The occurrence of adverse drug reactions reported for attention deficit hyperactivity disorder (ADHD) medications in the pediatric population: a qualitative review of empirical studies.

BACKGROUND: To review empirical studies of adverse drug reactions (ADRs) reported to be associated with the use of medications generally licensed for treatment of attention deficit hyperactivity disorder (ADHD) symptoms in the pediatric population.
METHODS: PubMed, Embase, and PsycINFO(®) databases were searched from origin until June 2011. Studies reporting ADRs from amphetamine derivates, atomoxetine, methylphenidate, and modafinil in children from birth to age 17 were included. Information about ADR reporting rates, age and gender of the child, type, and seriousness of ADRs, setting, study design, ADR assessors, authors, and funding sources were extracted.
RESULTS: The review identified 43 studies reporting ADRs associated with medicines for treatment of ADHD in clinical studies covering approximately 7000 children, the majority of 6- to 12-year-old boys, and particularly in the United States of America (USA). The most frequently reported ADRs were decrease in appetite, gastrointestinal pain, and headache. There were wide variations in reported ADR occurrence between studies of similar design, setting, included population, and type of medication. Reported ADRs were primarily assessed by the children/their parents, and very few ADRs were rated as being serious. A large number of children dropped out of studies due to serious ADRs, and therefore, the actual number of serious ADRs from use of psychostimulants is probably higher. A large number of studies were conducted by the same groups of authors and sponsored by the pharmaceutical companies manufacturing the respective medications.
CONCLUSION: Reported ADRs from use of psychostimulants in children were found in clinical trials of short duration. Since ADHD medications are prescribed for long-term treatment, there is a need for long-term safety studies. The pharmaceutical companies should make all information about ADRs reported for these medications accessible to the public, and further studies are needed on the impact of the link between researchers and the manufacturers of the respective products.

Introduction

Psychostimulants, such as amphetamine derivates, methylphenidate, and modafinil, as well as the nonstimulant medication atomoxetine, are considered first-line medication treatment of attention deficit hyperactivity disorder (ADHD) symptoms in the pediatric population.1 Case reports on serious cardiovascular adverse drug reactions (ADRs), sudden death, and psychiatric disorders led regulatory agencies to warn against the use of methylphenidate in the pediatric population in 2006 and 2007.2,3 In 2006, warnings were also linked to atomoxetine use due to reports of hepatotoxicity and suicidal thoughts in children.4 Concern has been raised about ADRs from long-term treatment with ADHD medications, such as psychosis, sensitization, dependency, and withdrawal reactions.1 The issue of appropriate warnings about possible ADRs to the use of methylphenidate and other ADHD medications is ever more important as usage continues to increase rapidly in many countries: an increase in the number of treated patients has been observed, as well as an increase in the average dispensed daily dose of psychostimulants.5

The use of psychostimulants, particularly methylphenidate, to treat ADHD symptoms in children has increased rapidly since the 1990s. Studies have shown that the prevalence of psychostimulant use in children in the Netherlands increased eight times from 1996 to 2006,6 and in Germany, prescription rates of methylphenidate increased by 96% from 2000 to 2007.7 From 1994 to 2004, the prevalence of psychostimulant use in Norwegian children increased five times,8 while the prevalence of stimulant medication increased ten times in American children from 1987 to 1996.9 Previous meta-analyses and reviews that evaluated the short-term efficacy of psychostimulants on ADHD symptoms in children concluded that psychostimulants are more effective than placebo with respect to treating disturbed attention and impulsivity.1,10 Several articles have reported information about the safety of methylphenidate and other psycho-stimulants in clinical studies,11 but to the current reviewers’ knowledge no articles have systematically reviewed the occurrence of ADRs following the use of ADHD medications in the pediatric population.

The objective of this study is to review published empirical studies on the occurrence of adverse drug reactions (ADRs) associated with the use of medications generally licensed for treatment of ADHD symptoms in the pediatric population.

Methods

Literature search

A literature search was performed in PubMed, Embase, and PsycINFO® (whole databases without language restriction) using the search terms “atomoxetine” (ATC group N06BA09), “methylphenidate” (ATC group N06BA04), “modafinil” (ATC group N06BA07), “amphetamine” (ATC group N06BA02), “psychostimulants,” and “nonstimulants” combined with any of the following: “adverse drug reaction,” “side effect,” and “adverse event.” Reference lists of identified articles were also screened for additional potentially relevant articles. For further details of the search strategy, please see Appendix 1. Literature searches were updated until September 2011.

Study selection

Using article titles as the selection basis, the first author retrieved and screened the abstracts to identify studies relevant to the study objective. Potentially relevant articles were retrieved in full text and screened for inclusion. To be considered relevant for this review, articles had to be peer reviewed and report ADRs in children in the age group 0–17 years of age associated with the use of psychostimulants.

Psychostimulants were specified as amphetamine derivates, methylphenidate, and modafinil, and nonstimulants as atomoxetine. Articles reporting ADRs from psychostimulants in mixed populations of children and adults were excluded if age-related ADR occurrence was not specified. Articles were excluded if they did not report data on ADR occurrence that made it possible to calculate rates. Hence, case reports, letters, commentaries, interim analyses, meta-analyses, and review articles were excluded. Further, articles reporting unintended events not classified as ADRs and articles on misuse were excluded, although reference lists of these studies were searched for relevant studies.

Data extraction

Data from included articles were extracted using a standard form, one for each article. The following information was recorded: authors, publication year, country, study design, dosage, comparator, monitoring period (weeks), size of study population, age and gender of included population, and ADR reporting rates in percentage. ADR reporting rates were indicated as reported in the original papers. In placebo-controlled studies, information about ADR reporting rates for placebo was also extracted. Information about who had assessed the ADRs, reported ADRs classified as being serious by the respective authors, and funding sources were also recorded. The first author extracted data, while the second author controlled and verified all cases.

Results

A total of 137 potentially relevant references were identified during the database searches and reference screenings. An overview of the review process and reasons for exclusion are displayed in Figure 1. Sixty-eight studies were excluded after screening abstracts. Sixty-nine studies were retrieved for full text review. Of these studies, four were later excluded as they reported mixed data on children and adults that could not be separated. Eight meta-analyses and four reviews of efficacy were excluded as they reported information from studies already included. Also excluded were two studies reporting data from a subgroup analysis of already included studies, and nine studies reporting ADRs as percent of children reporting an ADR.

Figure 1

Decision tree of the review process.

Note: *An overview of excluded studies is shown in Appendix 2.

Abbreviation: ADRs, adverse drug reactions.

Eventually 43 articles reporting ADRs from psycho-stimulants in the pediatric population were included. Table 1 displays an overview of the study characteristics of included articles. The majority of studies were conducted in the United States of America (USA), the remaining in Australia, Canada, Europe, Iran, and Latin America. Atomoxetine studies were published in the period from 2001 to 2009; amphetamine studies from 1997 to 2007; methylphenidate studies from 1997 to 2009; and modafinil studies from 2005 to 2009.

Table 1

Characteristics of included studies by country, design, study population, and funding

Studies (chronological order)	Country	Design	Setting	Dosage (mg/day)	Comparator	Treatment weeks (N)	Patients included (N)	Patients completed (N)	Age (y)	% Male	Type of assessor	Funding
Amphetamine
Biederman et al12	USA	R parallel	Naturalistic	30–70	Placebo	4	290	218	6–12	69	Parent	Industry
Spencer et al13	USA	R parallel	Naturalistic	10–40	Placebo	4	335	308	6–17	69	Parent	Industry
Wigal et al14	USA	R parallel	Laboratory	10–30	Atomoxetine	3	102	93	6–12	75	Parent	Industry
Biederman et al15	USA	R parallel	Naturalistic	10–30	Placebo	3	374	336	6–12	80	Parent	Industry
Efron et al16	AU	R crossover	Naturalistic	0.15 mg/kg	MPH	2	125	121	5–15	91	Teacher/Parent	Nonindustry
Total/Range						2–4	1226	1076	5–17	69–91
Methylphenidate
Arabgol et al17	IR	R parallel	Naturalistic	20–50	Reboxetine	6	16	12	7–16	66	Teacher/Parent	Nonindustry
Maayan et al20	USA	Open label	Naturalistic	10–30	NR	4	14	11	4–5	82	Self	NR
Amiri et al49	IR	R parallel	Naturalistic	20–30	Modafinil	6	32	30	6–15	80	Teacher/Parent	Nonindustry
Findling et al18	USA	R parallel	Naturalistic	10–54	Placebo	7	189	137	6–12	65	Self	Industry
Newcorn et al19	USA	R parallel	Naturalistic	18–54	Atomoxetine	6	220	180	6–16	71	Parent	Industry
Findling et al21	Various	R parallel	Naturalistic	10–60	Placebo	3	272	240	6–12	80	Self	Industry
Greenhill et al22	USA	R parallel	Naturalistic	5–30	Placebo	7	53	48	6–17	59	Self	Industry
McGough et al23	USA	R crossover	Laboratory	10–27	Placebo	5	42	41	6–12	73	Self	Industry
Gau et al24	TW	R open label	Naturalistic	10–40	None	4	64	64	6–15	91	Self	Industry
Silva et al25	USA	R crossover	Laboratory	20–40	Placebo	<1	54	53	6–12	70	Parent	Industry
Kemner et al26	USA	R open label	Naturalistic	18–72	Atomoxetine	3	891	850	6–12	74	Parent	Industry
Swanson et al27	USA	R crossover	Laboratory	18–60	Placebo	NR	184	181	6–12	74	Self/Parent	Industry
Biederman et al28	USA/CA	R parallel	Naturalistic	10–40	Placebo	2	65	61	6–14	80	Parent	Industry
Kratochvil et al29	USA/CA	R open label	Naturalistic	5–60	None	10	44	25	7–15	100	Parent	Industry
Pelham et al30	USA	R crossover	Lab/Nat	5–54	Placebo	3	70	68	6–12	NR	Teacher/Parent	Industry
Efron et al16	AU	R crossover	Naturalistic	0.3 mg/kg	Amphetamine	2	125	121	5–15	91	Teacher/Parent	Nonindustry
Total/Range						1–10	2303	2092	4–17	59–100
Atomoxetine
Svanborg et al31	SE	R parallel	Naturalistic	80	Placebo	10	49	49	7–15	80	Self	Industry
Block et al32	USA	R parallel	Naturalistic	0.47–1.81 mg/kg	Placebo	6	288	140	6–12	73	Parent	Industry
Tamayo et al33	Various	Open label	Naturalistic	35–120	None	10–11	1198	947	6–17	76	Self	Industry
Newcorn et al19	USA	R parallel	Naturalistic	0.8–1.8 mg/kg	MPH/Placebo	6	222	186	6–16	78	Parent	Industry
Bangs et al34	US	R parallel	Naturalistic	1.2–1.8 mg/kg	Placebo	9	72	71	12–17	72	Self	Industry
Geller et al35	USA	R parallel	Naturalistic	0.8–1.8 mg/kg	Placebo	12	87	66	8–17	62	Self	Industry
Gau et al36	TW	R parallel	Naturalistic	16–99	Placebo	6	72	72	6–16	90	Parent	Industry
Kratochvil et al37	USA	Open label	Naturalistic	0.5–1.8 mg/kg	None	8	22	20	5–6	86	Parent	Industry
Prasad et al38	UK	R open label	Naturalistic	0.5–1.8 mg/kg	SCT	10	104	78	7–15	89	Self	Industry
Arnold et al39	USA	R cross over	Naturalistic	2.5–40	Placebo	12	16	15	5–15	75	Self	Industry
Newcorn et al40	USA	R parallel	Naturalistic	1.2–1.8 mg/kg	None	32	229	160	6–16	72	Parent	Industry
Wigal et al14	USA	R parallel	Laboratory	10–60	Amphetamine	3	101	97	6–12	76	Parent	Industry
Allen et al42	USA	R parallel	Naturalistic	0.5–1.5 mg/kg	Placebo	18	76	74	7–17	92	Self	Industry
Kemner et al26	USA	R open label	Naturalistic	10–80	MPH	3	499	473	6–12	74	Parent	Industry
Escobar et al41	ES	Open label	Naturalistic	0.5–1.8 mg/kg	None	10	36	36	6–15	89	Parent	Industry
Kelsey et al43	USA	R parallel	Naturalistic	0.8–1.2 mg/kg	Placebo	8	133	107	6–12	71	Parent	Industry
Biederman et al44	USA	R parallel	Naturalistic	2 mg/kg	Placebo	9	31	31	7–13	0	Self	Industry
Michelson et al45	USA	R parallel	Naturalistic	0.5–1.0 mg/kg	Placebo	6	85	84	6–16	71	Parent	Industry
Kratochvil et al29	USA/CA	R open label	Naturalistic	0.2–2.0 mg/kg	None	10	184	118	7–15	91	Parent	Industry
Spencer et al46	USA	R parallel	Naturalistic	90	Placebo	12	129	127	7–13	76	Parent	Industry
Michelson et al47	USA	R parallel	Naturalistic	0.5–1.8 mg/kg	Placebo	8	213	176	8–17	71	Self	Industry
Total/Range						3–32	3846	3127	5–17	0–91
Modafinil
Kahbazi et al48	IR	R parallel	Naturalistic	200–300	Placebo	6	24	23	6–15	76	Teacher/Parent	Nonindustry
Amiri et al49	IR	R parallel	Naturalistic	200–300	MPH	6	32	30	6–15	78	Teacher/Parent	Nonindustry
Boellner et al50	USA	Open label	Naturalistic	100–400	None	8	220	166	6–14	72	Parent	Industry
Wigal et al51	USA	R parallel	Naturalistic	170–425	Placebo	9	423	411	10.2	72	Parent/Self	Industry
Biederman et al52	USA	R parallel	Naturalistic	300–400	Placebo	4	197	175	6–14	75	Parent	Industry
Greenhill et al53	USA	R parallel	Naturalistic	170–425	Placebo	9	133	100	6–16	73	Parent/Self	Industry
Biederman et al54	USA	R parallel	Naturalistic	170–425	Placebo	9	164	97	6–17	69	Parent	Industry
Total/Range						4–9	1137	949	6–17	69–75
Total all studies							8512	7244	–	–

Abbreviations: AU, Australia; CA, Canada; IR, Iran; lab, laboratory; MPH, methylphenidate; nat, naturalistic; NR, not reported; R, randomized; ES, Spain; SCT, standard current therapy; SE, Sweden; TW, Taiwan; USA, United States of America.

Design and setting

Information about ADRs was reported in clinical studies using different designs, ie, randomized parallel group studies (N = 28);12–15,17–19,22,23,28,31,32,34–36,40,42–47,48–54 randomized crossover studies (N = 6);16,23,25,27,30,39 and open-label designs (N = 9).20,26,29,33,37,38,41,50 The majority of studies were conducted in naturalistic settings at home and at school (N = 38);12,13,15–22,24,26,28–30,31–43,44–54 five articles reported ADRs from children participating in laboratory school protocols,14,23,25,27,30 in which classroom sessions were organized in cycles to include 12 hours of observation. This design consisted of daily schedules of alternating classroom, meals/snacks, recess, and research activities scheduled at specific times during the day. The largest number of studies (N = 21)31–47 concerned atomoxetine; followed by methylphenidate (N = 14;17–30 modafinil (N = 7);48–54 and amphetamine (N = 5).12–16

Dosage and comparator

The tested dosages varied from 10 to 70 mg/day in amphetamine studies; from 5 to 72 mg/day in methylphenidate studies; from 10 to 90 mg/day in atomoxetine studies; and from 100 to 425 mg/day in modafinil studies. Placebo was used as a comparator drug in the majority of studies (N = 28), while an active comparator, was administered in nine studies. Seven open-label studies did not include a control group.

Treatment period

Treatment duration varied from 1 to 32 weeks across studies. Treatment duration varied from 2 to 4 weeks in amphetamine studies; from 1 to 10 weeks in methylphenidate studies; from 3 to 32 weeks in atomoxetine studies; and from 4 to 9 weeks in modafinil studies.

Population

A total of 8512 children were included in the clinical studies, of which 7244 children completed treatment: amphetamine (1076); methylphenidate (2092); atomoxetine (3127); and modafinil (949). The reasons for noncompletion were many, but lack of efficacy and the appearance of ADRs were the most common. The ages of the included children varied from 4 to 17 years (median 6–12 years), and the share of male patients in the studies varied from 0 to 100% (median 69%).

Type of assessor

Parents rated information about ADRs in 20 studies,12–15,19,25–28, 32,35–36,40–46,48,50,52 54 patients in 15 studies,18,20–24,31,33–35,38–39,42,44,47 and a combination of teacher/parent (five studies),16–17,30,48–49 and patient/parent (three studies).27,49,51 The articles specified only limited information about applied ADR scales and the classification systems used.

Funding source

In almost all studies the funding source was the manufacturer of the respective medications, and only four studies were publicly funded. Additionally, a large number of the studies were conducted by the same groups of authors who declared conflicts of interest. The majority of the authors received contributions from the pharmaceutical companies producing the medications in return for activities, such as providing scientific advice and making oral and poster presentations at scientific meetings.

ADRs by type and occurrence

Tables 2–5 display the ADR reporting rates listed in the included studies for each type of psychostimulant. ADRs of similar type and wording were aggregated in a common category in order to clarify data presentation. The aggregated categories were: weight changes (changes in weight, weight decreased, weight increased, decrease in weight); gastrointestinal pain (abdominal pain, upper abdominal pain, gastrointestinal pain); anxiety (anxiety, anxiousness); influenza (influenza, flu syndrome); tics (tics, motor tics, facial tics); blood pressure changes (diastolic blood pressure, changes in blood pressure); sleeping problems (awake during the night, difficulty falling asleep, sleep disturbance, delayed onset of sleep); changes in heart rate (racing heart, changes in heart rate). Thirty-one categories of ADRs were reported for amphetamine derivates (Table 2); 65 categories for methylphenidate formulations (Table 3); 55 categories for atomoxetine (Table 4); and 38 categories for modafinil (Table 5). The following ADRs were most frequently reported for all four psychostimulants: decrease in appetite, gastrointestinal pain, and headache.

Table 2

Adverse drug reaction reporting rates (%) for amphetamine derivates by category and study

Reference number	12	13	14	15	16	Range	Placebo (range)
Adverse drug reaction
Accidental injury	–	5	–	–	–	5	5
Anorexia	–	25	17	22	–	17–25	2
Anxiety	–	–	–	–	68	68	–
Appetite decrease	39	–	28	–	59	28–59	4–5
Cough	1	–	–	5	–	1–5	5
Crying	–	–	–	–	76	76	–
Daydreams	–	–	–	–	62	62	–
Dizziness	5	–	6	–	32	5–32	–
Dry mouth	5	–	–	–	–	5	–
Emotional disturbance	–	–	–	–	59	59	–
Emotional lability	–	5	–	9	–	5–9	2
Fatigue	–	–	2	–	–	2	–
Fingernail biting	–	–	–	–	40	40	–
Gastrointestinal pain	12	11	19	14	–	11–19	5–10
Headache	12	19	15	18	30	12–30	10–21
Insomnia	19	20	28	17	–	17–28	2–8
Irritability	10	–	7	–	82	7–82	–
Nasal congestion	1	–	–	–	–	1	–
Nasopharyngitis	5	–	–	–	–	5	–
Nausea	6	–	7	5	–	5–7	3
Nervousness	–	6	–	6	–	6	2
Nightmares	–	–	–	–	28	28	–
Pharyngitis	–	7	–	7	–	7	5–20
Sleeping problems	–	–	–	–	70	70	–
Social withdrawal	–	–	–	–	64	64	–
Somnolence	–	–	5	–	–	5	–
Stomachache	–	–	–	–	40	40	–
Tics	–	–	–	–	26	26	–
Unusually happy	–	–	–	–	26	26	–
Vomiting	9	–	5	7	–	5–9	4
Weight changes	9	8	6	–	–	6–9	1

Table 5

Adverse drug reaction reporting rates (%) for modafinil by category and study

Reference number	48	49	50	51	52	53	54	Range	Placebo (range)
Adverse drug reaction
Abnormal behavior	–	–	–	–	–	–	–	–	4
Accidental injury	–	–	5	–	–	5	5	5	3–6
Anorexia	–	–	–	–	–	–	–	–	1–2
Anxiety	–	–	–	–	–	–	–	–	1–12
Appetite decrease	35	25	6	16	7	18	16	6–35	2–12
Asthenia	–	–	–	–	–	–	4	4	5
Cough	–	–	5	8	7	9	8	5–9	4–9
Diarrhea	–	–	–	–	–	–	–	–	1–2
Dry mouth	14	14	–	–	–	–	–	14	18
Dyspepsia	–	–	–	–	–	–	–	–	4
Emotional disturbance	–	–	–	–	–	–	–	–	12
Emotional lability	8	4	5	–	–	5	–	5–8	2–6
Fatigue	–	–	–	–	–	–	–	–	4
Fever	–	–	–	5	6	5	5	5–6	2–7
Gastroenteritis	–	–	–	–	–	5	–	5	4
Gastrointestinal pain	8	7	–	10	9	12	7	7–12	2–13
Headache	8	7	10	20	13	22	20	7–22	3–23
Infection	–	–	5	11	6	11	12	5–12	1–15
Influenza	–	–	–	–	–	–	–	–	9
Insomnia	–	–	13	27	12	28	29	12–29	2–10
Irritability	8	7	–	–	–	–	–	7–8	2–6
Nasal congestion	–	–	–	–	–	–	–	–	1
Nasopharyngitis	–	–	–	–	–	–	–	–	2–7
Nausea	4	7	–	–	–	5	–	4–7	2–12
Nervousness	6	7	–	5	–	5	4	4–7	2–6
Otitis media	–	–	–	–	–	–	–	–	2
Pain	–	–	–	–	–	–	5	5	1
Pharyngeal pain	–	–	–	–	–	–	–	–	1–7
Pharyngitis	–	–	–	7	–	8	9	7–9	3–13
Rash	–	–	–	–	–	–	6	6	4
Respiratory tract infection									2–6
Rhinitis	–	–	5	7	5	8	10	5–10	4–11
Sleeping problems	4	14	–	–	–	–	–	4–14	12
Somnolence	–	–	–	–	–	5	2	2–5	4–5
Tics	–	–	–	–	–	–	–	–	4
Urinary incontinence	–	–	–	–	–	–	–	–	3
Vomiting	–	–	–	5	–	6	6	5–6	2–9
Weight changes	6	7	–	–	–	5	–	5–7	1–6

Table 3

Adverse drug reaction reporting rates (%) for methylphendiate by category and study

Reference number	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	49	Range	Placebo (range)
Adverse drug reaction
Abnormal behavior	–	–	–	–	–	3	–	–	–	–	1	–	–	5	–	–	1–5	4
Accidental injury	–	–	–	–	–	–	–	–	–	–	–	3	–	13	3	–	3–13	3
Affect lability	–	–	3	–	–	–	4	–	–	–	–	–	–	–	–	–	3–4	–
Aggression	–	–	–	–	–	–	–	–	–	–	1	–	–	–	–	–	1	–
Anorexia	–	–	3	–	–	5	4	3	–	8	–	3	3	15	–	–	3–15	1–2
Anxiety	61	–	–	–	–	–	–	–	25	–	–	–	–	–	–	5	5–61	1
Appetite decrease	56	31	19	17	28	3	30	–	53	9	6	3	–	–	–	31	3–56	5–9
Asthenia	–	–	–	–	–	–	–	–	–	–	–	–	–	3	–	–	3	–
Blood pressure changes	–	–	–	18	–	–	–	3	–	–	–	–	–	–	–	–	3–18	–
Changes in heart rate	–	–	–	12	–	–	–	–	–	–	–	–	–	–	–	–	12	–
Changes in pulse rate	–	–	–	–	–	–	–	–	–	1	–	–	–	–	–	–	1	–
Chest pain	–	6	–	–	–	–	–	–	–	–	–	–	–	–	–	–	6	–
Cough	–	–	–	4	–	2	–	–	–	–	–	–	–	5	–	–	2–5	4
Crying	71	–	–	–	–	–	–	–	38	–	2	–	–	–	–	–	2–71	–
Daydreams	62	–	–	–	–	–	–	–	30	–	–	–	–	–	–	–	30–62	–
Depression	–	–	–	–	–	–	–	–	–	–	–	–	–	5	–	–	5	–
Diarrhea	–	–	–	–	–	–	4	–	–	–	–	–	–	3	2	–	2–4	1–2
Dizziness	30	13	–	–	–	–	–	–	13	–	1	–	–	–	2	–	1–30	4
Dry mouth	–	–	–	–	–	–	–	–	24	–	–	–	–	–	–	12	12–24	–
Dyspepsia	–	–	–	–	–	–	8	–	–	–	–	–	–	5	–	–	5–8	–
Emotional disturbance	56	–	–	–	–	–	–	–	–	–	1	–	–	–	–	–	1–56	–
Emotional lability	–	–	–	–	8	–	–	–	–	–	–	–	–	5	–	7	5–8	–
Euphoria	–	–	–	–	–	–	–	–	9	–	–	–	–	–	–	–	9	–
Eye redness	–	–	–	–	4	–	–	–	–	–	–	–	–	–	–	–	4	–
Eye twitching	–	–	–	–	4	–	–	–	–	–	–	–	–	–	–	–	4	–
Fatigue	–	–	–	2	4	–	4	–	–	4	1	–	–	–	–	–	1–4	4
Fever	–	–	–	–	4	2	–	–	–	–	–	–	–	10	–	–	2–10	7
Fingerrnail biting	45	–	–	–	–	–	–	–	22	–	–	–	–	–	–	–	22–45	–
Gastroenteritis	–	–	–	–	–	–	4	–	–	–	–	5	–	–	–	–	4–5	4
Gastrointestinal pain	–	–	–	10	12	10	19	–	–	6	4	4	–	18	15	8	4–19	2–13
Headache	24	–	–	11	4	16	25	4	28	2	4	3	2	33	14	8	2–33	3–23
Hyperkinesia	–	–	–	–	–	–	–	–	–	–	–	–	–	5	–	–	5	–
Increases in ALT/AST	–	–	–	–	–	–	–	–	–	–	–	5	–	–	–	–	5	–
Infection	–	–	–	–	4	–	–	–	–	–	–	2	–	8	–	–	2–8	1–4
Influenza	–	–	–	–	–	–	4	–	–	–	–	–	–	10	–	–	4–10	–
Insomnia	–	19	8	27	–	4	8	–	44	4	7	2	3	18	–	–	2–44	3–10
Irritability	80	6	–	6	4	3	4	–	16	–	1	1	–	–	–	7	1–80	2–6
Lymphadenopathy	–	–	–	–	–	–	–	3	–	–	–	–	–	–	–	–	3	–
Mood alteration	–	–	–	–	–	–	–	–	34	–	1	7	–	–	–	–	1–34	–
Nasal congestion	–	–	3	–	4	–	–	–	–	–	–	–	–	–	–	–	3–4	1
Nasopharyngitis	–	–	4	–	–	4	9	1	–	–	–	–	–	–	–	–	1–9	2–7
Nausea	–	–	8	6	–	–	11	4	–	–	1	–	–	5	–	5	1–11	2–6
Nervousness	–	–	–	–	–	–	–	–	–	–	–	4	–	10	–	–	4–10	–
Nightmare	21	–	–	–	–	–	–	–	16	–	–	–	–	–	–	–	16–21	–
Otitis media	–	–	–	–	–	–	4	–	–	–	–	–	–	–	–	–	4	2
Pain	–	–	–	–	–	–	–	–	–	–	–	–	–	3	–	–	3	–
Pallor	–	13	–	–	–	–	–	–	–	–	–	–	–	–	–	–	13	–
Palpitation	–	–	–	–	–	–	–	–	–	–	–	–	–	5	–	–	5	–
Pharyngeal pain	–	–	–	–	–	–	–	3	–	–	–	–	–	–	–	–	3	–
Pharyngitis	–	–	–	–	–	3	–	–	–	–	–	–	–	8	2	–	2–8	3
Rash	–	–	–	–	–	–	–	1	–	–	–	–	–	8	–	–	1–8	4
Respiratory tract infection	–	–	–	–	–	3	9	–	–	–	–	–	–	–	4	–	3–9	2–6
Rhinitis	–	–	–	–	–	–	–	3	–	–	–	–	–	20	2	–	2–20	–
Sensitivity	–	–	–	–	4	–	–	–	–	–	–	–	–	–	–	–	4	–
Sleeping problems	64	–	–	–	12	–	–	–	–	–	–	–	–	–	–	9	9–64	–
Socially withdrawn	59	–	–	–	–	–	–	–	27	–	–	–	–	–	–	–	27–59	–
Somnolence	–	–	–	2	–	–	–	–	–	–	1	–	–	–	–	–	1–2	–
Stomachache	32	–	–	–	–	–	4	–	28	–	–	–	–	–	–	–	4–32	–
Tachycardia	–	–	–	–	–	–	–	–	–	–	–	–	–	5	–	–	5	–
Tics	28	–	1	–	–	–	–	–	13	–	–	–	–	–	–	–	1–28	4
Twitching	–	–	–	–	–	–	–	–	–	–	–	–	–	–	3	–	3	–
Unusually happy	28	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	28	–
Urinary incontinence	–	–	–	–	–	–	–	–	–	–	–	–	–	–	1	–	1	3
Vomiting	–	–	10	4	4	3	4	–	–	–	1	1	–	–	3	–	1–10	2–5
Weight changes	–	–	8	1	–	–	–	–	–	–	–	–	–	5	–	8	1–8	4

Abbreviation: AST/ALT, aspartate aminotransferase/alanine aminotransferase.

Table 4

Adverse drug reaction reporting rates (%) for atomoxetine by category and study

Reference number	14	19	26	29	31	32	33	34	35	36	37	38	39	40	41	42	43	44	45	46	47	Range	Placebo (range)
Adverse drug reaction
Abnormal behavior	–	–	2	–	–	–	–	–	–	–	19	–	–	–	–	–	–	–	–	–	–	2–19	–
Accidental injury	–	–	–	5	–	–	–	–	–	–	–	–	–	–	–	–	–	–	6	–	3	3–6	5–8
Aggression	–	–	1	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	1	–
Anorexia	9	–	–	19	35	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	9–35	5
Appetite decrease	18	14	3	–	6	11	10	13	14	36	50	8	38	15	31	16	18	19	20	22	–	3–50	3–25
Asthenia	–	–	–	8	–	–	–	–	–	10	–	–	–	–	–	–	–	–	11	–	5	5–11	1–5
Blunted effect	–	–	–	–	–	–	–	–	–	–	5	–	–	–	–	–	–	–	–	–	–	5	–
Blood pressure changes	–	18	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	18	16
Changes in heart rate	–	11	–	–	–	–	–	–	–	–	–	–	19	–	–	–	–	–	–	–	–	11–19	12
Constipation	–	–	–	–	–	–	4	–	–	–	–	–	6	–	–	–	–	–	–	–	–	4–6	6
Cough	–	3	–	5	–	–	7	–	5	13	–	6	–	5	–	5	–	16	7	13	10	3–16	5–21
Crying	–	–	1	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	1	–
Depression	–	–	–	3	10	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	3–10	4–6
Diarrhea	–	–	–	7	–	–	–	1	–	6	5	–	6	–	–	4	2	3	–	–	5	1–7	5–13
Dilated pupils	–	–	–	–	–	–	–	–	–	–	5	–	–	–	–	–	–	–	–	–	–	5	–
Dizziness	2	–	2	–	–	–	6	13	–	10	–	–	–	–	–	–	–	3	6	–	5	2–13	1–5
Dry mouth	–	–	–	–	–	–	–	–	–	–	–	–	6	–	–	–	–	–	–	–	–	6	6
Dyspepsia	–	–	–	5	–	–	–	–	–	–	–	–	–	–	–	–	6	–	9	–	–	5–9	–
Emotional disturbance	–	–	1	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	1	–
Emotional lability	–	–	–	6	–	–	–	–	–	–	55	–	–	8	–	–	–	3	7	–	–	3–55	5–14
Fatigue	7	5	3	–	33	–	4	13	–	–	5	–	–	5	–	12	10	–	–	–	–	3–33	1–18
Fever/pyrexia	–	–	–	11	4	–	5	3	–	4	–	–	–	10	–	–	–	10	7	–	5	3–11	4–15
Gastroenteritis	–	–	–	–	–	–	–	–	–	–	–	–	31	–	–	–	–	–	–	–	–	31	–
Gastrointestinal pain	15	11	5	23	47	13	7	8	12	8	27	7	–	12	–	9	15	29	17	31	13	5–47	4–22
Headache	10	18	4	31	39	9	17	17	14	10	14	21	13	21	17	21	7	26	20	30	24	4–39	4–28
Hyperkinesia	–	–	–	2	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	2	–
Infection	–	–	–	4	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	4	4	1
Influenza	–	–	–	5	–	–	–	4	5	–	–	–	–	5	–	–	–	–	–	–	–	4–5	1–6
Insomnia	7	7	3	9	–	–	5	–	–	11	–	–	19	–	–	–	–	3	–	–	7	3–19	1–13
Irritability	4	6	2	–	12	–	–	6	7	–	–	–	–	10	–	–	–	–	–	–	–	2–12	1–4
Mood alteration	–	–	–	–	–	–	–	–	–	–	–	–	44	–	–	–	–	–	–	–	–	44	31
Nasal congestion	–	–	–	–	–	–	–	–	–	–	–	–	–	6	–	–	–	–	–	–	–	6	–
Nasopharyngitis	–	–	–	–	–	4	5	–	7	–	–	–	–	14	–	–	–	–	–	–	–	4–14	6–8
Nausea	9	4	5	10	29	6	9	22	7	17	5	17	31	5	–	16	12	7	12	10	6	4–31	1–14
Nervousness	–	–	–	16	–	–	–	–	–	–	–	–	–	–	–	–	–	7	–	13	6	6–16	5–7
Oppositional behavior	–	–	–	–	–	–	–	–	–	–	–	–	–	–	11	–	–	–	–	–	–	11	–
Pain	–	–	–	5	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	6	5–6	6
Palpitation	–	–	–	2	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	2	–
Pharyngeal pain	–	–	–	–	–	–	–	–	–	–	–	–	–	10	–	–	–	–	–	–	–	10	–
Pharyngitis	–	–	–	6	–	–	–	–	–	7	–	–	–	5	17	4	–	19	7	16	10	4–19	9–19
Pruritus	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	4	4	–
Rash	–	–	–	4	–	–	–	–	–	–	–	–	13	–	–	–	–	–	7	–	7	4–13	4–6
Respiratory tract infection	–	–	–	–	10	3	–	–	–	–	–	–	–	8	–	–	–	–	–	–	–	3–10	4
Restlessness	–	–	–	–	–	–	–	–	–	–	–	–	13	–	–	–	–	–	–	–	–	13	19–21
Rhinitis	–	–	–	18	–	–	–	–	–	8	–	–	–	–	–	–	–	26	17	26	14	8–26	15–38
Sinusitis	–	–	–	–	–	–	–	–	5	–	–	–	–	–	–	–	–	–	–	–	–	5	4
Somnolence	19	6	4	11	–	8	–	–	–	22	36	–	–	5	–	–	15	7	11	9	8	4–36	1–14
Stomachache	–	–	–	–	–	5	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	5	–
Tachycardia	–	–	–	6	–	–	–	–	–	–	–	–	–	–	14	–	–	–	–	–	–	6–14	–
Thirst	–	–	–	–	–	–	–	–	–	–	14	–	–	–	–	–	–	–	–	–	–	14	–
Throat pain	–	–	–	–	–	–	3	–	–	–	–	–	–	–	–	–	–	–	–	–	–	3	–
Tics	–	–	–	–	–	–	–	–	–	–	5	–	–	–	–	–	–	–	–	–	–	5	7
Tired	–	–	–	–	–	–	–	–	–	–	–	–	31	–	–	–	–	–	–	–	–	31	4–13
Vomiting	13	7	2	12	12	12	7	13	10	7	14	9	–	8	11	16	6	19	15	15	8	2–19	1–12
Weight changes	4	1	–	3	–	–	–	9	–	6	–	8	–	–	–	–	–	–	–	–	–	1–9	1–6

ADRs by seriousness

The majority of reported ADRs were categorized by the authors/investigators as nonserious. Table 6 shows information about the categories of serious ADRs reported in the clinical studies. Serious cases included aggression (amphetamine, methylphenidate);16 anxiety (amphetamine);16 emotional disturbances (amphetamine);14,15 insomnia (amphetamine, modafinil);13,15,37 and attempted suicide (amphetamine).13

Table 6

Serious ADRs reported for ADHD medications in identified studies

Medication (alphabetically)	Reference	Adverse drug reaction(s)
Amphetamine	Spencer et al13	Arthrosis
		Hyperkinesia
		Insomnia
		Nervousness
		Pharyngitis
		Suicide attempt
	Wigal et al14	Emotional disturbance
		Headache
	Biedermann et al15	Anorexia
		Emotional lability
		Insomnia
	Efron et al16	Agitation
		Aggression
		Anxiety
Methylphenidate	Efron et al16	Aggression
		Headache
		Tearful
	Greenhill et al22	Hypersomnia
	Kemner et al26	Mania
	Biederman et al28	Depression
Atomoxetine	Wigal et al14	Upper abdominal pain
	Arnold et al39	Aggression
Modafinil	Boellner et al50	Insomnia
	Biederman et al52	Insomnia
	Wigal et al51	Asthma
		Dehydration
		Duodenitis
		Erythema multiforme
		Hypertonia
		Influenza syndrome
		Peptic ulcer
		Stevens–Johnson syndrome

Abbreviations: ADRs, adverse drug reactions; ADHD, attention deficit hyperactivity disorder.

Discussion

This is the first study to systematically review the empirical literature on the occurrence of ADRs reported for ADHD medications in the pediatric population. Information about ADRs from psychostimulants and the nonstimulant atomoxetine was reported in clinical studies of short duration, primarily conducted in 6- to 12-year-old boys, and particularly in the USA. The most frequently reported ADRs were decrease in appetite, gastrointestinal pain, and headache. A large number of studies were conducted by the same groups of authors and sponsored by the pharmaceutical companies manufacturing the respective medications.

Although the review process found a large number of small clinical trials exploring the efficacy of ADHD medications in the pediatic population, only a minor share of these studies reported information about ADRs. The studies included in this article were similar in design and setting, treatment duration, as well as number, age, and gender of included patients. The reliability of the studies may be questioned as the number of reported ADRs varied widely for identical and similar study designs. Further exploration of these questions would require access to the original study material. Large variations in ADR reporting rates were observed between studies and therapeutic groups, and similar types of ADRs were reported for the individual ADHD medications. It is puzzling that large numbers of specific ADRs are reported in some studies, but few if any in others. These findings question the relevance of the many small clinical trials conducted on the medications, particularly atomoxetine and methylphenidate, as they are not designed to measure long-term efficacy and safety.55 Almost all of these clinical trials were sponsored by the pharmaceutical companies producing the subject medications, and therefore, the current reviewers encourage these companies to make information about the ADRs reported in said clinical trials accessible to the public.

Seriousness of reported ADRs

Only a small number of serious ADRs were reported. However, in several of the included studies a large number of children withdrew due to experiencing ADRs, and therefore, the actual number of serious ADRs occurring from the use of ADHD medications might be higher, and some types of ADRs may not have been reported. Information about ADR incidence in the monitored population was only reported if the incidence was above 2% and/or 5%; consequently, information about rarely occurring ADRs is not included. Another issue is that information about definitions and scales to define and evaluate events occurring during the clinical trials is not reported in the articles, thus making it impossible to react to this information. Therefore, the regulatory agencies are encouraged to allow access to the original clinical protocols, so that all information reported for ADHD medications can be made public. A previous study has shown that there are large discrepancies between the data reported in clinical trial protocols and data published in scientific journals.56

Long-term safety aspects of psychostimulant use

Psychostimulants and other ADHD medications are prescribed for long-term treatment in large populations and there is a need for long-term efficacy and safety studies.1 The lack of sufficient knowledge about ADRs at the point of licensing of new medicines makes spontaneous ADR reporting an important source of information about medicine safety.57 As clinical trials in the pediatric population are limited, clinicians and health authorities must rely on spontaneous reports as the main source of information about previously unknown ADRs.57 However, the current review did not find any studies about ADRs from the use of psychostimulants reported to any national ADR databases. Systematic analyses of ADRs reported to national databases are necessary, as these databases constitute a critical (and underestimated) source of important data, especially information about new, serious, and rarely occurring ADRs. Further studies of data from large databases, ie, the World Health Organization/Uppsala Monitoring Centre VigiBase™ (Uppsala, Sweden) or the European Medicines Agency EudraVigilance (London, United Kingdom [UK]) databases, are recommended in order to increase knowledge about ADRs from the use of ADHD medications.

Strengths and limitations of this review

The included studies were conducted over a period of approximately 20 years in different countries, with a great deal of inconsistency in observing and classifying the type and seriousness of reported ADRs. Information about the seriousness of the reported ADRs was extracted from the included studies, and it was not possible for the review to evaluate these ratings, nor to estimate ADRs in terms of effect sizes, as the review did not have access to the original data material. A major limitation of this study is that it is unknown to what extent the causality of these ADRs can be confirmed, and this has implications for the interpretation of the findings in the review. A large number of published clinical studies were not included in this review because these articles did not report information about ADRs, despite the fact that pharmaceutical companies had a legal obligation to monitor ADRs in clinical trials, and therefore, these data must exist. As the clinical trials were mainly sponsored by the pharmaceutical companies that produce the medications, these companies are urged to make these data accessible to the public.

Conclusion

Reported ADRs from the use of psychostimulants in the pediatric population were generally found in clinical trials of short duration. Since ADHD medications are prescribed for long-term treatment there is a need for long-term safety studies. Considering the widespread and increasing use of these medications in children, greater care must be taken when prescribing these medications for long-term use. Further studies of spontaneous reports submitted to national and international databases are recommended in order to increase knowledge about ADRs from the use of psychostimulants in the pediatric population. Pharmaceutical companies should make all information about ADRs reported for ADHD medications accessible to the public. Additionally, the impact of the link between researchers and the manufacturers of the medications needs to be studied.

Embase

Adverse event

Methylphenidate

Adverse drug reaction AND methylphenidate

(Adverse event OR adverse drug reaction) AND psychostimulant

Atomoxetine OR modafinil OR methylphenidate OR amphetamine

(Atomoxetine OR modafinil OR methylphenidate OR amphetamine)

AND adverse event

PubMed

Adverse event

Methylphenidate

Adverse event AND methylphenidate

Adverse event OR adverse drug reaction

Psychostimulant

(Adverse event OR adverse effect) AND psychostimulant

Atomoxetine OR modafinil OR methylphenidate OR amphetamine

(Atomoxetine OR modafinil OR methylphenidate OR amphetamine)

AND adverse event

PsycINFO®

Adverse event

Methylphenidate

Adverse event AND methylphenidate

Side effect

Adverse drug reaction

Psychostimulant

(Adverse drug reaction OR side effect) AND psychostimulant

Adverse drug reaction OR side effect OR adverse event

Atomoxetine OR modafinil OR methylphenidate OR amphetamine

(Adverse event OR side effect OR adverse drug reaction) AND

(Atomoxetine OR modafinil OR methylphenidate OR amphetamine)