In the 1920s, researchers had not yet determined the cause of celiac disease (CD), and the doctors who developed the Specific Carbohydrate Diet (SCD) in 1924 doubted the recently proposed “gluten theory", which implicated the wheat protein called gluten in the development of CD. They surmised, instead, that people with CD were intolerant to all disaccharide and polysaccharide sugars. 
Based on this theory, the SCD consisted primarily of ripe bananas and calcium caseinate, which were — according to Sydney V. Haas, MD — tolerable due to their low di- and polysaccharide content and provided adequate calories and protein to treat CD-associated weight loss. Certain foods could be reintroduced gradually, with tolerance indicated largely by bowel habits. The diet was intended to last 12 months, at which point the excluded carbohydrates could be gradually reintroduced. From the mid-1920s through the mid-1950s, Dr. Haas and his associate (another Dr. Haas — his son, Merrill P.) treated about 800 patients with the SCD. 
The diet was repopularized in the 1990s by Elaine Gottschall, a biochemist (and, purportedly, the mother of one of Sydney Haas’s patients). After her daughter was diagnosed with CD, Gottschall wrote a book about the SCD, claiming it could cure CD as well as other conditions, such as IBD, diverticulitis, and autism spectrum disorder (ASD). However, these claims have not been substantiated. 
In addition to bananas and low-lactose milk, moderate quantities of fat and protein were allowed, as was honey, but table sugar was to be avoided. Although other fruits could be introduced later on, it was thought that they could worsen diarrhea if given too early; eggs and vegetables were to be reintroduced gradually once diarrhea had subsided. Potatoes, corn, rice, wheat, rye, table sugar, corn syrup, and whole milk — though not sour cream — were permanently excluded. Early publications reported that most children tolerated starch after 12–18 months on the diet, and patients could return to a normal diet as long as their symptoms didn’t recur. 
In Gotschall’s version of the diet, most fresh fruits and vegetables, certain legumes, unprocessed meats, low-lactose dairy products, and honey are permitted. Potatoes, yams, canned fruits and vegetables, milk, and processed foods are excluded. 
Despite numerous anecdotal reports of improvement in Crohn’s disease, ASD, and inflammation, most of the SCD research comes from observational data and case studies. Early research using case studies and retrospective surveys failed to collect dietary data, and many patients were also taking medications to manage their IBD. A majority of patients reported occasionally eating restricted foods — or even reintroducing them entirely — but the majority also reported that their symptoms improved after implementing the SCD.
One prospective study followed children for up to 1 year after starting the SCD, and researchers observed healing of the intestinal lining in some participants, but others failed to improve, and some even worsened. That same year, a small retrospective study found no signs of mucosal healing (healing of the innermost layer of the bowel) in children who had been reported to be using the SCD. Retrospective chart reviews of patients following the SCD suggest that strict implementation of the diet is associated with symptom improvement, but they also highlight the difficulty that patients experience in adhering to the diet.
The SCD and other restrictive diets have been suggested as treatments for ASD, despite the lack of supporting evidence from placebo-controlled trials. Research on the SCD in ASD is currently limited to one 16-week case study and an online survey administered to 256 caregivers of children with ASD.
Surprisingly, despite the lack of evidence supporting the use of the SCD in children with ASD, half of the caregivers reported starting their children on the diet after a practicing physician recommended it. Additionally, 40% reported receiving guidance from dietitians or nutritionists trained in SCD application. Caregivers reported starting the children on the SCD at an average of about 5½ years old and following it for an average of 2.7 years. Prior to the SCD, 64% of the children were following gluten-free, casein-free diets (which have been studied more extensively and currently appear ineffective.) 
A large majority of the caregivers reported improvements in the children’s abdominal pain, constipation, diarrhea, and food variety, as well as self-injurious behavior, irritability, anxiety, toilet training, and language skills after SCD implementation. However, the questionnaire used in this study has not been validated for research; dietary intake data was never collected; and the findings weren’t controlled for the use of medications, time spent on the SCD, or the presence of IBD in some of the children. Likewise, the child in the 16-week case study also improved, but the researchers can’t rule out other explanations for the changes.
An exploratory, open-label study assigned 22 children with juvenile idiopathic (from an unknown cause) arthritis to the SCD for 4 weeks with no control group. The dropout rate was also high in this case, with only 15 children completing the intervention. They reported improvements in morning stiffness, pain, and physical function, and several of their inflammatory proteins decreased by the end of the trial. The short-chain fatty acid butyrate increased, as well, which is helpful for colon function.
So far, just two randomized trials (RCTs) have been performed in people with Crohn’s disease. Though preliminary findings suggest improvements in some markers of inflammation, a lack of quality RCTs makes it difficult to determine the efficacy of the SCD. 
In a 12-week RCT that enrolled nearly 200 participants with mild to moderate Crohn’s disease, researchers compared the effects of the SCD and a Mediterranean-style diet (MD). Both groups ate more fruits and vegetables and reported improvements in fatigue, pain, sleep quality, and social isolation. Though both groups also reported similar rates of remission (either lessening or going away entirely) of symptoms (46.5% in the SCD and 43.5% in the MD), their intestinal and systemic inflammatory markers didn’t change after either diet. Given the comparable effects of these diets, the researchers concluded that the MD may be a superior intervention, due to it being easier to follow. This study experienced a high dropout rate (30 participants from the MD group and 35 from the SCD group), and participants reported that they were unlikely to adhere to a restrictive diet for longer than 6 weeks.
Another 12-week trial randomized 18 pediatric patients with active Crohn’s disease to either a strict SCD, modified SCD (which included oats and rice), or whole-foods diet (which excluded only wheat, corn, sugar, milk, and food additives). There were high dropout rates in this study also; only 10 of the participants completed it. But researchers reported that all 10 achieved clinical remission and reductions in intestinal and systemic inflammation. 
The SCD is no more effective than the MD or a whole-foods diet for people with Crohn’s disease. The SCD is not currently a scientifically supported intervention for ASD, and more research is needed to determine its efficacy in juvenile idiopathic arthritis. Since participants knew about their dietary changes, and none of these studies included a control group, the results could have been influenced by the placebo effect and other changes associated with a new diet.
Though the Haases didn’t intentionally restrict gluten (or even believe it played a role in CD), they coincidentally eliminated it from their patients’ diets by removing grains. This early gluten-free diet could have been effective for the patients with CD, but formal clinical trials were never performed. Because the SCD eliminates grains and certain starchy foods, it may also reduce the FODMAP (fermentable oligosaccharides, disaccharides, monosaccharides, and polyols — carbohydrates that are poorly absorbed in the small intestine — ) content, which could explain some of the gastrointestinal (GI) improvements. However, the FODMAP content of the diet has not been studied.
Limited, low-quality evidence suggests that the SCD is as effective as the Mediterranean diet for improving symptoms of Crohn's disease, but it is not currently a scientifically supported intervention for autism spectrum disorder.
Individuals with IBD may already be at risk of malnutrition due to reduced nutrient absorption and dietary restrictions, while children with ASD may be at increased risk due to extremely selective food preferences.  Restrictive diets, such as the SCD, could exacerbate this by reducing their food options.
One small, prospective study in children with mild or moderate IBD analyzed their 3-day food records to determine the nutritional content of the SCD. Though their energy needs were met, most of the children’s intakes of calcium and vitamin D were much lower than the RDA, so the researchers acknowledged the importance of supplementation while following the SCD.
The diet may also be hard to sustain. Many participants dropped out of studies or reported difficulty adhering to the diet. Dropout rates in the RCTs ranged from about 20 to 40%, and the surveyed caregivers reported spending 6–9 hours per week on food preparation for the children’s SCD.