Diet and autism: What the science says

Autism is a highly controversial subject with a very vocal lobby group, and one expects differing viewpoints.  But when some professionals claim that the available research shows no indication that diet or parenting causes the disorder, this misconception needs to be corrected.

It is critical to note up front that we cannot obtain causal proof for lifestyle choices influencing the development and severity of autism (however we define autism, which is another important and difficult question) without performing unethical and illegal experiments on mothers and children.  In the absence of this data, what we must rely upon are correlations and associations that, while not proof of causation, are definitely strongly suggestive.  And the scientific literature is replete with such examples, a select group of which are presented below.

As part of a study into the "dietary patterns and body mass index in children with autism and typically developing children," a team from Tufts University, The Ohio State University, the University of Massachusetts, and Boston University reported the following:

Children with ASD [autism spectrum disorder] were found to consume significantly more daily servings of sweetened beverages and snack foods and significantly fewer daily servings of fruits and vegetables than typically developing children ... two studies show that children with ASD have a preference for starchy foods and energy-dense foods such as chicken nuggets, hot dogs, cake, French fries, macaroni, pizza, and ice cream over other food groups. Our results are consistent with these findings, in that children with ASD were found to have more energy dense dietary intake patterns compared to typically developing children. On average, children with ASD consumed significantly more juice and sweetened non-dairy beverages and snacks and significantly fewer servings of vegetables than typically developing children.

In a study of "food preferences and factors influencing food selectivity for children with autism spectrum disorders," researchers from Penn State University-Harrisburg and the Hershey Medical Center came to the following conclusions:

[I]t was found that food selectivity of children with autism was related to their family's reported eating preferences. Specifically, the fewer food items the family reportedly ate -- the fewer food items the child ate ... In this study, when the typical food menus were examined according to food group, food items preferred by 50% or more of the children with autism were very rare. The actual food items included foods that are high in sugar content, such as cakes, cookies, grapes, pizza, white bread, ice cream, etc ... the results of this study may indicate that children with autism consume food items that result in quick 'sugar highs'.

A recent review article by a group from the University of Pennsylvania looked at "eating behaviors, diet quality, and gastrointestinal symptoms in children with autism spectrum disorders" and discussed these issues at great length.  The article clearly shows extensive linkages between diet and autism, some of which are highlighted below:

Two of the most common problems in children with ASD are difficult eating behaviors and gastrointestinal (GI) symptoms. As early as 1979, mealtime difficulties, such as increased food selectivity (e.g., picky eating), food refusal, and disruptive mealtime behaviors were reported in as many as 90% of children with ASD. More recently, cross-sectional data from the 2003-2004 National Survey of Children's Health indicate that children with ASD were 40% more likely to be obese compared to typically developing children (TDC) ... Mealtime difficulties not only put children with ASD at increased risk for nutritional deficiencies, but they also put a strain on parents and caregivers.

Findings from a longitudinal study indicated that, compared to TDC, children who were subsequently diagnosed with ASD, were more likely to be described by their parents as "slow feeders" at 6 months of age and were reported to have difficulty transitioning to solid foods. As children grew older (15 to 54 months of age), children with ASD were described as "difficult to feed" and "very choosy" eaters. They also showed a significantly greater prevalence of pica (i.e., craving for and consumption of nonfood substances) at 38 and 54 months of age compared to TDC. Eating difficulties in children with ASD are of concern because they not only increase caregiver stress, but they also put children with ASD at a greater risk for nutritional deficiencies, which may adversely affect their growth and development ...

Similarly, another study showed that, based on parental report using a modified food frequency questionnaire, children with ASD refused significantly more foods as a percentage of those offered than did TDC (42 versus 19%). This finding applied especially to vegetables ...

When using 3-day food records to examine 3- to 11-year- old children's dietary intake, data showed that inadequate intakes of vitamin D, calcium, and vitamin A are more common in children with ASD than in TDC ... Data from another study ... showed that selective eaters with ASD were at a greater risk for inadequate nutrient intakes than non-selective eaters with ASD and TDC. Specifically, selective eaters with ASD were at an increased risk for developing nutrient deficiencies ... for calcium, zinc, vitamin D, and vitamin B12 ... A study in Egyptian children showed that children with ASD had significantly lower serum vitamin D levels and significantly lower calcium levels compared to TDC, further pointing to potential nutritional deficiencies in children with ASD ... findings from another study showed that children with ASD had significantly more essential plasma amino acid deficiencies than TDC (58 versus 4%), which is suggestive of poor protein nutrition among children with ASD ...

GI problems, such as abdominal pain, constipation, diarrhea, gastrointestinal reflux, are common in children with ASD ... Together these data suggest that children with ASD may show alterations in the bacterial composition of the gut flora.

A study with 90 children with ASD showed that abnormal intestinal permeability was present in 37% of children with ASD compared with 21% first-degree relatives, 5% adult controls, and none of the child controls. This suggests that a subgroup of children with ASD may indeed show a "barrier function deficit" and may therefore benefit from a gluten-free diet, however, more systematic research is needed on this topic.

Similar dietary concerns were found in a study by researchers from the University of Newcastle and The Ohio State University:

Children exhibiting more severe autism-specific disruptive behaviors at mealtimes were most at risk for suboptimal intake of select nutrients such as biotin, vitamin K, iodine, linolenic omega-3 fatty acids, and choline, which play a role in metabolism and bone and brain health ... children with ASD and specific problem eating behaviors display suboptimal daily nutrient intake.

Work done at the State University of New York reported on some clear parenting choices and autism:

Many ... studies link ... in utero smoking exposure to higher levels of autism ... much recent research has suggested that nutritional ... factors also play a major role in the development of autism and ADHD disorders ... A study of 861 autistic children and 123 control children found that feeding infant formulas lacking added docosahexaenoic acid (DHA) and arachidonic acids was associated with a 4.41 increased odds of developing autistic spectrum disorder as compared to breastfed infants ... Some studies have reported that nutritional interventions can significantly help many patients with autism and/or ADHD ...

Finally, a substantive review on "how nutritional status, diet and dietary supplements can affect autism" by a group from Wroclaw Medical University in Poland concluded the following:

Such an extensive array of symptoms in the digestive tract in patients with ASD, strongly suggests the need for introducing nutritional therapy in addition to routine treatment. Adequate nutrition can lead to relief of symptoms, both digestive and metabolic as well as psychological ones. Unfortunately, still very few doctors and nutrition experts monitor the diet and nutrition of autistic patients ...

In a case-controlled study, Krakowiak et al. evaluated whether the incidence of metabolic disorders during pregnancy, (such as: diabetes, hypertension and obesity) is associated with the prevalence of autism risk in children. It was concluded that this risk and other developmental disorders could increase if the mother is obese or suffers from type 2 diabetes. It was found that the incidence of the latter and gestational diabetes was higher in mothers who bore a child with ASD. Furthermore, the incidence of hypertension and obesity were respectively higher in mothers of children with ASD compared to normal ...

Ho and Eaves analysed differences related to nutritional status and diet of Canadian children, aged on average 13.3 years, with ASD ... 42.6% of the examined children were obese of whom almost half demonstrated low physical activity levels. The study showed that obesity may be correlated with increased symptoms of autism.

This is but a small sampling of the literature on autism, and it reveals a number of key points.

While researchers acknowledge the "chicken and the egg" problem (i.e., identifying the true causation), it appears highly likely that dietary choices strongly influence the development and severity of autism among many children.  As well, lifestyle/parenting choices before and during pregnancy (e.g., smoking, obesity, etc.) result in significantly higher risks of having autistic children.

Thus, we are left with the following conclusion.  For many children, if the parents would have enforced a healthier diet and general lifestyle during pregnancy and after birth (i.e., better parenting), either the onset of autism could have been avoided or the severity of the effects could have been greatly reduced.

If you experience technical problems, please write to