ADHD involves inattention, impulsivity, and hyperactivity, whereas autism is characterized by stereotyped behavior and impaired social and communication skills. On the surface, there seems to be little in common between these two disorders at first glance.

One feature they do have in common is that they rarely present alone. In the words of two researchers:^[12] “It is the exception, not the rule, to encounter cases with ‘pure’ ADHD”, and the same also seems to hold true for ASD. It’s not uncommon for attention issues and full-blown ADHD to be found^[13] in children with ASD. The reason aspects of these conditions can overlap may come down to where they occur in the brain. Both involve some of the same regions of the brain, in what is known as the frontostriatal system. Disorders that arise from this region are thus known as frontostriatal disorders,^[14] which include ADHD and ASD. Recent neuropsychological evidence^[15] suggests that ADHD and ASD share some similar brain circuits and both involve problems with managing cognitive processes and emotions, self-control, and executing complex tasks (collectively known as executive dysfunction), implying that both disorders may have some underlying similarities.

Individuals with ASD may have a variety of co-occurring conditions, including epilepsy,^[26][27] bipolar-disorder,^[28] and immunological problems,^[29][30][31] such as asthma and atopic dermatitis.

Other conditions associated with ASD are fragile-x-syndrome and rett-syndrome.

Fragile X syndrome is a genetic condition that causes intellectual disability. Symptoms of Fragile X syndrome include ear, nose, and throat problems (ear infections or sinusitis), developmental delays, motor dysfunction, and the physical appearance of a long face with prominent forehead and protrusile ears.^[32] These symptoms typically do not display until early childhood. Roughly 50% of individuals with Fragile X meet the diagnostic criteria for ASD and tend to have more severe developmental and behavioral problems.^[33]

Rett syndrome is a genetic disorder that occurs predominantly in girls. The first signs of Rett syndrome are a failure to meet development milestones, usually between 6 and 18 months of age. Active regression occurs between 1 and 4 years of age, in which there is a loss of motor function and language skills. Eventually, this loss plateaus, although some individuals experience improvements in ASD traits through early adulthood. About 50% of people with Rett syndrome meet the criteria for ASD during the regression period, but this decreases with time.^[34]

Men and boys are more likely to be diagnosed with ASD than women and girls.^[17] There are a variety of explanations for this discrepancy — some biological; some methodological.

Genetics and prenatal androgen exposure have both been implicated in differing rates of autism between sexes. It has been hypothesized that the paternal X chromosome may play a role in protecting women and girls from developing ASD ^[18]. During fetal development, one X chromosome in female babies’ cells expresses its 1,100 genes while the other is inactivated. The inactivation of this X chromosome makes it less likely for mutated genes to be expressed in developing female cells.^[19][20]

Some researchers propose that women and girls who were exposed to excess androgens in utero are at greater risk of developing autistic traits,^[21][22] but this is controversial.^[23]

It’s also possible that the current assessment tools for ASD are less effective at capturing features of ASD that are unique to women and girls (a phenomenon known as “diagnostic bias”).^[24] For example, women and girls with ASD are more likely to camouflage or mask^[25] their behaviors, meaning that they may adopt a socially acceptable persona by copying facial expressions and purposefully making eye contact. Since social behavior, eye contact, and facial expressions all factor in to an ASD diagnosis, someone with ASD who masks their behaviors may not be diagnosed as having the condition.

Two diagnostic criteria are frequently used by clinicians to make an ASD diagnosis: the Diagnostic and Statistical Manual of Mental Disorders-5^[16](DSM-5) and the International Disease Classification (ICD). Because both criteria are similar, this section will focus only on the DSM-5 criteria. Keep in mind that, no matter which criteria are used, the clinical manifestation of ASD will vary across individuals.

According to the DSM-5, a diagnosis of ASD requires the presence of persistent difficulty with social communication and interaction in all three of the following categories: social-emotional reciprocity (e.g., failing to engage in mutually agreeable back-and-forth conversation, difficulty initiating or responding to social interactions); nonverbal communication behaviors (e.g., poor eye contact, lack of facial expressions, misunderstanding use of gestures); and developing, maintaining, and understanding relationships (e.g., difficulty adjusting behavior to suit various social contexts). The DSM-5 diagnosis also requires the presence or history of at least two restricted and repetitive behavior patterns, interests, or activities. These include repetitive movements, inflexible adherence to routines, fixated interests, and hyper-reactivity or hypo-reactivity to sensory input.

Any diet that decreases food variability has the potential to increase the risk of nutrient deficiencies — if it is not well planned. With a GFCF diet, nearly all dairy, many grains, and grain-based products (wheat, barley, and rye) will be eliminated. The removal of dairy and gluten-containing foods can subsequently decrease the intake of vitamins and minerals these products contain naturally and are typically fortified with. Namely calcium, fiber, and the B vitamins.

While it may seem simple for an adult to just get wheat or dairy associated nutrients from other foods, nutrient intake issues can be tricky in developing children, who typically don’t make their own eating decisions, and especially children with specific health conditions. One study found that boys with ASD^[11] had lower bone mineral density and vitamin D intake from food as well as measured in serum. With the elimination of dairy, particular attention is needed to ensure adequate intakes of calcium and vitamin D are being consumed to maintain bone health.

Narrow food preferences and avoidant-restrictive food intake disorder (ARFID) are common among young children with ASD, as many have oral sensory sensitivity ^[51][52]. This puts these children at a greater risk for nutrient deficiencies.^[53][54]. Thus, elimination or restricted diets must be carefully planned. To date, the evidence surrounding elimination-style diets (e.g., the gaps-diet, gluten-free/casein-free diet, and spec-carb-diet) for improving the core features of ASD have either not been researched or are inconclusive.^[55][56]

Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized by impaired social communication and repetitive behavior patterns. While the exact pathology of ASD is unknown, there may be a connection between diet and ASD behavioral symptoms.

Many dietary supplements have been investigated as potential treatments, including omega-3 fatty acids,^[5] melatonin,^[6] vitamin-d,^[7] and a combination of vitamin-b6 and magnesium;^[8] but the gluten-free and casein-free (GFCF) diet has become one of the more popular interventions among parents with ASD children.

Survey have shown that anywhere from 20% to 70% of respondents^[57] have tried a GFCF diet. Parents often report symptom improvement when placing their children on this diet,^[58] but what do results from randomized controlled trials (RCTs) tell us?

Many families try decreasing autism spectrum disorder (ASD) symptoms with a gluten-free, casein-free (GFCF) diet. Several trials have been conducted to examine this connection.

What is the theory behind the GFCF diet?

The opioid-excess theory of ASD has long been a popular hypothesis for explaining how a GFCF diet may alleviate ASD behavioral symptoms.^[59] It has three main components:

1. The incomplete breakdown of the proteins making up gluten and casein can form excess opioid peptides.

2. These peptides can enter the body through an abnormally permeable intestinal border, which is speculated to be more common in people with ASD.^[60]

3. If produced in sufficient quantities, these peptides could theoretically cross the blood–brain barrier and interfere with normal brain development.^[60]

This hypothesis, however, is being disputed. Some studies did report greater gut permeability in people with ASD,^[61][62] but others saw no difference.^[63] Moreover, highly sensitive measurement techniques consistently failed to find detectable concentrations of opioid peptides in the urine samples of patients with ASD.^[64][65] If significant amounts of opioid peptides were making it past the gut and into the bloodstream, urine tests should reveal their high levels as the body worked to eliminate them.

According to the opioid-excess theory, the incomplete breakdown of gluten or casein (two proteins) can form opioid peptides that may exacerbate ASD symptoms. However, the literature does not consistently support this hypothesis.

What do GFCF studies show?

Since the 1970s, more than 30 trials have tried to ascertain the role a GFCF diet could play in ASD therapy, but many suffered from poor methodological quality.^[57][66]

The trials often

• Were short in duration.
• Had small sample sizes.
• Lacked a control group.
• Were single-blinded or not blinded at all.
• Performed no power calculation.

Moreover, changes in behavior were usually reported only by the parents, yet in most trials, the parents knew which group (placebo or GFCF) their child was in. This knowledge introduced possible bias into the studies, since the parents who believed that the GFCF diet would benefit their child were more likely to report positive results even if there were none. Unconscious bias such as this can lead to false positives in a study’s findings.

To reduce the risk of bias, we’ll examine the most rigorously controlled GFCF trials currently available: those that are double-blinded, randomized, and placebo-controlled. At present, five such trials have been conducted.

2006 Study

The gluten-free, casein-free diet in autism: results of a preliminary double-blind clinical trial.^[67]

Standard assessment questionnaires were administered, such as the Childhood Autism Rating Scale (CARS) and the Ecological Communication Orientation Scale (ECOS). Both scales monitor the frequency of behavioral patterns, such as social initiating, social responding, intelligible words spoken, and non-speech vocalizations.

Additionally, research assistants videotaped the child interacting with their primary caretaking parent. Two independent coders, who were blinded to the dietary treatment status of the child, rated each taped assessment. No significant differences were observed for any of the behavioral endpoints measured between the GFCF diet and the control diet.

2007 Study

The gluten- and casein-free diet and autism: communication outcomes from a preliminary double-blind clinical trial.^[68]

This study is a retrospective analysis of the “2006 Study” above. Its authors sought to perform a more in-depth analysis of the verbal and nonverbal communication outcomes than the one mediated by the ECOS scale, used in the original study. With this aim in mind, they re-evaluated the existing videotapes of parent-and-child interactions, with a focus on verbal communication. Again, however, no significant changes were observed between the GFCF diet and the control diet.

2014 Study

Are “leaky gut” and behavior associated with gluten- and dairy-containing diet in children with autism spectrum disorders?^[69]

For the first 2 weeks, all participants followed a GFCF diet. For the remaining 4 weeks, they were randomized to the intervention group (brown rice flour) or the control group (gluten powder and non-fat dried milk). Blinded to the group a child was in, parents and investigators assessed hyperactivity, irritability, and inattention. No significant behavioral differences were observed between groups.

2015 Study

Gluten and casein supplementation does not increase symptoms in children with autism spectrum disorder.^[70]

Researchers used the Approach Withdrawal Problems Composite (AWPC) subtest of the Pervasive Developmental Disorder Behavior Inventory (PDDBI) to monitory maladaptive behavior before and after supplementation. In both groups, the AWPC score decreased significantly after supplementation. But the researchers noted that “the change in the degree of maladaptive behavior was not significantly different between the two groups (p = 0.971)” (emphasis added).

2016 Study

The gluten-free/casein-free diet: a double-blind challenge trial in children with autism.^[71]

The main conclusion of this study was that the GFCF diet did not change measures of “physiologic function, behavioral disturbance (sleep disruption and over-activity), or ASD-related behaviors”. The researchers also examined the children’s individual data to see if any individual results were being masked by the overall group result, but again found no clear pattern suggesting that the dietary challenge had had any effect, positive or negative, on any of the behavioral outcomes measured. A subset of children actually experienced fewer negative social relationship symptoms on the days gluten and casein were co-administered, but this trend never reached statistical significance.

Conclusion

Five double-blind, randomized, placebo-controlled trials examining a total of 89 children have found that a GFCF diet did not improve ASD behavior symptoms.

Is GFCF the best therapy for ASD? Unlikely.

Two recent systematic reviews^[72][73] and a consensus report by the American Academy of Pediatrics^[74] have concluded that cutting out gluten and casein isn’t likely to help in the treatment of ASD.

Anecdotal reports of improvement could be due to casein- and gluten-free diets accompanying some other treatment or some healthy lifestyle habits. The potential benefits of other, more comprehensive diet changes have not been rigorously tested.

Given the current lack of evidence that a GFCF diet benefits people with ASD, emphasis on behavioral therapies is likely to provide greater benefit.^[75][76]

A GFCF diet is unlikely, at least one its own, to improve the behavior of children with ASD. Time, energy, and resources may be better spent pursuing other treatment options.

Genetics, epigenetics, and environmental factors are all implicated in the development of ASD.^[10]

From a genetic standpoint, ASD is associated with a number of identifiable genetic variants (e.g., polygenic variants, single nucleotide variants, noncoding variants), and as many as 102 genes have been identified as ASD risk factors.^[35][36]

Maternal/paternal health and toxic exposure are also associated with the development of ASD. Such exposures may modulate the expression of the genetic factors that place an individual at higher risk for ASD. Examples include maternal infections (particularly in the second and third trimesters of pregnancy),^[37] prenatal exposure to valproate^[38] (an anti-seizure drug), older maternal and paternal age,^[39][40][41] maternal hypertension, ^[42] maternal obesity,^[43][44] neonatal jaundice,^[45] elevated maternal c-reactive protein,^[46] and maternal polycystic ovary syndrome^[47], all of which have been associated with an increased risk of ASD in offspring.

An often-discussed potential risk factor for ASD is vaccination, whether that be the vaccine itself (e.g., MMR) or the vaccine constituents (e.g., thimerosal). There is no evidence to support this association, and multiple epidemiological studies and systematic reviews show no causal relationship between vaccination and ASD.^[48][49][50]