D'Adamo's Blood Type Diet has met with several criticisms. The fundamental criticisms are, for one, that none of his hundreds of citations to others' research on blood groups directly support his claims of differential food tolerances and, secondly, that he provides no comparative clinical trials demonstrating efficacy of his diet.
One criticism of D'Adamo's hypotheses and recommendations claims that he provided inadequate evidence. For example, his first book, Eat Right 4 Your Type, published in 1997, contains only a bibliography. Most of his subsequent books, however, have been thoroughly referenced as far as his general theory. However, despite his providing general reasons for the classifications of various foods within his established categories of "beneficials", "neutrals" and "avoids", his specific process for reaching these conclusions of classification remain undocumented.
Although D'Adamo claims there are many ABO specific lectins in foods, this claim is, for a number of his cited cases, unsubstantiated by established biochemical research, which has not found differences in how the lectins react with a given human ABO type. A common criticism is that lectins which are preferential for a particular ABO type are not found in foods (except for one or two rare exceptions, e.g. lima bean), and that lectins with ABO specificity are more frequently found in non-food plants or animals.
Another criticism is that there are no clinical trials of the Blood Type Diet. In his first book Eat Right 4 Your Type, D'Adamo mentions being in the eighth year of a 10 year cancer trial, but no results of this trial have ever been published. In his book Arthritis: Fight It With the Blood type Diet, D'Adamo mentions an impending clinical trial of the Blood Type Diet in order to determine its effects on the outcomes of patients with rheumatoid arthritis, but no results of this trial have yet been published.
In the article "Genetic of the ABO blood system and its link with the immune system", Luiz C. de Mattos and Haroldo W. Moreira point out that D'Adamo's assertion that the O blood type was the first human blood type requires that the O gene evolved before the A and B genes in the ABO locus. Instead, phylogenetic networks of human and non-human ABO alleles show that the A gene was the first to evolve. The authors argue that, in the evolutionary sense, it would be extraordinary for normal genes (those for types A and B) to have evolved from abnormal genes (for type O).
Yamamoto et al. further note:
Although the O blood type is common in all populations around the world, there is no evidence that the O gene represents the ancestral gene at the ABO locus. Nor is it reasonable to suppose that a defective gene would arise spontaneously and then evolve into normal genes.
In May 2004, Transfusion published a study which concluded that: "Assuming constancy of evolutionary rate, diversification of the representative alleles of the three human ABO lineages (A101, B101, and O02) was estimated at 4.5 to 6 million years ago." This finding declares that ABO did not evolve in the near past, essentially contradicting that which D'Adamo suggests.