The effect of dietary approaches to stop hypertension (DASH) diet on attention-deficit hyperactivity disorder (ADHD) symptoms: a randomized controlled clinical trial

The dietary approaches to stop hypertension (DASH) diet have several components like high amounts of fruits, vegetables, low-fat dairy products, and vitamin C and low amounts of simple sugars that might improve attention-deficit hyperactivity disorder (ADHD) symptoms. We aimed to investigate the effect of a DASH diet on children (aged 6–12 years) with ADHD, for the first time. Participants were randomized to receive a DASH or a control diet for 12 weeks. The severity of ADHD symptoms [determined by abbreviated 10-item Conner’s scale (ACS), 18-item Swanson, Nolan and Pelham (SNAP-IV) scale and strengths and difficulties questionnaire (SDQ)] were assessed every four weeks. Eighty children completed the study. After adjustment for confounders, parent (− 4.71 for the DASH group vs. − 3 for the control group) and teacher-reported (− 5.35 vs. − 1.87) ACS scores, parent-, teacher-, child-reported hyperactivity (− 1.40 vs. − 0.66, − 1.95 vs. -0.63, − 1.60 vs. − 0.43, respectively), emotional symptoms (− 1.50 vs. − 0.45, − 1.42 vs. − 0.63, and − 1.09 vs. − 0.61, respectively), and total SDQ scores (− 3.81 vs. − 1.65, − 4.11 vs. − 1.23, − 4.44 vs. − 1.26, respectively), teacher-reported of conduct problems (− 1.42 vs. − 0.63), peer relationship problems (− 0.87 vs. − 0.07), and prosocial behaviors (1.36 vs. 0.08) assessed by the SDQ were significantly improved in the DASH group compared with the control group (P < 0.05). Adherence to a DASH-style diet might improve ADHD symptoms. Further RCTs which include participants from both sexes and with longer follow-up periods are needed to warrant current findings (The trial registration code: IRCT20130223012571N6; http://irct.ir/trial/12623). Trial registration Trial registration number: The trial was registered in the Iranian registry of clinical trials (registration code: IRCT20130223012571N6), URL: http://irct.ir/trial/12623.


Introduction
Attention-deficit hyperactivity disorder (ADHD), a functional impairment due to persistent inattention, hyperactivity, and impulsivity, is one of the most common psychiatric disorders in school-aged children [1] which might lead to a lower educational, occupational, and social level in later life [2][3][4]. The prevalence of ADHD is estimated to be 5.29% worldwide [5]. This amount was shown to be 12.6% in Tehran [6] and 16.3% for children living in Yazd city, Iran [7]. Several genetic and environmental factors are proposed to be associated with the development of the disease [8].
Dietary interventions including restriction and elimination diets (the removal of food colors and other additives, sugars, and sweeteners), few-foods diet, as well as supplementation of essential fatty acids, amino acids, vitamins, and minerals are investigated for their possible beneficial effect on ADHD symptoms [9]. Studies on the effect of supplementation with amino acids (phenylalanine, tyrosine and tryptophan) [10,11], vitamins (B complex and vitamin C) [12,13], minerals (iron, zinc and magnesium) [14,15], and essential fatty acids [omega-3, omega-6, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)] [16,17] have led to conflicting results. Investigations on the effect of elimination diets were also inconsistent in their findings [18][19][20]. A recent systematic review and metaanalysis suggested that the few foods diet (FFD) followed by food challenge might be an effective strategy in controlling ADHD [21,22]. Indeed, the FFD is not regarded as a treatment and the approach of starting with diet and following the food challenge might take at least one year [22]. As the diet puts a high burden on children and their families, it is recommended for children with ADHD who do not respond to medications, those who are too young for medication [21], and children living in highly motivated families [22]. Furthermore, the FFD is a restrictive diet and might not fully meet children's nutritional needs and growth [9].
The Dietary Approaches to Stop Hypertension (DASH) is a diet rich in fruits, vegetables, whole grains, fish, and low-fat dairy products [23]. This diet was initially designed for treating hypertension; however, later investigations observed that the DASH might beneficially affect blood glucose [24] and lipid profile [25]. It is revealed that a fewer fruits and vegetables consumption is associated with more ADHD-like symptoms in children [26,27]. Furthermore, a clinical trial showed that the inactivity score in children with ADHD is significantly related to fruits and vegetables consumption [28]. Fish intake that helps to provide essential fatty acids, such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), is also recommended in the DASH diet. The consumption of these n-3 fatty acids might significantly reduce ADHD symptoms [29,30]. The diet also contains high amounts of calcium and magnesium, which their supplementation is suggested to be effective in ADHD treatment [31][32][33]. This diet is also rich in vitamin C that might reduce the total hyperactivity score [34]. While it can provide enough amounts of nutrients for supporting children's growth, the DASH diet is low in food colors, simple sugars, and artificial sweeteners [35]; therefore, we hypothesized that this diet might be a good choice for ADHD treatment. Therefore, the present randomized controlled clinical trial (RCT) was conducted to examine the possible effect of the DASH diet compared with an isocaloric control diet in children with ADHD.

Study protocol
The present study was a randomized controlled clinical trial conducted in children aged 6 to 12 years who attended psychology clinics of Shahid Sadoughi University of Medical Sciences and were recently diagnosed with ADHD according to the Conner's teacher questionnaire and the DSM-4 criteria by a psychiatrist. The children were excluded from the study if: (1) they used medication or behavioral therapy for ADHD before or during the intervention period; (2) had a history of any other neurological diseases associated with ADHD-like anxiety; (3) had a low intelligence quotient score (less than 70) (assessed by the Wechsler IQ test); (4) were adopted or foster child; (5) experienced a recent major dietary change; (6) themselves or their parents decided not to participate or refused to continue the study with any reason. Parents were given verbal and written information about the study and signed informed consent before participation. The present study was conducted following the declaration of Helsinki and its protocol was approved by the ethics committee of Shahid Sadoughi University of Medical Sciences, Yazd, Iran (ethical approval code: IR.SSU.SPH.REC.1395.106) and registered in the Iranian registry of clinical trials (IRCT) with the registry code of IRCT20130223012571N6 on 8th February 2018 (URL: http:// irct. ir/ trial/ 12623). We followed the Consolidated Standards for Reporting Trials (CON-SORT) statement for reporting the current study [36].

Study design
The eligible children were randomly assigned to receive a DASH or a control diet for three months (12 weeks). Simple randomization method was conducted using the statistical package for social sciences (SPSS) software. The randomization concealment was done by putting the allocated group in sealed and opaque envelopes until the assignment. Both randomization and randomization concealment were conducted by an independent researcher. Children and their parents were visited each month for up to three months, to determine the severity of ADHD symptoms using the abbreviated 10-item Conner's scale (ACS), 18-item Swanson, Nolan and Pelham (SNAP-IV) scale, and strengths and difficulties questionnaire (SDQ), and to assess the dietary intake, physical activity, and anthropometric measurements. As the purpose of the study was to compare the effect of two diets on ADHD symptoms, parents and children could not be blinded; however, teachers were not aware of the assigned intervention group.

Intervention diets
The estimated energy requirement (EER) was calculated using a formula proposed by the American Institute of Medicine (IOM) based on weight, height, age, and physical activity level for each participant [37]. Intervention diets were designed by a trained nutritionist. Indeed, investigators designed DASH and control diets for each participant according to their EER. Then, researchers explained to parents and children how to implement the diet. Parents were also provided with sample diets and food substitution lists. Therefore, they could choose foods based on prescribed diets. Either DASH or control diets were designed to provide 50-60% of EER from carbohydrates, 25-30% from fats, and 15-20% from proteins. The DASH and the control diet were different in the prescribed dietary food groups. The DASH diet was designed to contain higher amounts of whole grains, fruits, vegetables, low-fat dairy products, nuts, and beans, as well as low amounts of saturated fats, cholesterol, refined grains, sweets, and red meat [23]. The control diet was designed to be similar to the usual diet of Iranian children [38]. Indeed, the control diet allowed for refined grains, full-fat dairy, and meats. It also had lower amounts of fruits and vegetables. Simple sugars were also allowed in the control diet [39,40]. Table 1 compares food groups as well as their servings for a 1500 kcal DASH and a 1500 kcal control diet prescribed in the present study.

ADHD severity assessment
Three questionnaires including abbreviated 10-item Conner's scale (ACS), 18-item SNAP-IV scale, and SDQ were used to assess the ADHD severity. Conner's questionnaire consists of ten behavioral questions focusing on hyperactivity and inattention. According to Conner's questionnaire, children with a score of 15 and above might have attention-deficit/hyperactivity disorder. The validity and reliability of this questionnaire were shown in Iranian children [41]. The questionnaire was filled by the participants' teachers and parents in each visit. The 18-item Swanson, Nolan, and Pelham (SNAP-IV) Rating scale has 9 questions about attention-deficit disorder (ADD), 5 questions about hyperactivity, and 4 questions about impulsivity. The overall hyperactivity and impulsivity were called hyperactivity disorder (HD). Three measures were assessed using this questionnaire: total score (combined), attention-deficit disorder (ADD), and hyperactivity and impulsivity (HD) score. The validity and reliability of this questionnaire were also shown to be acceptable for Iranian children [42]. This questionnaire was also filled by parents and teachers. The SDQ questionnaire which has 25 questions was filled by children, their parents, and teachers in each visit. The SDQ provides data on five subscales of emotional symptoms, hyperactivity, conduct problems, peer relationship problems, and prosocial behavior. The total subscale is generated by adding all components except prosocial behavior together. The reliability and validity of the questionnaire were also approved in Iranian children [43].

Anthropometric measurements
Anthropometric measurements were performed for each person at baseline and each month up to three months. The weight of participants was assessed using a digital body analyzer (Omron Inc. Osaka, Japan; model no: BF511), up to the nearest 0.1 kg, with minimum possible clothing. Height was also measured using a wall-fixed height gauge to the nearest 0.5 cm. Body mass index (BMI) was calculated by dividing the weight (in kilogram) by height (in meter squared). Waist circumference (WC) and hip circumference were measured using a non-stretchable measuring tape with a precision of 0.5 cm by following the standard guidelines. Body fat and muscle percentages were also measured using Omron BF511 (Omron Inc. Osaka, Japan) body analyzer. All anthropometric measurements were performed at the same time of the day and three times on each visit by a trained nutritionist and the mean values of them were recorded.

Dietary assessment
A trained nutritionist filled a 24-h dietary recall by interviewing parents at baseline. This was done to assess the detailed participants' dietary data and to make sure that they can adapt to intervention diets. Furthermore, parents were trained to keep 3-day dietary records (2 workdays and a weekend day), at the beginning and each month up to 3 months. The recorded food items were converted to grams using household measures and then the daily intake of macro-and micronutrients were calculated using Nutritionist IV software (version 3.5.2, Axxya Systems, Redmond, Washington, USA).

Physical activity measurement
Parents were recommended to maintain their children's usual physical activity during the study period. This was checked by asking parents to record the daily physical activity of children at the beginning and each month up to the end of the intervention period using three-day records (2 workdays and a weekend day). Recorded physical activities and their duration were converted to metabolic equivalent-hour/day (Met-h/day) using the MET intensity of each activity [44].

Sample size calculation and statistical analysis
The sample size was calculated to be at least 19 participants in each intervention group based on the following formula: [45] and assuming the mean difference (μ1-μ2) of 8.4 in teachers' ACS score between intervention and control groups with standard deviation (ó) of 9.14, alpha of 0.05, and power of 80% based on a previously published study [20]. We anticipated a high attrition rate and lower effects because of the nature of the diet; furthermore, the study team had access to children with ADHD. Therefore, we aimed to include at least 40 participants in each study arm. The normal distribution of quantitative data was evaluated by looking at histograms and incorporating the Kolmogorov-Smirnov test. Independent samples t test and Chi-square test were used for comparing qualitative and quantitative variables between the DASH and the control group, respectively. A paired-samples t test was used for intra-group comparisons of quantitative variables between baseline and after-intervention periods. A linear mixed model was used to assess the time, group, and time*group interaction effects on ADHD symptoms assessed by ACS, SNAP-IV, and SDQ scales after adjustment for participants' age, sex, energy intakes, parents' job, and education, and the baseline values. The change in ADHD symptoms' scores and scores reported at each follow-up visit were also compared between the intervention and control groups using the analysis of covariance (ANCOVA) considering the participants' age, sex, energy intakes, parents' job and education, and the baseline values as covariates. The Quantitative values are reported as means ± standard deviations (SDs) unless indicated. Statistical package for social sciences (SPSS) software version 20 was used for data analysis and P values less than 0.05 were considered as statistically significant.

Results
A total of 86 participants meeting the inclusion criteria were allocated to intervention and control groups (n = 43 per group) and 80 children completed the study. Three participants from each group left the study during the follow-up period for these reasons: lack of motivation (n = 4), taking medication (n = 1), and migration (n = 1). Therefore, 80 participants (n = 40 per group) were included in the statistical analyses. The study flow process is detailed in Fig. 1.
With respect to the baseline characteristics, there were no significant differences in age, anthropometric measurements, sex, and parental education, and occupation between the two groups ( Table 2).
The baseline and after follow-up dietary macro-and micro-nutrient intake, as well as their change values, are provided in Table 3. There was no difference in the two groups regarding nutrients intake, at baseline. The mean changes in dietary total fat, fiber, pyridoxine, vitamin C, vitamin A, vitamin K, and potassium intakes were significantly increased in the DASH diet group (P < 0.05). Also, in the control group, the energy, total fat, total carbohydrates, sugar, vitamin A, vitamin K, potassium and calcium intakes were significantly increased (P < 0.05). After the follow-up Fig. 1 The study flow diagram period, children in the DASH diet consumed higher amounts of protein, dietary fiber, thiamin, riboflavin, niacin, pyridoxine, folate, vitamin C, vitamin K, potassium, calcium, magnesium, and iron when compared to controls (P < 0.05). It should be noted that only the mean changes in dietary fiber and vitamin C intakes were significantly different between the two groups (P < 0.05).
The baseline and after-intervention ADHD symptoms' scores, assessed using ACS, SNAP-IV, and SDQ questionnaires are summarized in Table 4. There was no difference between the two groups at baseline (P > 0.05). After adjustment for age, sex, energy intake, parental education and occupation, and baseline values, significant improvements were observed in scores based on Conner's scale (P time < 0.05, Table 4). A significant time effect was also shown in some of SDQ subscales including parent, teacher and child-reported hyperactivity scores, conduct Problems, and total scores, parent and child-reported emotional symptoms, parent-reported scores of peer relationship problems, teacher-reported scores of prosocial behaviors (P time < 0.05); however, time effects were not significant for SNAP-IV questionnaire and its subclasses (P time > 0.05). The group effect was significant for parent and teacher-reported ACS score (P group < 0.05). Also, a significant group effect was indicted in all subscales of the SNAP-IV questionnaire (P group < 0.05) except for teacher-reported scores of HD. Significant group effects were also observed in some SDQ subscales including parent and child-reported scores of hyperactivity, emotional symptoms, peer relationship problems, and the total score. Moreover, the group effect for parent and teacher-reported scores of prosocial behaviors was significant (P group < 0.05). The group*time effect was not significant for all ADHD symptoms (P group*time > 0.05).
According to Table 5, after adjustment for possible confounders, the mean changes in ACS score were significantly higher in the DASH group when compared with the control group either for parents or teachers (P < 0.05). However, there was no significant difference between the two groups based on SNAP-IV questionnaire except for teacher-reported combined scores. The reduction in mean hyperactivity score, emotional symptoms, and the total SDQ score was higher in the intervention group compared to the control group, according to reports provided by parents, teachers, and children (P < 0.05). Furthermore, teachers reported more improvements in conduct problems, peer relationship problems, and prosocial behavior in children assigned to the DASH diet compared with those received the control diet (P < 0.05).
The age, sex, energy intake, and baseline values-adjusted means for ACS, combined subscale of SNAP-IV, and hyperactivity component of SDQ scores at baseline, months 1, 2, and 3 of follow-up, are depicted in Supplemental figures 1-3, respectively. In the second and the third months of the follow-up, parent-and teacher-reported ACS scores were significantly lower in the children who were assigned to the DASH diet (Supplemental figure 1, P < 0.05). The total SNAP-IV score was significantly different only in scores provided by parents in the third month of follow-up (Supplemental figure 2, P < 0.05). According to the parents, the SDQ hyperactivity score was significantly lower in the DASH group compared to the control group in the second and the third months of follow-up (Supplemental figure 3A, P < 0.05). The teacher-and children-reported scores were significantly different in the third month of follow-up (Supplemental figures 3B and C, P < 0.05).

Sensitivity analysis
We tried to include male and female children in the present clinical trial; however, only one girl was referred to the study team. Therefore, we performed a sensitivity analysis to assess if the study results change after the exclusion of the girl from all analyses. It was revealed that the group effect for teacher-reported emotional symptoms assessed by the SDQ questionnaire became statistically significant (P group = 0.04). Also, the group*time effect became significant for teacherreported hyperactivity score (P group*time = 0.04). No other changes were observed after sensitivity analysis.

Discussion
The present study revealed that adherence to the DASH diet might significantly improve ADHD symptoms assessed by ACS and SDQ questionnaire in children when compared to a control diet. This is while the effect was found to be marginal for the combined inattention and hyperactivity score assessed by the SNAP-IV questionnaire. To the best of our knowledge, the present study is the first randomized controlled clinical trial investigating the effect of the DASH diet on children with ADHD. The previous investigations have usually examined the effect of dietary components or restrictive diets on ADHD. For instance, in a study on 100 children aged 4-8 years, Pelsser et al. [46] reported that a restricted elimination diet might significantly improve ADHD symptoms assessed by the ACS. In a recent meta-analysis, it was proposed that restriction diets might have notable improving effects on ADHD symptoms [47]. As restrictive diets may not fully meet nutritional needs of children, long-term adherence to this diet might not be applicable [20].
It is indicated that examining the effect of dietary patterns might be more informative than nutrients especially in health promotion, such as mental health [48,49]. The DASH diet was initially designed to improve blood pressure and emphasizes high vegetables, fruits, and low-fat dairy products, fish, whole grains, poultry, and nuts intake. The diet is also limited in red meat, salt, and sweetened beverages [23]. In line with our results, Ghanizadeh et al. [28] have reported that high vegetables and dairy products consumption may improve behavioral problems in children with ADHD. Also, Park et al. [27] found that higher dairy products and vegetables intake and low adherence to a diet high in fried food, sweetened desserts, and salt might beneficially affect ADHD symptoms.
It is asserted that inflammation causes neurodegeneration due to the activation of immune cells in the brain which in turn might lead to the production of pro-inflammatory factors [50]. Therefore, fruit-and vegetable-rich diets' beneficial effects on cognitive behavior might be due to their high antioxidant content and anti-inflammatory properties [51]. Also, the beneficial effects of plant-rich diets may be due to increasing short-chain fatty acids (SCFAs) levels [52] that may trigger leptin production that might be reduced in impulsivity disorders [53][54][55].
The DASH dietary pattern also emphasizes the consumption of dairy products, especially low-fat dairy products, which might have positive effects on cognitive behaviors by improving glucose and calcium regulation [56]. Indeed, some studies indicated that poor glucose regulation is associated with cognitive impairments [57,58]. Besides, calcium dysregulation has been suggested to be a notable factor in neurodegeneration [59,60]. In addition, alpha-lactoalbumin, a protein found in dairy products, has beneficial effects on increasing serotonin levels and therefore, improving mood and cognitive behavior [56,61,62].
Fish intake, that provides a good source of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), is also recommended in the DASH diet. Indeed, omega-3 polyunsaturated fatty acids (PUFAs) have important roles in regulating cell survival, synaptic function, and neurogenesis [63,64]. Therefore, omega-3 PUFAs deficiency which might occur in children with ADHD, may alter dopaminergic and serotonergic neurotransmission systems and this leads to mood and cognitive disorders [63,65]. Moreover, it is suggested that DHA is associated with dopamine production which is shown to be dysregulated in ADHD [66]. The DASH is also rich in magnesium-containing foods. It is mentioned that magnesium deficiency is associated with several cognitive disorders which usually can be manifested in ADHD including lack of concentration, fatigue, and aggression [67]. Reduced serum magnesium levels have been indicated by several studies in patients with ADHD [31,[68][69][70].
There are some limitations in the current study which should be taken into account when interpreting its results. We could only use 3-day dietary food records to assess the adherence to the intervention diets; although no consensus exists about the gold standard method to assess DASH compliance, 24-h urinary sodium, potassium and their ratio might be better tools for assessing the compliance of the study participants, [71]. Furthermore, although we tried to include eligible children from both sexes, only one girl was included because, we did not use a stratified randomization based on the participants' sex. Therefore, the study results might not be attributed to female children with ADHD. In the present study, we tried to design a diet similar to the usual diet of Iranian children which allowed for a slightly higher amounts of refined grains, full-fat dairy and meats, and also simple sugars. However, it still provided high amounts of fruits and vegetables. Therefore, the control diet might not be completely different from the intervention diet; this might explain the improvements in ADHD symptom scores observed in the control group. The beneficial effects of the DASH diet might be even stronger if compared with unhealthy diets; however, it is not ethically acceptable to prescribe unhealthy diets for children. In addition, the current study was not a feeding trial and no food was provided for the study participants. Although this approach might reduce the control of investigators on dietary intakes of participants, it might be more generalizable because it enabled us to examine the effect of prescribing a DASH-style diet on ADHD symptoms in real life.
In conclusion, the present study provides evidence that adherence to the DASH dietary pattern might be effective in improving ADHD symptoms as it significantly improved ACS and SDQ scores in children when compared to a control diet. Future clinical trials on children with both sexes and also longer follow-up periods are recommended to confirm these findings.