Prevalence of obesity and overweight among Chinese children with attention deficit hyperactivity disorder: a survey in Zhejiang Province, China

Attention deficit hyperactivity disorder (ADHD) is recognized as one of the most common childhood psychiatric disorders, and is characterized by age-inappropriate inattention, hyperactivity-impulsivity, or both [1, 2].The condition is estimated to affect 5–10% of school-aged children worldwide, depending on different geographic areas, ethnic background, culture, and diagnostic criteria [3, 4]. ADHD is often comorbid with psychiatric and developmental disorders such as anxiety disorders, oppositional defiant disorder, conduct disorder, and learning disorders [5‐9], and is associated with impairment of academic and social functions [10, 11].

In recent years, ADHD was found to be linked with another increasing worldwide childhood disorder: obesity [12‐15]. The prevalence of ADHD in obese subjects is higher than that of the normal population. In 2002, Altfas [16] showed that 27.4% of obese adults were diagnosed with ADHD, which was higher than the prevalence in general adult subjects. In children and adolescents, similar results were obtained [14, 15]. In addition, children and adolescents with ADHD had a higher body mass index (BMI) than expected for the age- and gender-specific group. Holtkamp et al. [17] found that the mean BMI standard deviation scores of children diagnosed with ADHD were significantly higher than the reference values of the normal German population. In 2007, Lam and Yang [18] reported that subjects with high scores of ADHD tendency (but not diagnosed with ADHD) had an increased risk for obesity of 1.4 times that of subjects with low ADHD tendency in the Chinese adolescent population. In a large sample survey in the United States of America, Waring and Lapane [19] reported that children with ADHD were much heavier than those without ADHD.

Although many studies suggest a potential comorbidity between obesity and ADHD, other results have shown that the rates of overweight and obesity among children with ADHD were not higher than the rates in the general population [20, 21]. However, most of the studies were performed in specific subgroups [16, 17, 22‐24], such as American adults and children, German adults and children, Turkish adolescents, and so on. Research has shown that individuals from diverse ethnic backgrounds engage in lifestyle practices that may influence the prevalence of both conditions [3, 12], so it is important to investigate the possible comorbidity in populations with different ethnic and cultural backgrounds. To our knowledge, the prevalence of obesity and overweight in Chinese children with a diagnosis of ADHD has not been explored. One study conducted in China in 2007 by Lam and Yang [18] showed a close relationship between ADHD tendency and obesity. However, it was evaluated in a general Chinese population of adolescents, not in an ADHD-diagnosed group.

Taking this research into consideration, the present study was performed in Zhejiang Province, China. We hypothesized that the prevalence of obesity was higher in ADHD patients, and that a specific subtype of ADHD was associated with obesity. Our aims were as follows: 1) to investigate the prevalence of obesity or overweight among Chinese children with ADHD, and to compare this prevalence with what is reported in the general Chinese population, and 2) to analyze the prevalence of obesity/overweight in children with different subtypes of ADHD to identify which dimension(s) of ADHD may specifically be associated with obesity/overweight.

There were 162 children diagnosed with ADHD between March 1, 2010 and August 31, 2010. Four children were excluded from the study because either their consents were not obtained, or the information they provided was incomplete. The characteristics of the subjects are summarized in Table 1. The mean age was 9.2 years, with standard deviation of 2.0 (they ranged from 6–16.6 years), and most (94.3%, n = 149) were in primary school. Of the children with ADHD, 79.7% (n = 126) were from 7–10 years of age, 7.0% (n = 11) were younger than 7 years, and 13.3% (n = 21) were older than 11 years. The sample consisted mainly of children with ADHD inattentive subtype (50.6%, n = 80). There were more males (77.8%, n = 123) than females. The ratio was 3.5:1, similar to other epidemiologic reports [3].

The prevalences of obesity, overweight, and combined obesity/overweight were 12.0% (19/158), 17.1% (27/158), and 29.1% (46/158), respectively, in this group of children with ADHD. The prevalence depended on the pubertal developmental stage and ADHD subtype (Table 2).

Anthropometric data of the study samples, including the distribution of height and BMI in percentile, are presented in Table 3. The mean z-scores for the children’s height, weight, and BMI were -0.12 ± 1.01, 0.05 ± 1.36, and 0.20 ± 1.71, respectively. The mean BMI z-score (0.20 ± 1.71) of the children with ADHD was greater than 0. However, the difference did not reach statistical significance (P = 0.147). The proportion of children with a BMI below the 5^th percentile was 8.23% (13/158).

Univariate analysis showed that three potential risk factors were associated with obesity/overweight. Children in the pubertal stage were more likely to be overweight/obese (OR = 3.162, P = 0.027). Children with ADHD combined subtype had a greater possibility of being overweight/obese (OR = 2.192, P = 0.048). Gender was not a risk factor for obesity/overweight.

Multivariate analysis (Table 4) demonstrated that children in puberty who had ADHD had a four-fold increase in the odds ratio of obesity/overweight than those in the pre-pubertal stage (95% CI: 1.337-12.191). Children with ADHD combined subtype were 2.8 times more likely to be obese/overweight than those with either of the other two ADHD subtypes (95% CI: 1.225–6.434).

In the present study, we found that the prevalence of obesity/overweight in Chinese children in Zhejiang Province with ADHD is 29.1%, which is higher than the figure of 6.6% found in the national survey of the normal population in 2002 [28]. When we compared our results with the survey conducted in Zhejiang Province in 2005 (published in 2006), the finding was similar [29]. Table 5 shows the recent surveys on the prevalence of pediatric obesity throughout China [28‐38]. The present data reveal that Chinese children with ADHD have a higher prevalence of obesity/overweight. We re-evaluated the children’s growth data using the World Health Organization criteria. The rate of overweight was 15.8% (25/158), and that of obesity was 11.4% (18/158). This prevalence was much higher than that reported in other Provinces of China (Table 5), as well as that in Zhejiang Province reported by Yang [29]. In addition, the absolute mean BMI z-score of children with ADHD is larger than 0, which confirms that the children with ADHD are more likely to be associated with obesity/overweight. Our finding is in line with previous reports by others [18, 22, 23] and strongly suggests that obesity is another comorbidity of ADHD in Chinese children.

Here, we want to emphasize several strengths of this study. First, our study was conducted in children who were diagnosed with ADHD for the first time and who were medication-naive, which minimized the potential influence of drugs on growth status [39, 40]. Second, BMI z-score was employed as an indicator, which was a practical, convenient, and widely accepted screening tool and could be used to compare different age groups, avoiding mathematical distortion [41]. Third, we used the latest Chinese reference for children’s and adolescents’ growth to assess growth status, which could define the growth level more accurately. All of these points strengthen the results with more persuasiveness [26].

Several hypotheses [12, 42] have tried to explain the potential mechanism of ADHD and obesity, such as an imbalance in the dopaminergic reward system [43] and excessive daytime sleepiness [44]. For the imbalance of the dopaminergic reward system, an insufficient dopamine-related natural reward leads to the use of “unnatural” immediate rewards, like inappropriate eating [45]. Changes in dopamine receptor (DR) D2 [46] and DRD4 [47, 48] are found to be associated with this imbalance in the dopaminergic reward system in obese patients [49, 50] and ADHD patients [49, 51, 52]. Therefore, obesity and ADHD may result from this common pathway of dysfunctions in the dopaminergic system. For sleep problems, alertness alterations, such as sleep apnea and daytime sleepiness, have been found in ADHD as well as in obesity [53‐55]. It was hypothesized that excessive daytime sleepiness might begin to explain the association between ADHD and obesity [44], which had been confirmed in one study [56]. However, this hypothesis has a long way to go to explain this mechanism. However, if a subtype of ADHD (inattention, hyperactivity/impulsivity, or combined type) is found to be associated with obesity, the connection could offer an understanding of the underlying mechanism. The specific relationship is still unclear, though, as there are only a few reports on the subject [57, 58]. In our survey, children with ADHD combined type were 2.8 times as likely to be obese/overweight as those with the other two ADHD subtypes. It is difficult to explain our results using the existing theory. Previous studies suggested that ADHD impulsivity might foster abnormal eating behaviors, which may contribute to maintaining a sufficient energy balance and the development of obesity [58]. In addition, the inattention symptoms are closely related to poor executive function, resulting in irregular eating patterns and obesity. It is proposed that hyperactivity-impulsivity and inattentive behaviors have cooperation effects on the subjects that would develop into obesity, yet individually the aspects contributed little to obesity.

Interestingly, this study found that pubertal development was associated with a higher prevalence of obesity/overweight in children with ADHD. When the sample of the ADHD group was restricted, children in puberty were associated with a four-fold increase in the odds ratio of obesity/overweight compared with those in the pre-pubertal stage. van Egmond-Frohlich and colleagues [59] found that ADHD symptoms (assessed by the Strengths and Difficulties Questionnaire with the hyperactivity/inattention subscale) were related to overweight in adolescent girls. Our results confirmed the association in a clinic-based sample of pubertal children, but were not confined to pubertal girls because the present survey did not allow us to calculate that specific association. It was speculated that other factors, in addition to impulsive and inattentive symptoms, or perhaps hormones affecting children with ADHD, play important roles in producing obesity during puberty.

Another interesting finding was that the proportion of children with BMI below the 5^th percentile was 8.23% (13/158), suggesting that the prevalence of underweight was also higher in the ADHD group. Most of them were males (12/13) who were diagnosed with ADHD inattentive type (8/13), followed by ADHD hyperactive-impulsive type (3/13), and ADHD combined type (2/13). This result was first reported in a retrospective survey with only three cases [60] that showed two of the three ADHD patients were underweight. Because our samples were medication-naive, the possibility that medications induced underweight by appetite suppression was excluded. We speculated that the hyperactivities resulted in excessive energy expenditure, which contributed to underweight.

The potential limitations of this study should be taken into consideration. First, the cross-sectional design precludes determining any causal inferences in the relationship between ADHD and obesity. Second, no control group was set to compare with the study group. However, in addition to comparing our results with the national survey, other investigations in China were used for comparison, including one conducted in Zhejiang Province, which could produce relatively objective conclusions. It should also be taken into consideration that the prevalence of obesity is rising quickly, and the data may not be completely up to date. Third, demographic and family information, obesity family history, and other parental characteristics of the ADHD group were not collected, and our population may have different lifestyles, food sources, and general structure, which would affect the children’s nutritional status. Finally, pubertal stages were estimated based on self-report, and data on the variety of pubertal stages were not collected, which might produce bias and affect the results. Therefore, a scientific design using a multi-center and random sampling method with larger sample size could be used to better explore this comorbidity.

Our findings demonstrated that the prevalence of obesity in children with ADHD was higher than in the general population, which strongly supports obesity as another comorbidity of ADHD in Chinese children. Children with ADHD of the combined subtype and in puberty had a higher risk for obesity/overweight. These findings extend the comorbidity of ADHD and obesity to Chinese populations, consistent with previous findings reported in other populations.