Effects of attention-deficit hyperactivity disorder on growth in children and adolescents: A systematic review and meta-analysis

Abstract

BACKGROUND

Attention-deficit hyperactivity disorder (ADHD) and its pharmacological treatments may influence growth in children and adolescents. This meta-analysis aimed to clarify their effects on the physical development, especially weight and height.

AIM

To investigate the effects of ADHD and its treatment on growth in children and adolescents.

METHODS

Researchers reviewed 18 studies published up to September 2023 from databases such as PubMed, EMBASE, Cochrane, and Web of Science. They analyzed changes in body weight, height, and body mass index (BMI) before and after ADHD treatment, along with the risks of overweight and obesity.

RESULTS

Children with ADHD undergoing long-term medication therapy showed decreased actual weight [weighted mean difference (WMD) = -9.50] and height (WMD = -0.15), along with a slight increase in weight standard deviation scores (WMD = 0.23) and height z scores (WMD = 0.10). BMI showed a non-significant downward trend (WMD = -1.72). Regarding overweight and obesity risks, the pooled odds ratios were 1.37 and 1.16, respectively, but these were not statistically significant.

CONCLUSION

Overall, the study suggests that long-term pharmacological treatment for ADHD may be associated with reduced growth in weight and height among young patients. However, no clear link was found between ADHD and increased risk of overweight or obesity. These findings highlight the importance of monitoring growth in children receiving medication for ADHD.

Key Words: Attention deficit hyperactivity disorder; Body mass index; Body weight; Height; Medication; Meta-analysis

Core Tip: This meta-analysis of 18 studies explored how long-term use of medication for attention-deficit hyperactivity disorder (ADHD) affects growth in children and adolescents. Results show that medication use is linked to decreases in weight and height, with slight increases in weight and height z scores, while body mass index showed a non-significant downward trend. No significant increase in overweight or obesity risk was observed. These findings underscore the need for careful growth monitoring in children on ADHD medication, as long-term treatment may impair physical development without substantially affecting obesity risk. The study highlights important considerations for clinicians managing young patients with ADHD.

INTRODUCTION

Attention-deficit hyperactivity disorder (ADHD) is a neurodevelopmental dysfunction arising from a complex interplay between genetics and environmental factors, impairing executive functions essential for planning and cognitive flexibility[1,2]. Characterized by inattention, hyperactivity, and impulsivity[3], ADHD is commonly prevalent among children and adolescents worldwide. A meta-analysis of 171756 subjects from major regions around the world estimated a global ADHD prevalence of 5.3% in this population[4]. Diagnosis primarily depends on clinical symptoms reported by patients, relatives, or teachers, with a detailed clinical interview serving as the diagnostic gold standard[2].

Pharmacological therapy, alone or combined with behavioral interventions, is effective in managing ADHD symptoms, especially when structured drug management is applied[5]. Both stimulant medications, such as amphetamine (AMP) and methylphenidate (MPH), and non-stimulants, such as atomoxetine (ATX), guanfacine, and clonidine, are commonly used, with stimulants generally being more effective[6,7]. Consequently, stimulants are the preferred first-line treatment for patients with ADHD across all age groups[8-10]. The side effects of ADHD medications are typically mild and transient, and severe adverse events remain rare[11]. Nonetheless, there is an ongoing discourse regarding the potential long-term side effects, which may include stunted growth and reduced weight gain[12]. For example, during the first 3 years of stimulant therapy, children diagnosed with ADHD exhibited an annual height deficit of 1 cm[13]. In a 16-year follow-up study of patients who had received stimulant therapy during childhood, those who received treatment for more than 50% of the days were found to be 4.1 cm shorter than their counterparts who received treatment for less than 50% of the days[14]. However, a study conducted by Man et al[15] reported no significant difference in height over a two-year period between children receiving MPH and those not receiving the medication. Similarly, long-term administration of ATX in Korean children was reported to adversely affect height growth, while the use of stimulant therapy has been associated with weight loss[16].

It is essential to understand growth changes in children with ADHD beyond the effects of medication, as research suggests that ADHD may independently influence growth, thereby increasing the risk of overweight or obesity[17]. Of note, results from a recent cross-sectional cohort study by Takım and Gökçay[18] indicated that ADHD pharmacotherapy significantly reduced mind-wandering. Likewise, in a recent study, narcissistic traits were ameliorated and empathy was enhanced in patients with ADHD who received psychostimulant treatment[19]. Moreover, treatment with MPH was associated with a decrease in both impulsivity and premature ejaculation symptoms in patients with ADHD, particularly with immediate-release formulations[20].

To date, no systematic review and meta-analysis has examined the impact of ADHD on the growth of children and adolescents. Existing meta-analyses have focused on the effect of individual pharmacological treatments on growth outcomes in this population[21,22]. This study aims to conduct a meta-analysis that establishes a robust foundation for understanding: (1) whether an ADHD diagnosis itself is associated with growth alterations and the risk of overweight/obesity, independent of medication; and (2) how pharmacological treatment for ADHD affects growth parameters.

MATERIALS AND METHODS

Literature search and selection

This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses 2020 guidelines[23].

A comprehensive literature search was performed in four electronic databases: PubMed, EMBASE, the Cochrane Library, and Web of Science, from their inception to September 29, 2023. The search strategy incorporated a combination of MeSH terms and free-text keywords related to ADHD and growth outcomes, including: “ADHD”, “amphetamine”, “methylphenidate”, “atomoxetine hydrochloride”, “modafinil”, “bupropion”, “guanfacine”, “body height”, and “body weight”. The complete search strategies for each database are provided in Supplementary Tables 1-4.

Inclusion criteria for studies evaluating the association between ADHD treatment and growth were as follows: (1) Patients < 18 years old diagnosed with ADHD; (2) Medication use for ADHD; (3) Pre-treatment vs post-treatment measurements; (4) Changes in weight (actual value, standard deviation score [SDS], or z score), height (actual value, SDS, or z score), body mass index (BMI; actual value, SDS, or z score); and (5) Randomized controlled trials (RCTs) or observational studies. Inclusion criteria for studies evaluating the association between ADHD and growth were as follows: (1) Patients < 18 years old and diagnosed with ADHD; (2) Intervention: Not applicable; (3) Children or adolescents without ADHD; (4) Outcomes: Odds ratios (ORs) for overweight and obesity between the ADHD and control groups; and (5) Study design: RCTs or observational studies.

Exclusion criteria for the two aspects of this meta-analysis were: (1) studies with non-extractable data; (2) case reports, reviews, conference proceedings, or animal studies; (3) duplicate research populations; (4) non-English citations; and (5) gray literature. Literature search and study selection were conducted by two independent researchers, and any disagreements were resolved through consultation with a third reviewer. Titles and abstracts of all retrieved citations were first screened, followed by an in-depth full-text review of candidate studies according to the predefined inclusion criteria. Furthermore, the reference lists of relevant review articles were also screened to find more eligible studies.

Data extraction and quality assessment

Two reviewers independently extracted relevant data from each included study, including the first author, publication year, study design, country, sample size, mean participant age, proportion of female participants, type of ADHD medication, dosage, and treatment duration.

The risk of bias in RCTs was assessed using the revised Cochrane Risk of Bias tool (RoB 2), which evaluates five domains: The randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of the reported result. Each domain was rated as having low risk, some concerns, or high risk of bias[24]. For non-randomized observational studies, the Newcastle–Ottawa Scale was used. This scale evaluates study quality across three broad domains: Selection of study groups, comparability, and outcome/exposure assessment. Studies were rated as low quality (0-3), moderate quality (4-6), or high quality (7-9)[25]. For non-randomized interventional studies, methodological quality was evaluated using the Methodological Index for Non-Randomized Studies (MINORS), which includes 12 criteria such as a clearly stated aim, consecutive patient inclusion, and appropriate statistical analyses. Each item was scored from 0 (not reported) to 2 (adequately reported), with higher total scores indicating better methodological quality[26]. All assessments were conducted independently by two reviewers, with disagreements resolved through discussion and consensus.

Statistical analysis

Data analyses were performed using STATA version 15.0 (StataCorp, College Station, TX, United States). For continuous outcomes, such as changes in weight, height, and BMI from pre- to post-intervention, weighted mean differences (WMDs) with corresponding 95% confidence intervals (95%CIs) were calculated. A positive WMD indicated an increase in the post-intervention value relative to the baseline, whereas a negative WMD indicated a decrease.

For dichotomous outcomes, such as the risk of overweight and obesity, ORs were computed to compare the ADHD and control groups. An OR greater than 1 suggested an elevated risk in the ADHD group, while an OR less than 1 indicated a lower risk relative to controls.

Meta-analyses were conducted using a random-effects model to account for potential heterogeneity across studies. Statistical heterogeneity was assessed using the I² statistic, with values greater than 50% and a P value < 0.05 indicating substantial heterogeneity. Publication bias was evaluated visually using funnel plots and statistically using Begg’s and Egger’s regression tests. Trim-and-fill analysis was performed only for outcomes with significant publication bias, as indicated by Egger’s test. In addition, sensitivity analyses were performed by sequentially omitting individual studies to assess the robustness of the pooled estimates. A two-tailed P-value < 0.05 was considered statistically significant in all analyses.

RESULTS

Study selection and baseline characteristics

The database search yielded 6584 citations, from which 1409 duplicates and 5143 ineligible studies were excluded after initial screening. Eighteen studies were identified for inclusion following full-text reading[27-43], as depicted in Figure 1. Among these, 1 was an RCT, 3 were prospective studies, and 14 were retrospective studies. These studies were conducted in various countries, including China, Iran, Italy, South Korea, Norway, Poland, Spain, Sweden, Turkey, and the United States. The baseline characteristics of the included studies are detailed in Tables 1 and 2.

Figure 1  Flowchart of literature search.

Table 1 Characteristics of included studies with regard to the impact of attention-deficit hyperactivity disorder on growth.

Ref.	Study design	Country	Sample size	Age (year)	Female (%)	Mean BMI	ADHD		Control		Obesity		Overweight		Normal
							n	Age	n	Age	n	Age	n	Age	n	Age
Pagoto et al[32], 2009	Retrospective study	United States	6737	31	51.6	-					1475	-	3294	-	1968	-
Fuemmeler et al[33], 2011	Retrospective study	United States	11666	28.8	49	29.1					4343	28.9	3464	28.8	3849	28.6
Aguirre Castaneda et al[37], 2016	Retrospective study	United States	1001	24.41	16.98	24.36	336	-	665	-
Hanć et al[36], 2015	Retrospective study	Poland	615	10.94	0	-	219	-	396	-
Byrd et al[34], 2013	Retrospective study	United States	3050	-	-	-	412	-	2638	-

ADHD: Attention-deficit hyperactivity disorder; BMI: Body mass index.

Table 2 Characteristics of included studies with regard to the impact of medications for attention-deficit hyperactivity disorder on growth.

Ref.	Study design	Country	Sample size	Age (year)	Female (%)	ADHA drug	Dose	Duration of medication
Deng et al[42], 2021	Retrospective study	China	86	8.9	17.4	MPH and ATX	MPH: 18 mg/day; ATX: Weight < 70 kg, 0.5 mg/kg/day, weight > 70 kg, 40 mg/day	6 months
Çevikaslan et al[41], 2021	Retrospective study	Turkey	31	7.3	0	MPH	18 mg/day	6 months
Spencer et al[30], 2005	Retrospective study	United States	419	10.6	22.2	MPH	1.2 mg/kg/day	2 years
Sund and Zeiner[29], 2001	Retrospective study	Norway	91	-	0	AMP and MPH	-	5 years
Díez-Suárez et al[39], 2017	Retrospective study	Spain	342	10.7	19.9	MPH	1.25 mg/kg/day	2 years
Mellström et al[28], 2020	Retrospective study	Sweden	724	10	27.2	MPH	0.81 mg/kg/day	3 years
Fast et al[43], 2021	Retrospective study	Sweden	44	9.9	25	MPH or lisdexamfetamine	-	1 year
Germinario et al[35], 2013	Prospective study	Italy	1758	-	11.4	MPH and ATX	MPH: 0.48 mg/kg/day; ATX: 38.7 mg/day	2 years
Mei et al[44], 2022	Prospective study	China	155	8	37.58	ATX	0.5 mg/kg/day	3 months
Spencer et al[31], 2007	Retrospective study	United States	1312	11	23.5	ATX	0.5 mg/kg/day	5 years
Gurbuz et al[38], 2016	Prospective study	Turkey	89	9.75	0	ATX	0.5 mg/kg/day	3 months
Jahangard et al[40], 2017	RCT	Iran	42	8.54	28.57	MPH + RISP	MPH + RISP (1 mg/kg/d + 0.5 mg/d)	2 months
Lee et al[27], 2022	Retrospective study	South Korea	82	8.17	22	MPH	-	2 years

ADHD: Attention-deficit hyperactivity disorder; MPH: Methylphenidate; ATX: Atomoxetine; AMP: Amphetamine; RISP: Risperidone.

The overall risk of bias in included studies was generally low (Tables 3, 4, and 5). The RCT[40] showed low risk across all five RoB 2 domains. The 13 non-randomized interventional studies typically scored high, receiving 2 points in the relevant domains, except for the study by Germinario et al[35], which lacked a control group. The non-randomized observational studies were rated high quality according to the Newcastle–Ottawa Scale; however, the comparability of study groups was not adequately specified in the studies by Fuemmeler et al[33] and Hanć et al[36].

Table 3 Risk-of-bias assessment using the RoB 2 tool.

Ref.	D1	D2	D3	D4	D5	Overall
Jahangard et al[40], 2017	Low	Low	Low	Low	Low	Low

Table 4 Risk-of-bias assessment using the Methodological Index for Non-Randomized Studies tool.

Ref.	A stated aim of the study	Inclusion of consecutive patients	Prospective collection of data	Endpoint appropriate to the study aim	Unbiased evaluation of endpoints	Follow-up period appropriate to the major endpoint	Loss to follow-up not exceeding 5%	Prospective calculation of the sample size	A control group having the gold standard intervention	Contemporary groups	Baseline equivalence of groups	Statistical analyses adapted to the study design	Total
Deng et al[42], 2021	2	2	2	2	2	2	2	2	-	-	-	-	16
Çevikaslan et al[41], 2021	2	2	2	2	2	2	2	2	-	-	-	-	16
Spencer et al[30], 2005	2	2	2	2	2	2	2	2	-	-	-	-	16
Sund and Zeiner[29], 2001	2	2	2	2	2	2	2	2	-	-	-	-	16
Díez-Suárez et al[39], 2017	2	2	2	2	2	2	2	2	-	-	-	-	16
Mellström et al[28], 2020	2	2	2	2	2	2	2	2	-	-	-	-	16
Fast et al[43], 2021	2	2	2	2	2	2	2	2	-	-	-	-	16
Germinario et al[35], 2013	2	2	2	2	2	2	2	2	2	2	2	2	24
Mei et al[44], 2022	2	2	2	2	2	2	2	2	-	-	-	-	16
Spencer et al[31], 2007	2	2	2	2	2	2	2	2	-	-	-	-	16
Gurbuz et al[38], 2016	2	2	2	2	2	2	2	2	-	-	-	-	16
Byrd et al[34], 2013	2	2	2	2	2	0	2	2	-	-	-	-	14
Lee et al[27], 2022	2	2	2	2	2	2	2	2	-	-	-	-	16

Table 5 Risk-of-bias assessment using the Newcastle–Ottawa Scale tool.

Ref.	Is the case definition adequate?	Representativeness of the cases	Selection of controls	Definition of controls	Comparability of cases and controls based on the design or analysis	Ascertainment of exposure	Same method of ascertainment for cases and controls	Non-response rate	Total
Pagoto et al[32], 2009	1	1	1	1	2	1	1	1	9
Fuemmeler et al[33], 2011	1	1	1	1	1	1	1	1	8
Aguirre Castaneda et al[37], 2016	1	1	1	1	2	1	1	1	9
Hanć et al[36], 2015	1	1	1	1	1	1	1	1	8

Effects of pharmacological treatment for ADHD on growth

Effects on weight: The pooled WMD for actual weight (6 studies) was -9.50 (95%CI: -15.47 to -3.53; I² = 96.7%, P < 0.001; Figure 2A). Subgroup analyses revealed that WMDs of actual weight for follow-up duration ≤ 1 year and > 1 year were -0.74 (95%CI: -2.66 to 1.17; I² = 28.9%, P = 0.236) and -13.54 (95%CI: -19.33 to -7.76; I² = 92.4%, P < 0.001; Supplementary Figure 1). The WMD for weight in SDS (3 studies) was 0.23 (95%CI: -0.03 to 0.49; I² = 79.8%, P = 0.007; Supplementary Figure 2). Subgroup analyses revealed that WMDs of weight in SDS for follow-up duration ≤ 1 year and > 1 year were 0.08 (95%CI: 0.00 to 0.16; I² = 0%, P = 0.347) and 0.40 (95%CI: 0.20 to 0.60; I² = NA, P = NA; Supplementary Figure 3). The WMD for weight in z score (4 studies) was 0.21 (95%CI: 0.06 to 0.37; I² = 45.9%, P = 0.136; Supplementary Figure 4). Subgroup analyses revealed that WMDs of weight in z score for follow-up duration ≤ 1 year and > 1 year were 0.29 (95%CI: 0.04 to 0.53; I² = 0%, P = 0.673) and 0.16 (95%CI: -0.04 to 0.37; I² = 79.2%, P = 0.028; Supplementary Figure 5).

Figure 2 Forest plot. A: Actual weight; B: Actual height; C: Body mass index.

Effects on height: The pooled WMD for height in actual value (five studies) was -0.15 (95%CI: -0.24 to -0.06; I² = 97.4%, P < 0.001; Figure 2B). WMDs of actual height for follow-up duration ≤ 1 year and > 1 year were -0.01 (95%CI: -0.06 to 0.03; I² = NA, P = NA) and -0.18 (95%CI: -0.26 to -0.09; I² = 96.8%, P < 0.001; Supplementary Figure 6). The WMD for height in SDS (3 studies) was 0.09 (95%CI: -0.03 to 0.20; I² = 0%, P = 0.926). The follow-up duration of all included studies was ≤ 1 year (Supplementary Figure 7). The WMD for height in z score (5 studies) was 0.10 (95%CI: -0.04 to 0.24; I² = 0%, P = 0.661; Supplementary Figure 8). Subgroup analyses revealed that the WMDs of height in z score for follow-up duration ≤ 1 year and > 1 year were 0.10 (95%CI: -0.13 to 0.34; I² = 0%, P = 0.401) and 0.10 (95%CI: -0.08 to 0.28; I² = 0%, P = 0.466; Supplementary Figure 9).

Effects on BMI: The WMD for BMI in actual value (three studies) was -1.72 (95%CI: -4.10 to 0.65; I² = 93.8%, P < 0.001; Figure 2C). WMDs of actual BMI for follow-up duration ≤ 1 year and > 1 year were -1.41 (95%CI: -1.90 to -0.91; I² = NA, P = NA) and -1.87 (95%CI: -2.41 to -1.33; I² = 94.0%, P < 0.001; Supplementary Figure 10). The WMD for BMI (five studies) in SDS was 1.43 (95%CI: 0.64-2.22; I² = 98.5%, P < 0.001; Supplementary Figure 11). The WMD for BMI in SDS for follow-up duration ≤ 1 year and > 1 year were 1.75 (95%CI: -0.25 to 3.76; I² = 98.7%, P < 0.001) and 1.05 (95%CI: 0.12, 1.98; I² = 97.8%, P < 0.001; Supplementary Figure 12). The WMD for BMI in z score (3 studies) was 0.18 (95%CI: -0.05 to 0.42; I² = 52.7%, P = 0.121; Supplementary Figure 13). Subgroup analyses revealed that WMDs of height in z score for follow-up duration ≤ 1 year and > 1 year were 0.11 (95%CI: -0.15 to 0.37; I² = NA, P = NA) and 0.23 (95%CI: -0.16 to 0.62; I² = 73.6%, P = 0.052; Supplementary Figure 14).

Association between ADHD diagnosis and risk of overweight or obesity

Pooled OR for overweight between ADHD and control (three studies) was 1.37 (95%CI: 0.90-2.08; I² = 79.9%, P = 0.007; Supplementary Figure 15). Pooled OR for obesity between ADHD and control (five studies) was 1.16 (95%CI: 0.80-1.67; I² = 71.3%, P = 0.004; Supplementary Figure 16).

Publication bias

Egger’s tests demonstrated no significant publication bias in any of the pooled analyses except for BMI in SDS (P = 0.035; Supplementary Table 5). The WMD for BMI in SDS was 1.08 (95%CI: 0.35-1.82) using the trim-and-fill method, which indicated that the overall findings for this outcome were not significantly influenced by publication bias (Supplementary Table 5 and Supplementary Material).

Sensitivity analysis

Sensitivity analysis confirmed that the synthesized outcomes were robust (Supplementary Figures 15-25).

DISCUSSION

This meta-analysis synthesized data from 18 studies to evaluate the effects of pharmacological treatment for ADHD on growth trajectories in children and adolescents, and the association between ADHD diagnosis and the risk of overweight or obesity. These two aspects were analyzed independently to distinguish between the impact of medical intervention and that of the disorder itself. However, we acknowledge that complete separation of these effects remains challenging in real-world observational data, where medication use is often intertwined with the diagnosis.

While actual values reflect absolute changes, SDS and z-scores represent deviations from age- and sex-adjusted population norms. The observed stability or modest increases in standardized scores (height z-score WMD = 0.10, 95%CI: -0.04 to 0.24; weight SDS WMD = 0.23, 95%CI: -0.03 to 0.49) paradoxically suggest that pharmacological treatment may attenuate but not fundamentally alter growth trajectories relative to population expectations. The observed paradoxical discrepancy—a reduction in absolute weight and height alongside stable or slightly increased standardized scores (SDS/z-scores)—is most plausibly explained by transient suppression of growth velocity followed by catch-up growth. This pattern suggests that pharmacological treatment, particularly stimulants, may temporarily decelerate growth, potentially through appetite suppression and metabolic alterations. However, during treatment pauses or discontinuation, a compensatory acceleration (catch-up) likely occurs, allowing children to partially recalibrate their trajectory toward their genetic potential. Consequently, while absolute measurements may remain lower, the standardized scores—which compare the child to age- and sex-matched peers—can remain stable, indicating a growth trajectory that runs parallel to the population norm, albeit from a lower baseline.

The results of this study were consistent with previous reports. Mei et al[44] reported significant weight loss after ATX treatment, while another study reported that body weight SDS was statistically significantly reduced in children with ADHD who were treated with MPH[38]. Similarly, Spencer et al[31] observed that children with ADHD weighed less than expected from their baseline percentiles compared to standard norms and were shorter than expected. Early side effects of ATX, such as nausea and stomachache, might reduce appetite, a phenomenon also reported with MPH[45-47]. In one longitudinal study, a significant decrease in weight gain was observed at the first follow-up (after 1 month of continuous treatment), followed by a very slight decrease in height in the first year. However, the average height returned to the expected value at the end of treatment in the second year of follow-up, and the average weight normalized by the third year, indicating that these developmental impacts are not long-lasting or irreversible[31]. Longer follow-up (> 1 year) showed continued reductions in weight, height, and BMI, aligning with meta-analyses reporting a modest, clinically minimal impact of long-term MPH on growth. Importantly, the impact is generally small and of limited clinical concern[24]. According to the National Institute for Health and Care Excellence guidelines, short-term weight loss does not need intervention unless it becomes clinically significant. In such cases, medication should be taken during or after meals. An additional morning or evening snack is also recommended for children experiencing weight loss[44].

These findings underscore the importance of regular monitoring during ADHD treatment in children and adolescents. A 6-monthly or annual check-up is recommended, during which, in addition to evaluating the medication’s effect, height, weight, and BMI should be measured and plotted against local growth charts to evaluate growth trajectories[48].

We also explored the link between ADHD and childhood overweight/obesity and found no significant association with either outcome, consistent with prior research[32,33]. However, the non-significant obesity risk must be interpreted conservatively given the heterogeneity observed (I² = 71.3%) and the potential influence of unmeasured confounders such as diet and physical activity. A possible behavioral explanation is that children with ADHD may be less responsive to food cues and more likely to consume more calories when distracted by stimuli such as television or video games, compared to adults[33,49,50]. Fuemmeler et al[51] reported bidirectional associations between ADHD symptoms and obesity-related eating behaviors, finding that early symptoms of ADHD in children (e.g., at age 4) predicted greater increases in obesity-related behaviors and BMI in later stages of childhood, but the opposite was not observed. Interestingly, Martins-Silva et al[52] utilized Mendelian randomization to control for reverse causality and residual confounding factors in observational studies, revealing a causal impact of BMI on the risk of ADHD[53]. Children with ADHD often face stigma, family stress, and academic difficulties, all of which can negatively affect eating habits, sleep, and stress physiology, and subsequently, growth and development[54]. Further causal studies are necessary to clarify these mechanisms.

Strengths and limitations

This meta-analysis investigated the impact of ADHD medications and ADHD itself on the growth of children and adolescents—a topic not previously well-explained. The overall quality of the included studies was high relative to their respective study designs. A random-effects model was used to provide a conservative estimate appropriate for heterogeneous data.

Insignificant publication bias was detected for most outcomes, except for BMI SDS. Sensitivity analyses, conducted by sequentially omitting individual studies, further demonstrated the stability of the findings.

Nevertheless, this meta-analysis has several limitations. First, the number of included studies for each outcome was limited, and many were retrospective in design. Significant heterogeneity was observed in most pooled outcomes, which may be attributed to differences in medication types, dosing protocols, participant characteristics, and growth measurement intervals (some < 6 months). Although this variability reflects the diversity of real-world clinical practices, it may limit the generalizability of the findings. Future studies using standardized protocols and prospective designs are needed to better clarify the impact of ADHD treatment on growth. Second, significant reductions were detected in actual values of body weight, height, and BMI instead of SDS or z scores, and these findings should therefore be interpreted with caution. Third, the included studies evaluated only one or two types of anti-ADHD agents with fixed doses, such as MPH, ATX, AMP, risperidone, and lisdexamfetamine. Positive results were reported in studies examining MPH, MPH+AMP, MPH+ATX, and ATX alone. Although we planned to perform subgroup analysis by drug type and dosage, the limited number of enrolled studies precluded meaningful stratification by these factors. MPH was the predominant agent in our included studies (13 out of 18), restricting broader generalization. Nevertheless, existing evidence suggests that different agents may exert differential effects on growth. For example, cohorts exposed to ATX[17,30] exhibited a more pronounced reduction in height SDS (-0.38) compared to those receiving MPH (-0.21), indicating potential agent-specific impacts that merit further investigation. Fourth, restricting inclusion to English-language studies may have omitted relevant data from non-English publications, though this aligns with standard meta-analytic practice to ensure accurate data extraction. Fifth, the observed decreases in actual weight and height are statistically significant; however, their clinical importance is limited as they are generally modest. It should also be noted that the extent of these changes varies markedly across studies. Finally, while publication bias was quantitatively assessed using Egger’s test only for outcomes with sufficient study numbers and was significant only for BMI in SDS, we cannot fully rule out the potential for unpublished negative or null findings, particularly for outcomes with high heterogeneity. The generally symmetrical appearance of the funnel plots for other outcomes offered visual reassurance, but the possibility of selective outcome reporting remains a common limitation in this meta-analysis. More prospective studies with standardized protocols are warranted to compare the effects of different medication types, doses, and formulations over the long term. Such evidence will be essential to establish causality and develop evidence-based treatment guidelines.

CONCLUSION

This meta-analysis indicates that children and adolescents with ADHD undergoing long-term pharmacological treatment may experience reduced growth, underscoring the need for careful monitoring of growth parameters at baseline, 3 months after treatment initiation, and every 6 months thereafter. If significant growth deceleration is detected, mitigation strategies such as nutritional counseling, medication dose adjustment, or consideration of drug holidays are recommended. Additionally, no significant association was observed between ADHD and overweight/obesity.