.27)	330 (177-1234)
15	−0.19 (−0.25 to −0.13)	−11.26 (−11.95 to −10.57)	−0.39 (−0.52 to −0.27)	−13.55 (−15.05 to −12.18)	1.39 (1.18-1.64)	1.38 (1.18-1.62)	128 (78-275)	1.24 (1.06-1.45)	1.23 (1.06-1.44)	207 (111-823)
20	−0.25 (−0.32 to −0.17)	−11.95 (−12.75 to −11.03)	−0.49 (−0.64 to −0.34)	−14.70 (−16.43 to −12.98)	1.55 (1.25-1.94)	1.53 (1.24-1.90)	91 (54-198)	1.33 (1.08-1.64)	1.32 (1.08-1.62)	151 (78-618)
25	−0.30 (−0.39 to −0.21)	−12.52 (−13.55 to −11.48)	−0.56 (−0.73 to −0.39)	−15.51 (−17.46 to −13.55)	1.74 (1.32-2.28)	1.71 (1.31-2.22)	68 (40-155)	1.43 (1.11-1.85)	1.42 (1.111.82)	116 (59-449)
30	−0.35 (−0.45 to −0.25)	−13.09 (−14.25 to −11.95)	−0.61 (−0.79 to −0.42)	−16.09 (−18.16 to −13.90)	1.94 (1.39-2.69)	1.90 (1.38-2.60)	54 (30-128)	1.54 (1.13-2.09)	1.52 (1.132.04)	92 (46-380)
35	−0.40 (−0.51 to −0.29)	−13.67 (−14.94 to −12.41)	−0.63 (−0.81 to −0.44)	−16.32 (−18.39 to −14.13)	2.16 (1.47-3.18)	2.11 (1.46-3.04)	44 (24-106)	1.65 (1.15-2.37)	1.63 (1.15-2.30)	77 (37-330)
40	−0.44 (−0.55 to −0.33)	−14.13 (−15.39 to −12.87)	−0.63 (−0.82 to −0.44)	−16.32 (−18.50 to −14.13)	2.42 (1.56-3.75)	2.35 (1.54-3.55)	36 (19-89)	1.77 (1.18-2.68)	1.74 (1.18-2.59)	65 (30-275)
45	−0.47 (−0.59 to −0.36)	−14.47 (−15.86 to −13.21)	−0.63 (−0.82 to −0.43)	−16.32 (−18.50 to −14.02)	2.7 (1.65-4.42)	2.61 (1.63-4.13)	30 (15-77)	1.91 (1.20-3.03)	1.87 (1.20-2.91)	55 (25-248)
50	−0.50 (−0.61 to −0.39)	−14.82 (−16.09 to −13.55)	−0.62 (−0.82 to −0.42)	−16.20 (−18.50 to −13.90)	3.01 (1.74-5.21)	2.89 (1.71-4.79)	26 (13-68)	2.05 (1.23-3.42)	2.01 (1.22-3.26)	48 (21-216)
55	−0.53 (−0.63 to −0.42)	−15.17 (−16.32 to −13.90)	−0.62 (−0.83 to −0.40)	−16.20 (−18.62 to −13.67)	3.36 (1.84-6.15)	3.20 (1.81-5.56)	22 (11-60)	2.20 (1.25-3.87)	2.15 (1.24-3.65)	42 (18-198)
60	−0.55 (−0.65 to −0.44)	−15.39 (−16.55 to −14.13)	−0.61 (−0.83 to −0.39)	−16.09 (−18.62 to −13.55)	3.76 (1.95-7.25)	3.56 (1.91-6.42)	19 (9-53)	2.36 (1.28-4.38)	2.30 (1.27-4.09)	37 (16-177)
65	−0.57 (−0.67 to −0.46)	−15.62 (−16.77 to −14.36)	−0.60 (−0.84 to −0.37)	−15.97 (−18.73 to −13.32)	4.20 (2.06-8.56)	3.94 (2.02-7.40)	16 (8-48)	2.54 (1.30-4.95)	2.46 (1.29-4.58)	33 (14-165)
70	−0.58 (−0.69 to −0.47)	−15.74 (−17.00 to −14.47)	−0.60 (−0.85 to −0.34)	−15.97 (−18.84 to −12.98)	4.68 (2.17-10.09)	4.35 (2.12-8.49)	14 (6-43)	2.73 (1.33-5.6)	2.64 (1.325.11)	30 (12-151)
75	−0.60 (−0.72 to −0.48)	−15.97 (−17.35 to −14.59)	−0.59 (−0.86 to −0.32)	−15.86 (−18.96 to −12.75)	5.23 (2.30-11.91)	4.81 (2.24-9.72)	13 (6-39)	2.93 (1.36-6.33)	2.82 (1.35-5.70)	27 (10-138)
80	−0.62 (−0.75 to −0.48)	−16.20 (−17.70 to −14.59)	NA	NA	5.84 (2.43-14.04)	5.31 (2.36-11.06)	11 (5-36)	NA	NA	NA
85	−0.63 (−0.78 to −0.49)	−16.32 (−18.04 to −14.70)	NA	NA	6.52 (2.57-16.56)	5.85 (2.49-12.53)	10 (4-32)	NA	NA	NA

Open in a new tab

Abbreviations: ADHD, attention-deficit/hyperactivity disorder; ADHD-RS, attention-deficit/hyperactivity disorder rating scale; NA, not applicable; NNH, number needed to harm; RR, relative risk.

^{^a}

Doses in mg/d refers to immediate-release methylphenidate hydrochloride and mixed amphetamine salts for methylphenidate and amphetamines, respectively.

Figure 3. — Standardized mean difference (SMD) for change in attention-deficit/hyperactivity disorder (ADHD) symptoms (A) and odds ratio (OR) for tolerability (B) in network meta-analysis. All values are presented with 95% CIs. RCTs indicates randomized clinical trials.

Tolerability

The dose-response curve based on 7 fixed-dose RCTs and 17 stimulant arms (dose range 15 to 82.5 mg per day) indicated a log-linear association with increments of 2.23% (95% CI, 1.11%-3.35%) on an exponential scale in the risk of dropping out due to adverse events for every 1 mg per day increase throughout the entire dose range for which data were available (Figure 2B; Table, which also reports number needed to harm values). NMA corroborated that unlicensed doses were associated with decreased tolerability in comparison with licensed doses (16 studies; OR, 2.02; 95% CI, 1.19-3.43; P = .01; moderate certainty of evidence) (Figure 3B; eFigure 1, eFigure 3, and eTable 4 in Supplement 1).

Acceptability

The dose-response curve for acceptability indicated similar findings to those for tolerability. NMA did not indicate significant differences in acceptability (20 studies; OR, 1.59; 95% CI, 0.93-2.73; P = .09; low certainty of evidence) between unlicensed and licensed doses (eResults 3 in Supplement 1).

Heterogeneity, Incoherence, and Sensitivity Analyses

There was low (VPC of approximately 0%) (change in ADHD symptoms; acceptability) and low to moderate (VPC less than 50%) (tolerability) heterogeneity in the dose-response meta-analyses (eFigure 4, eFigure 5, and eResults 3 in Supplement 1). There was low-moderate (tolerability: τ² = 0.05; I² = 8%) and moderate-high (change in ADHD symptoms: τ² = 0.05; I² = 68%; acceptability: τ² = 0.24; I² = 63%) heterogeneity in NMAs. There was no evidence of incoherence in any of the NMAs (eTable 5, eTable 6, and eResults 3 in Supplement 1). Contour-enhanced funnel plots indicated some asymmetry for change in ADHD symptoms (eFigure 6, eFigure 7, eFigure 8, and eResults 3 in Supplement 1). The findings from the dose-response meta-analyses (eFigures 9 through 13 and eResults 3 in Supplement 1) or the NMAs (Figure 3; eResults 3 in Supplement 1) did not change considerably in any of the sensitivity analyses.

Amphetamines

Change in ADHD Symptoms

The dose-response curve based on 10 fixed-dose RCTs and 21 stimulant arms (dose range 12.5 to 75 mg per day) demonstrated an initial sharp decrease followed by a plateau beyond 30 to 35 mg per day (Figure 4A; Table). Consistently, the ED50 was estimated at 12.5 mg per day while the ED95 was estimated at 30 mg per day. NMA corroborated that unlicensed doses were not associated with larger reductions of symptoms in comparison with licensed doses (14 studies; SMD, −0.08; 95% CI, −0.24-0.08; P = .31; very low certainty of evidence) (Figure 3A; eFigure 14, eFigure 15, and eTable 4 in Supplement 1).

Figure 4. — Dose-response curve for change in attention-deficit/hyperactivity disorder symptom severity (A) and tolerability (B). The curves are presented until the maximum dose for which data were available for equivalent doses of amphetamines. The shaded areas indicate 95% CIs. The black dotted line indicates the US Food and Drug Administration (FDA) maximum recommended dose for lisdexamfetamine; the blue dashed line indicates the FDA recommended maximum dose for immediate release mixed amphetamine salts.

Tolerability

The dose-response curve based on 7 fixed-dose RCTs and 15 stimulant arms (dose range 12.5 to 75 mg per day) indicated a log-linear association with increments of 1.44% (95% CI, 0.41%-2.49%) on an exponential scale in the risk of dropping out due to adverse events for every 1 mg per day increase throughout the entire dose range for which data were available (Figure 4B; Table). NMA did not indicate significant differences in tolerability (9 studies; OR, 1.19; 95% CI, 0.71-2.02; P = .51; low) between unlicensed and licensed doses (Figure 3B; eFigure 14, eFigure 16, and eTable 4 in Supplement 1).

Acceptability

The dose-response curve did not indicate a significant association with acceptability (χ² = 0.07; P = .79). NMA corroborated there were no differences in acceptability (10 studies; OR, 0.81; 95% CI, 0.54-1.22; P = .31; low) between unlicensed and licensed doses (eResults 3 in Supplement 1).

Heterogeneity, Incoherence, and Sensitivity Analyses

There was low (tolerability), low to moderate (acceptability), and moderate to high (VPC 50% or more) (change in ADHD symptoms) heterogeneity in the dose-response meta-analyses (eFigure 17, eFigure 18, and eResults 3 in Supplement 1). There was low heterogeneity in the NMAs (change in ADHD symptom severity: τ² = 0.002; I² = 55%; tolerability: τ²<0.0001; I² = 0%; acceptability: τ² = 0.04; I² = 10%). There was some evidence of global incoherence (Q = 5.09; P = .07) in the NMA for acceptability. There was no further evidence of incoherence in any of the NMAs (eTable 5, eTable 6, and eResults 3 in Supplement 1). Contour-enhanced funnel plots did not indicate asymmetry for change in ADHD symptom severity (eFigure 19 in Supplement 1). The findings from the dose-response meta-analyses (eFigures 20 through 26 and eResults 3 in Supplement 1) or the NMAs (Figure 3, eResults 3 in Supplement 1) did not change considerably in any of the sensitivity analyses.

Discussion

This study included data from 47 RCTs involving 7714 participants to investigate treatment outcomes with licensed and unlicensed doses of stimulants in adults with ADHD. Based on averages, our findings indicated that higher doses of stimulants were generally associated with larger reductions of symptoms within the licensed range. For methylphenidate, unlicensed doses were associated with greater reductions of symptoms in comparison with licensed doses; however, the incremental benefit was small and accompanied by considerable increase in the risk of treatment discontinuation due to adverse events. For amphetamines, unlicensed doses were not associated with additional reductions of symptoms in comparison with licensed doses. There were especially few RCTs evaluating unlicensed doses for amphetamines and it is possible that differences across stimulant classes at the unlicensed doses could be related to data availability; therefore, our results could be interpreted as indicative that there are signals for greater improvement above the licensed range for stimulants in general, albeit those were clearer for methylphenidate than amphetamines based on the currently available data. Regardless, taken together our findings still indicated that, based on averages, unlicensed doses of stimulants may not have positive risk benefits for adults with ADHD.

Further advances in precision medicine⁴⁵ may enable the identification of individualized optimal doses and the stratification of patients by predicted positive/negative risk benefits with unlicensed doses without the need for a trial-and-error process. Nevertheless, in the current absence of such personalized evidence, intention-to-treat data from all RCTs provide the best evidence to guide decision-making in routine clinical practice at the group level (ie, for a generic adult with ADHD). Based on our findings, practitioners should be aware that adults with ADHD may experience the most reductions of symptoms with stimulants at licensed doses and that additional reductions of symptoms with unlicensed doses may be small and possibly nonclinically meaningful (eg, less than 10 to 15 ADHD-RS reductions).⁴⁶ However, the tolerability of stimulants may continuously decrease with increments in dose. Therefore, practitioners should generally consider unlicensed doses cautiously.

Importantly, because the risk benefits based on averages from our study will not generalize to every patient, our study does not provide evidence that practitioners should not trial unlicensed doses of stimulants in routine clinical practice. Indeed, individuals differ in their response to these medications⁴⁷ and some adults with ADHD may experience larger reductions of symptoms, or less evident adverse events, with unlicensed doses than the average patient as reflected in our findings. Therefore, for a subgroup of patients, unlicensed doses may be well tolerated and required for the maximal reductions of symptoms and improvement in functioning. Besides, stimulants are short acting and their effects wash out within the day, which enables the assessment of benefits and harms quickly, within 5 to 7 days.^9,12 If needed (ie, if harms outweigh the benefits), practitioners may go back to lower doses. It should also be considered that for some medications (eg, mixed amphetamine salts) the licensed doses of extended-release formulations are smaller than those of immediate-release formulations or of extended-release formulations in children, so that the unlicensed doses are not particularly high.

While practitioners may trial unlicensed doses of stimulants in adults with ADHD who have shown partial response and good tolerability at the maximum licensed dose, they should be aware that there may not be large gains in the patients’ response to the medication with those further increments in dose. Other strategies to augment response to stimulants (eg, switching to other stimulant class) should be considered and may be implemented rather than simply raising the dose beyond the licensed range.¹⁰ Therefore, our study provides support for expert opinion that licensed doses may be sufficient for most patients^10,11,48 and calls into question the widespread use of unlicensed doses in routine practice,⁴⁹ despite earlier reports.⁵⁰

The quality of evidence was not high, which underscores the importance of additional studies evaluating the impact of unlicensed doses of stimulants on the treatment outcomes of adults with ADHD. Flexible-dose RCTs may be particularly appropriate to study the risk-benefits of escalation to unlicensed doses because they ensure that participants receive optimal doses. Intriguingly, our sensitivity analyses focused on flexible-dose RCTs did not yield conclusive evidence of stronger associations between reductions of symptoms and unlicensed doses in comparison to associations from the main analyses (Figure 3). For methylphenidate, the point estimate was rather similar; for amphetamines, there was considerable imprecision and the 95% CI was compatible with a relevant effect (absolute SMD more than 0.2) in both directions. While the scarcity of flexible-dose RCTs, particularly for amphetamines, may have limited the statistical power to detect differences, if those were large, they would likely be detected anyway.

Variation in how flexible-dose RCTs operationalized unlicensed doses may have contributed to the underestimation of the benefits of such high doses. For instance, for methylphenidate, for which most of the data were available, RCTs evaluating unlicensed doses, which were predominantly flexible-dose studies (80%), were more likely to use weight-adjusted doses. Weight-adjusted doses have been shown to correlate poorly with reductions of symptoms in youth⁵¹ and current practices are generally based on fixed-dose schedules. Nevertheless, current recommendations from CPGs about stimulant dosing in adults with ADHD have been based on data from RCTs that adopted weight-adjusted dosing^14,15 and yielded the largest effect sizes across RCTs.¹⁶ RCTs evaluating unlicensed doses were also smaller and older than RCTs evaluating licensed doses; these characteristics may have contributed to larger effect sizes favoring unlicensed doses in relation to licensed doses.⁵² Regardless, our study highlights the need for additional rigorously conducted flexible-dose RCTs, which is in contrast with current FDA guidance on the development of stimulant drugs.

Strengths and Limitations

The most notable strengths of our study were our methodological rigor and the consistent results across different analytical strategies. However, our study also has limitations. Although we made our best efforts to identify all available RCTs, we cannot rule out the possibility of having missed relevant studies. Our approach to compare doses of stimulants may not be directly translated into routine clinical practice because practitioners must also account for the differential impact of ADHD symptoms on activities along the day when prescribing stimulants for ADHD. For instance, extended-release medications have differing pharmacokinetic properties that could influence treatment outcomes, even if given in the same total daily dose as immediate-release medications.^53,54 Indeed, the COMACS study⁵⁵ showed that 2 bioequivalent doses of different extended-release formulations of methylphenidate may have different effects on ADHD symptoms across the day depending on their immediate-/extended-release ratios. For amphetamines, our conversions based on total amphetamine base equivalence did not account for differences in potency of d-/l-amphetamine as a dopamine/norepinephrine reuptake inhibitor,⁵⁶ one of the proposed mechanisms of action for amphetamines.⁵⁷ Nevertheless, other researchers in the field (eg, investigators from the Multimodal Treatment Study of ADHD) have adopted similar approaches when assessing stimulant doses.^58,59,60 Additionally, our sensitivity analyses for the dose-response curves considering other conversion factors did not yield considerable differences. Our definition of tolerability did not consider individual adverse events specifically.⁶¹ Nevertheless, using dropout rates due to adverse events as a proxy for tolerability is a typical procedure.^13,23,52 Lastly, our analyses based on averages do not account for individual factors that could contribute to different dose-outcome associations (eg, baseline symptoms,⁶² genetic background⁶³). However, others have adopted similar procedures to investigate dose-outcome associations in psychiatry,^64,65,66 general medicine^67,68 and public health.^69,70

Conclusion

The findings from this systematic review and meta-analysis provide the best evidence available regarding treatment outcomes with licensed and unlicensed stimulants doses for adults with ADHD. Our results, based on averages, showed that most of the reductions of symptoms were associated with stimulants at licensed doses. Unlicensed doses were not only associated with small and possibly nonclinically meaningful additional reductions in symptoms, but also with marked decreased tolerability. Therefore, based on averages, unlicensed doses of stimulant may not have positive risk benefits. In general, practitioners should consider unlicensed doses cautiously. Practitioners may trial unlicensed doses if needed and tolerated but should be aware that there may not be large gains in the response to the medication with those further increments in dose. Our findings should be incorporated in future recommendations from clinical practice guidelines to ensure transparent, evidence-based discussions with patients about expected risk-benefits with unlicensed doses of stimulants.

Supplement 1.

eMethods 1. PRISMA checklist

eMethods 2. Protocol amendments

eMethods 3. Search strategy

eMethods 4. Eligibility criteria

eMethods 5. Data extraction

eMethods 6. Conversion of stimulant doses

eMethods 7. Conversion of effect sizes

eMethods 8. Distribution of clinical and methodological characteristics across comparisons for change in ADHD symptom severity.

eMethods 9. Within-study risk of bias and evaluation of certainty of evidence

eMethods 10. Description and rationale for the sensitivity analyses

eResults 1. List of included studies

eResults 2. Risk of bias assessment of individual studies for reduction in ADHD symptoms

eResults 3. Results for acceptability (dropouts due to any reason)

eTable 1. FDA and clinical practice guideline recommended dose ranges and conversion factors

eTable 2. Characteristics of studies

eTable 3. Dosing procedures of flexible-dose studies

eTable 4. Certainty of evidence assessment for change in ADHD symptom severity and tolerability

eFigure 1. Network estimates for the network meta-analyses of change in ADHD symptom severity and tolerability for methylphenidate

eFigure 2. Network plot for the network meta-analysis of change in ADHD symptom severity for methylphenidate

eFigure 3. Network plot for the network meta-analysis of tolerability for methylphenidate

eFigure 4. Variance partition coefficient plot (VPC) for the dose-response meta-analysis of change in ADHD symptom severity for methylphenidate

eFigure 5. Variance partition coefficient plot (VPC) for the dose-response meta-analysis of tolerability for methylphenidate

eFigure 6. Contour-enhanced funnel plot for the “within FDA against placebo” comparison for change in ADHD symptom severity for methylphenidate

eFigure 7. Contour-enhanced funnel plot for the “beyond FDA against placebo” comparison for change in ADHD symptom severity for methylphenidate

eFigure 8. Contour-enhanced funnel plot for the “within FDA against placebo” comparison for tolerability for methylphenidate

eFigure 9. Dose-response curve for change in ADHD symptom severity with knots at the 25th, 50th, and 75th percentiles for methylphenidate

eFigure 10. Dose-response curve for change in ADHD symptom severity without randomized controlled trials rated at high risk of bias for methylphenidate

eFigure 11. Dose-response curve for change in ADHD symptom severity without crossover randomized controlled trials for methylphenidate

eFigure 12. Dose-response curve for change in ADHD symptom severity with different conversion factor for methylphenidate extended-release

eFigure 13. Dose-response curve for tolerability with different conversion factor for methylphenidate extended-release

eFigure 14. Network estimates for the network meta-analyses of change in ADHD symptom severity and tolerability for amphetamines

eFigure 15. Network plot for the network meta-analysis of change in ADHD symptom severity for amphetamines

eFigure 16. Network plot for the network meta-analysis of tolerability for amphetamines

eFigure 17. Variance partition coefficient plot (VPC) for the dose-response meta-analysis of change in ADHD symptom severity for amphetamines

eFigure 18. Variance partition coefficient plot (VPC) for the dose-response meta-analysis of tolerability for amphetamines

eFigure 19. Contour-enhanced funnel plot for within FDA against placebo comparison for change in ADHD symptom severity for amphetamines

eFigure 20. Dose-response curve for change in ADHD symptom severity with knots at the 25th, 50th, and 75th percentiles for amphetamines

eFigure 21. Dose-response curve for change in ADHD symptom severity without randomized controlled trials rated at high risk of bias for amphetamines

eFigure 22. Dose-response curve for change in ADHD symptom severity without crossover randomized controlled trials for amphetamines

eFigure 23. Dose-response curve for change in ADHD symptom severity without lisdexamfetamine for amphetamines

eFigure 24. Dose-response curve for tolerability without lisdexamfetamine for amphetamines

eFigure 25. Dose-response curve for change in ADHD symptom severity with mixed amphetamine salts only

eFigure 26. Dose-response curve for tolerability with mixed amphetamine salts only

Click here for additional data file.^{(2.6MB, pdf)}

Supplement 2.

Data sharing statement

Click here for additional data file.^{(12.1KB, pdf)}

References

1.Canadian ADHD Resource Alliance . Canadian ADHD practice guidelines (4.1 edition). Accessed September 15, 2023.
2.National Institute for Health and Care Excellence . Attention deficit hyperactivity disorder: diagnosis and management. Accessed September 15, 2023. https://www.nice.org.uk/guidance/ng87 [PubMed]
3.Bolea-Alamañac B, Nutt DJ, Adamou M, et al. ; British Association for Psychopharmacology . Evidence-based guidelines for the pharmacological management of attention deficit hyperactivity disorder: update on recommendations from the British Association for Psychopharmacology. J Psychopharmacol. 2014;28(3):179-203. doi: 10.1177/0269881113519509 [DOI] [PubMed] [Google Scholar]
4.Kessler RC, Adler L, Barkley R, et al. The prevalence and correlates of adult ADHD in the United States: results from the National Comorbidity Survey Replication. Am J Psychiatry. 2006;163(4):716-723. doi: 10.1176/ajp.2006.163.4.716 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Bernardi S, Faraone SV, Cortese S, et al. The lifetime impact of attention deficit hyperactivity disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Psychol Med. 2012;42(4):875-887. doi: 10.1017/S003329171100153X [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Lichtenstein P, Halldner L, Zetterqvist J, et al. Medication for attention deficit-hyperactivity disorder and criminality. N Engl J Med. 2012;367(21):2006-2014. doi: 10.1056/NEJMoa1203241 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Dalsgaard S, Østergaard SD, Leckman JF, Mortensen PB, Pedersen MG. Mortality in children, adolescents, and adults with attention deficit hyperactivity disorder: a nationwide cohort study. Lancet. 2015;385(9983):2190-2196. doi: 10.1016/S0140-6736(14)61684-6 [DOI] [PubMed] [Google Scholar]
8.Schein J, Adler LA, Childress A, et al. Economic burden of attention-deficit/hyperactivity disorder among adults in the United States: a societal perspective. J Manag Care Spec Pharm. 2022;28(2):168-179. doi: 10.18553/jmcp.2021.21290 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.McGough JJ. Treatment controversies in Adult ADHD. Am J Psychiatry. 2016;173(10):960-966. doi: 10.1176/appi.ajp.2016.15091207 [DOI] [PubMed] [Google Scholar]
10.Cortese S, Newcorn JH, Coghill D. A practical, evidence-informed approach to managing stimulant-refractory attention deficit hyperactivity disorder (ADHD). CNS Drugs. 2021;35(10):1035-1051. doi: 10.1007/s40263-021-00848-3 [DOI] [PubMed] [Google Scholar]
11.Greenhill LL, Pliszka S, Dulcan MK, et al. ; American Academy of Child and Adolescent Psychiatry . Practice parameter for the use of stimulant medications in the treatment of children, adolescents, and adults. J Am Acad Child Adolesc Psychiatry. 2002;41(2)(suppl):26S-49S. doi: 10.1097/00004583-200202001-00003 [DOI] [PubMed] [Google Scholar]
12.Wilens TE, Morrison NR, Prince J. An update on the pharmacotherapy of attention-deficit/hyperactivity disorder in adults. Expert Rev Neurother. 2011;11(10):1443-1465. doi: 10.1586/ern.11.137 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Cortese S, Adamo N, Del Giovane C, et al. Comparative efficacy and tolerability of medications for attention-deficit hyperactivity disorder in children, adolescents, and adults: a systematic review and network meta-analysis. Lancet Psychiatry. 2018;5(9):727-738. doi: 10.1016/S2215-0366(18)30269-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Spencer T, Wilens T, Biederman J, Faraone SV, Ablon JS, Lapey K. A double-blind, crossover comparison of methylphenidate and placebo in adults with childhood-onset attention-deficit hyperactivity disorder. Arch Gen Psychiatry. 1995;52(6):434-443. doi: 10.1001/archpsyc.1995.03950180020004 [DOI] [PubMed] [Google Scholar]
15.Spencer T, Biederman J, Wilens T, et al. A large, double-blind, randomized clinical trial of methylphenidate in the treatment of adults with attention-deficit/hyperactivity disorder. Biol Psychiatry. 2005;57(5):456-463. doi: 10.1016/j.biopsych.2004.11.043 [DOI] [PubMed] [Google Scholar]
16.Faraone SV, Spencer T, Aleardi M, Pagano C, Biederman J. Meta-analysis of the efficacy of methylphenidate for treating adult attention-deficit/hyperactivity disorder. J Clin Psychopharmacol. 2004;24(1):24-29. doi: 10.1097/01.jcp.0000108984.11879.95 [DOI] [PubMed] [Google Scholar]
17.Huss M, Ginsberg Y, Tvedten T, et al. Methylphenidate hydrochloride modified-release in adults with attention deficit hyperactivity disorder: a randomized double-blind placebo-controlled trial. Adv Ther. 2014;31(1):44-65. doi: 10.1007/s12325-013-0085-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Medori R, Ramos-Quiroga JA, Casas M, et al. A randomized, placebo-controlled trial of three fixed dosages of prolonged-release OROS methylphenidate in adults with attention-deficit/hyperactivity disorder. Biol Psychiatry. 2008;63(10):981-989. doi: 10.1016/j.biopsych.2007.11.008 [DOI] [PubMed] [Google Scholar]
19.Spencer TJ, Adler LA, McGough JJ, Muniz R, Jiang H, Pestreich L; Adult ADHD Research Group . Efficacy and safety of dexmethylphenidate extended-release capsules in adults with attention-deficit/hyperactivity disorder. Biol Psychiatry. 2007;61(12):1380-1387. doi: 10.1016/j.biopsych.2006.07.032 [DOI] [PubMed] [Google Scholar]
20.Evidence-Based Medicine Working Group . Evidence-based medicine. a new approach to teaching the practice of medicine. JAMA. 1992;268(17):2420-2425. doi: 10.1001/jama.1992.03490170092032 [DOI] [PubMed] [Google Scholar]
21.Furukawa TA, Cipriani A, Cowen PJ, Leucht S, Egger M, Salanti G. Optimal dose of selective serotonin reuptake inhibitors, venlafaxine, and mirtazapine in major depression: a systematic review and dose-response meta-analysis. Lancet Psychiatry. 2019;6(7):601-609. doi: 10.1016/S2215-0366(19)30217-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Furukawa TA, Salanti G, Cowen PJ, Leucht S, Cipriani A. No benefit from flexible titration above minimum licensed dose in prescribing antidepressants for major depression: systematic review. Acta Psychiatr Scand. 2020;141(5):401-409. doi: 10.1111/acps.13145 [DOI] [PubMed] [Google Scholar]
23.Farhat LC, Flores JM, Behling E, et al. The effects of stimulant dose and dosing strategy on treatment outcomes in attention-deficit/hyperactivity disorder in children and adolescents: a meta-analysis. Mol Psychiatry. 2022;27(3):1562-1572. doi: 10.1038/s41380-021-01391-9 [DOI] [PubMed] [Google Scholar]
24.Cipriani A, Higgins JPT, Geddes JR, Salanti G. Conceptual and technical challenges in network meta-analysis. Ann Intern Med. 2013;159(2):130-137. doi: 10.7326/0003-4819-159-2-201307160-00008 [DOI] [PubMed] [Google Scholar]
25.Higgins JPT, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions version 6.3. Accessed September 15, 2023. https://training.cochrane.org/handbook/archive/v6.3
26.Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Syst Rev. 2021;10(1):89. doi: 10.1186/s13643-021-01626-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.DuPaul GJ, Power TJ, Anastopoulos AD, Reid R. ADHD Rating Scale—IV: Checklists, Norms, and Clinical Interpretation. Guilford Press; 1998. [Google Scholar]
28.Conners CK, Erhardt D, Sparrow EP. Conners’ Adult ADHD Rating Scales (CAARS): Technical Manual. Multi-Health Systems North Tonawanda; 1999. [Google Scholar]
29.Crippa A, Discacciati A, Bottai M, Spiegelman D, Orsini N. One-stage dose-response meta-analysis for aggregated data. Stat Methods Med Res. 2019;28(5):1579-1596. doi: 10.1177/0962280218773122 [DOI] [PubMed] [Google Scholar]
30.Harrell FE. Regression Modelling Strategies With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Springer International Publishing; 2015, doi: 10.1007/978-3-319-19425-7. [DOI] [Google Scholar]
31.Crippa A, Orsini N. Multivariate dose-response meta-analysis: the dosresmeta R package. J Stat Softw. 2016;72(1):1-15. doi: 10.18637/jss.v072.c01 [DOI] [Google Scholar]
32.Crippa A, Orsini N. Dose-response meta-analysis of differences in means. BMC Med Res Methodol. 2016;16:91. doi: 10.1186/s12874-016-0189-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Schwarzer G, Carpenter JR, Rücker G. Network Meta-Analysis. In: Schwarzer G, Carpenter JR, Rücker G, eds. Meta-Analysis with R. Springer International Publishing; 2015:187-216. doi: 10.1007/978-3-319-21416-0_8 [DOI] [Google Scholar]
34.Rhodes KM, Turner RM, Higgins JP. Predictive distributions were developed for the extent of heterogeneity in meta-analyses of continuous outcome data. J Clin Epidemiol. 2015;68(1):52-60. doi: 10.1016/j.jclinepi.2014.08.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Turner RM, Davey J, Clarke MJ, Thompson SG, Higgins JPT. Predicting the extent of heterogeneity in meta-analysis, using empirical data from the Cochrane Database of Systematic Reviews. Int J Epidemiol. 2012;41(3):818-827. doi: 10.1093/ije/dys041 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Krahn U, Binder H, König J. A graphical tool for locating inconsistency in network meta-analyses. BMC Med Res Methodol. 2013;13:35. doi: 10.1186/1471-2288-13-35 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.König J, Krahn U, Binder H. Visualizing the flow of evidence in network meta-analysis and characterizing mixed treatment comparisons. Stat Med. 2013;32(30):5414-5429. doi: 10.1002/sim.6001 [DOI] [PubMed] [Google Scholar]
38.Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 28 2019;366:l4898. doi: 10.1136/bmj.l4898 [DOI] [PubMed] [Google Scholar]
39.Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Contour-enhanced meta-analysis funnel plots help distinguish publication bias from other causes of asymmetry. J Clin Epidemiol. 2008;61(10):991-996. doi: 10.1016/j.jclinepi.2007.11.010 [DOI] [PubMed] [Google Scholar]
40.Nikolakopoulou A, Higgins JPT, Papakonstantinou T, et al. CINeMA: an approach for assessing confidence in the results of a network meta-analysis. PLoS Med. 2020;17(4):e1003082. doi: 10.1371/journal.pmed.1003082 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Papakonstantinou T, Nikolakopoulou A, Higgins JPT, Egger M, Salanti G. CINeMA: software for semiautomated assessment of the confidence in the results of network meta-analysis. Campbell Syst Rev. 2020;16(1):e1080. doi: 10.1002/cl2.1080 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. Evid Based Ment Health. 2019;22(4):153-160. doi: 10.1136/ebmental-2019-300117 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Balduzzi S, Rücker G, Nikolakopoulou A, et al. netmeta: An R package for network meta-analysis using frequentist methods. J Stat Softw. 2023;106(2):1-40. doi: 10.18637/jss.v106.i0237138589 [DOI] [Google Scholar]
44.Farhat LC, Flores JM, Avila-Quintero VJ, et al. List of excluded records. Open Science Framework. Accessed September 25, 2023. doi: 10.17605/OSF.IO/47FS9 [DOI] [Google Scholar]
45.Buitelaar J, Bölte S, Brandeis D, et al. Toward precision medicine in ADHD. Front Behav Neurosci. 2022;16:900981. doi: 10.3389/fnbeh.2022.900981 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Goodman D, Faraone SV, Adler LA, Dirks B, Hamdani M, Weisler R. Interpreting ADHD rating scale scores: linking ADHD rating scale scores and CGI levels in two randomized controlled trials of lisdexamfetamine dimesylate in ADHD. Prim Psychiatry. 2010;17(3):44. [Google Scholar]
47.Spencer T, Biederman J, Wilens T, Harding M, O’Donnell D, Griffin S. Pharmacotherapy of attention-deficit hyperactivity disorder across the life cycle. J Am Acad Child Adolesc Psychiatry. 1996;35(4):409-432. doi: 10.1097/00004583-199604000-00008 [DOI] [PubMed] [Google Scholar]
48.Asherson P, Manor I, Huss M. Attention-deficit/hyperactivity disorder in adults: update on clinical presentation and care. Neuropsychiatry (London). 2014;4(1):109-128. doi: 10.2217/npy.14.16 [DOI] [Google Scholar]
49.Zhang L, Lagerberg T, Chen Q, et al. Prediction of treatment dosage and duration from free-text prescriptions: an application to ADHD medications in the Swedish prescribed drug register. Evid Based Ment Health. 2021;24(4):146-152. doi: 10.1136/ebmental-2020-300231 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Olfson M, Marcus SC, Zhang HF, Wan GJ. Stimulant dosing in the community treatment of adult attention-deficit/hyperactivity disorder. J Clin Psychopharmacol. 2008;28(2):255-257. doi: 10.1097/JCP.0b013e31816740be [DOI] [PubMed] [Google Scholar]
51.Rapport MD, DuPaul GJ, Kelly KL. Attention deficit hyperactivity disorder and methylphenidate: the relationship between gross body weight and drug response in children. Psychopharmacol Bull. 1989;25(2):285-290. [PubMed] [Google Scholar]
52.Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet. 2018;391(10128):1357-1366. doi: 10.1016/S0140-6736(17)32802-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Weisler RH. Review of long-acting stimulants in the treatment of attention deficit hyperactivity disorder. Expert Opin Pharmacother. 2007;8(6):745-758. doi: 10.1517/14656566.8.6.745 [DOI] [PubMed] [Google Scholar]
54.López FA, Leroux JR. Long-acting stimulants for treatment of attention-deficit/hyperactivity disorder: a focus on extended-release formulations and the prodrug lisdexamfetamine dimesylate to address continuing clinical challenges. Atten Defic Hyperact Disord. 2013;5(3):249-265. doi: 10.1007/s12402-013-0106-x [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Swanson JM, Wigal SB, Wigal T, et al. ; COMACS Study Group . A comparison of once-daily extended-release methylphenidate formulations in children with attention-deficit/hyperactivity disorder in the laboratory school (the Comacs Study). Pediatrics. 2004;113(3 Pt 1):e206-e216. doi: 10.1542/peds.113.3.e206 [DOI] [PubMed] [Google Scholar]
56.Heal DJ, Smith SL, Gosden J, Nutt DJ. Amphetamine, past and present–a pharmacological and clinical perspective. J Psychopharmacol. 2013;27(6):479-496. doi: 10.1177/0269881113482532 [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Caye A, Swanson JM, Coghill D, Rohde LA. Treatment strategies for ADHD: an evidence-based guide to select optimal treatment. Mol Psychiatry. 2019;24(3):390-408. doi: 10.1038/s41380-018-0116-3 [DOI] [PubMed] [Google Scholar]
58.Swanson JM, Arnold LE, Molina BSG, et al. ; MTA Cooperative Group . Young adult outcomes in the follow-up of the multimodal treatment study of attention-deficit/hyperactivity disorder: symptom persistence, source discrepancy, and height suppression. J Child Psychol Psychiatry. 2017;58(6):663-678. doi: 10.1111/jcpp.12684 [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Greenhill LL, Swanson JM, Hechtman L, et al. ; MTA Cooperative Group . Trajectories of growth associated with long-term stimulant medication in the multimodal treatment study of attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry. 2020;59(8):978-989. doi: 10.1016/j.jaac.2019.06.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Molina BSG, Kennedy TM, Howard AL, et al. Association between stimulant treatment and substance use through adolescence into early adulthood. JAMA Psychiatry. 2023;80(9):933-941. doi: 10.1001/jamapsychiatry.2023.2157 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Tomlinson A, Efthimiou O, Boaden K, et al. Side effect profile and comparative tolerability of 21 antidepressants in the acute treatment of major depression in adults: protocol for a network meta-analysis. Evid Based Ment Health. 2019;22(2):61-66. doi: 10.1136/ebmental-2019-300087 [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Weisler RH, Biederman J, Spencer TJ, et al. Mixed amphetamine salts extended-release in the treatment of adult ADHD: a randomized, controlled trial. CNS Spectr. 2006;11(8):625-639. doi: 10.1017/S1092852900013687 [DOI] [PubMed] [Google Scholar]
63.Froehlich TE, Epstein JN, Nick TG, et al. Pharmacogenetic predictors of methylphenidate dose-response in attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry. 2011;50(11):1129-1139.e2. doi: 10.1016/j.jaac.2011.08.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Leucht S, Bauer S, Siafis S, et al. Examination of dosing of antipsychotic drugs for relapse prevention in patients with stable schizophrenia: a meta-analysis. JAMA Psychiatry. 2021;78(11):1238-1248. doi: 10.1001/jamapsychiatry.2021.2130 [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Sabé M, Pallis K, Solmi M, Crippa A, Sentissi O, Kaiser S. Comparative effects of 11 antipsychotics on weight gain and metabolic function in patients with acute schizophrenia: a dose-response meta-analysis. J Clin Psychiatry. 2023;84(2):22r14490. doi: 10.4088/JCP.22r14490 [DOI] [PubMed] [Google Scholar]
66.Hsu CW, Tsai SY, Tseng PT, et al. Differences in the prophylactic effect of serum lithium levels on depression and mania in bipolar disorder: a dose-response meta-analysis. Eur Neuropsychopharmacol. 2022;58:20-29. doi: 10.1016/j.euroneuro.2022.01.112 [DOI] [PubMed] [Google Scholar]
67.Lin DSH, Yu AL, Lo HY, et al. Differential effects of sodium-glucose cotransporter 2 inhibitors on cardiovascular and renal outcomes according to renal function: a dose-response meta-analysis involving 10 randomized clinical trials and 71 553 individuals. Eur J Endocrinol. 2023;189(1):S17-S25. doi: 10.1093/ejendo/lvad078 [DOI] [PubMed] [Google Scholar]
68.Smith AD, Crippa A, Woodcock J, Brage S. Physical activity and incident type 2 diabetes mellitus: a systematic review and dose-response meta-analysis of prospective cohort studies. Diabetologia. 2016;59(12):2527-2545. doi: 10.1007/s00125-016-4079-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Salanti G, Peter N, Tonia T, et al. ; MHCOVID Crowd Investigators; MHCOVID Crowd Investigators† . The impact of the COVID-19 pandemic and associated control measures on the mental health of the general population: a systematic review and dose-response meta-analysis. Ann Intern Med. 2022;175(11):1560-1571. doi: 10.7326/M22-1507 [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Crippa A, Discacciati A, Larsson SC, Wolk A, Orsini N. Coffee consumption and mortality from all causes, cardiovascular disease, and cancer: a dose-response meta-analysis. Am J Epidemiol. 2014;180(8):763-775. doi: 10.1093/aje/kwu194 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials