Blood Type Diet Research Paper

ABSTRACT

Background: Diets that are based on the ABO blood group system have been promoted over the past decade and claim to improve health and decrease risk of disease. To our knowledge, the evidence to support the effectiveness of blood type diets has not previously been assessed in the scientific literature.

Objective: In this current systematic review, published studies that presented data related to blood type diets were identified and critically appraised by using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach.

Design: A systematic search was performed to answer the following question: In humans grouped according to blood type, does adherence to a specific diet improve health and/or decrease risk of disease compared with nonadherence to the diet? The Cochrane Library, MEDLINE, and Embase were systematically searched by using sensitive search strategies.

Results: Sixteen articles were identified from a total of 1415 screened references, with only one article that was considered eligible according to the selection criteria. The identified article studied the variation between LDL-cholesterol responses of different MNS blood types to a low-fat diet. However, the study did not directly answer the current question. No studies that showed the health effects of ABO blood type diets were identified.

Conclusions: No evidence currently exists to validate the purported health benefits of blood type diets. To validate these claims, studies are required that compare the health outcomes between participants adhering to a particular blood type diet (experimental group) and participants continuing a standard diet (control group) within a particular blood type population.

INTRODUCTION

The Blood Service of Belgian Red Cross–Flanders occasionally receives inquiries regarding the validity of diets that are based on blood group systems, which are often endorsed as a program to improve health. To our knowledge, evidence for these claims has not been substantiated in the scientific literature, and this current review is the first systematic review performed on the topic.

Blood can be classified according to the presence or absence of certain antigens on the surface of red blood cells, and these antigens are controlled at specific gene loci (1). Relevant to this study are the following 2 of ∼30 recognized blood grouping systems: the ABO system (the most important antigens being A, B, and H) on chromosome 9 and the MNS system (the most important antigens being M, N, S, s, and U) on chromosome 4 (2). ABO blood typing is typically correlated with blood transfusions because ABO blood product incompatibility can potentially prove fatal. More recently, an extensive collection of epidemiologic studies have assessed the significance of ABO status in relation to physiologic variations and pathologic processes. From among the varied results (with some consistent and some inconsistent findings), the use of genome-wide association studies have supported a number of associations between ABO blood type and certain diseases, including pancreatic cancer, venous thromboembolism, and myocardial infarction in the presence of coronary atherosclerosis (3). Therefore, it appears possible that the ABO blood group system plays a role in determining an individual’s susceptibility to certain diseases.

This established association between blood types and disease has been translated into the basis for a range of diets. Of the many authors of blood type diets (4–8), D’Adamo is arguably the most prolific (9–23). Within his initial ABO blood type diet book entitled Eat Right 4 Your Type (9), which was published in 1996, D’Adamo claims that each ABO blood type processes food differently, and adherence to a diet specific to an individual’s ABO blood group could improve health, wellbeing, and energy levels and reduce risk of developing diseases such as cancer and cardiovascular disease. D’Adamo diets, one for each ABO blood type, are based on a theory that each blood type contains the genetic message of the diets and behaviors of our ancestors, and these traits still have an impact on us today. The validity of the scientific basis for any of the blood type diets is beyond the scope of this systematic review, which is solely focused on the related health outcomes of adherence to the prescribed diets.

Considering the substantial reach of the blood type diets [eg, there are >7 million copies (24) of D’Adamo’s 1997 book in print in >60 languages (25)], it seemed pertinent to be able to substantiate the health claims of blood type diets so that inquiries to blood services, physicians, and dietitians can be adequately addressed.

METHODS

Search strategy

The population, intervention, comparison, and outcome (PICO) question was as follows: In humans grouped according to blood type (population), does adherence to a specific diet (intervention) improve health and/or decrease risk of disease (outcome) compared with nonadherence to the prescribed diet (comparison)?

All searches were performed on 3 October 2012 within the following databases: The Cochrane Library (via Wiley: http://onlinelibrary.wiley.com/cochranelibrary/search), MEDLINE (http://www.ncbi.nlm.nih.gov/pubmed), and Embase (http://www.embase.com/). All databases were searched from inception to ensure that all relevant articles would be retrieved.

The development of search strategies was performed independently by 2 reviewers (LC and EDB). Final search strategies were combined to ensure maximal search sensitivity before the commencement of study selection. The subsequent study selection was performed independently by the same 2 reviewers in 2 phases. Initially, the titles and abstracts identified by the search were scanned for relevance to the PICO question. Full texts of any relevant references were obtained to assess whether exclusion and inclusion criteria were met. Independent results from the selection phase were compared to ensure that all relevant studies had been identified. See the online Appendix under “Supplemental data” in the online issue for search strategies.

Finally, references from identified studies, along with other articles that cited the identified studies, were assessed for relevance. The first 20 relevant studies (as suggested by MEDLINE) were also assessed for pertinence to the current review.

Study selection: inclusion and exclusion criteria

Language

All languages were included.

Study type

All experimental and observational studies were included (ie, randomized controlled trials, controlled clinical trials, cohort studies, case-control studies, and case series). In vitro studies were excluded, along with narrative reviews, commentaries, letters, and opinions.

Population

Studies were included if they contained humans grouped according to a blood group system. Animal studies were excluded.

Intervention

Adherence to a prescribed diet that was designed to improve health was necessary for inclusion. Any studies with another intervention or no intervention were excluded.

Outcome

A decreased incidence of disease, improved BMI, and any other quantified measure of health were included.

Data collection

Relevant information, including the study design, study population, sample size, and details of the intervention and outcomes, was extracted from the identified study (by LC).

Quality of evidence

The Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach was used to grade the overall quality of evidence identified in this systematic review. The GRADE approach took into consideration any limitations of included studies that may have had an effect on the quality of the evidence including the study design (lack of allocation concealment and blinding, incomplete accounting of patients and outcome events, selective outcome reporting, and other limitations), inconsistency between different studies (because of differences in populations, interventions, or outcomes), indirectness (of the population, intervention, or outcome), imprecision of outcomes and publication bias (which can be difficult to determine when limited evidence is available). After the appropriate downgrading for each of the previously mentioned criteria (or upgrading in certain circumstances), the quality of the evidence was established as either high (A: further research is very unlikely to change our confidence in the estimate of effect), moderate (B: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate), low (C: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate), or very low (D: any estimate of effect is very uncertain) (26).

RESULTS

The following 1415 articles were identified: 6 articles from The Cochrane Library, 639 articles from MEDLINE, and 770 articles from Embase. Once duplicates were removed, a total of 1003 articles were screened by title and abstract for relevance to the PICO question. The remaining 16 articles were reviewed to assess inclusion and exclusion criteria. Reasons for inclusion or exclusion of each identified article are shown in Table 1.

TABLE 1

Assessment of each identified reference according to inclusion and exclusion criteria1

Reference Study type Population Intervention Outcome Meets criteria? 
No author listed, 2000 (27) Narrative review NA NA NA No 
Baldwin, 2004 (28) Narrative review NA NA NA No 
Birley et al, 1997 (29) Experimental Humans grouped according to MNS blood group system Low-fat diet Quantitative measurements of reduction in LDL cholesterol Yes 
Cowley and King, 1997 (30) Opinion piece NA NA NA No 
Langman et al, 1966 (31) Experimental Humans grouped according to ABO blood group system Fatty breakfast followed by a small bar of milk chocolate2Serum concentrations of intestinal alkaline phosphatase3No 
Langman et al, 1967 (32) Narrative review NA NA NA No 
Larhammar, 2005 (33) Opinion piece NA NA NA No 
Mäkivuokko et al, 2012 (34) Observational Humans grouped according to ABO blood group system No intervention Determination of microbiota profiles No 
McGowan, 2001 (35) Letter NA NA NA No 
Meltzer et al, 2002 (36) Narrative review NA NA NA No 
Moen, 2001 (37) Narrative review NA NA NA No 
Mysterud, 2002 (38) Letter NA NA NA No 
Nylund et al, 2004 (39) Narrative review NA NA NA No 
Poleszynski, 2001 (40) Narrative review NA NA NA No 
Power, 2007 (41) Experimental Humans grouped according to ABO blood group system Food-allergy tests (mRAST and ELISA/ACT LRA tests) Quantitative measurements of IgG, IgE, and T cells No 
Schroder, 2005 (42) Narrative review NA NA NA No 
Reference Study type Population Intervention Outcome Meets criteria? 
No author listed, 2000 (27) Narrative review NA NA NA No 
Baldwin, 2004 (28) Narrative review NA NA NA No 
Birley et al, 1997 (29) Experimental Humans grouped according to MNS blood group system Low-fat diet Quantitative measurements of reduction in LDL cholesterol Yes 
Cowley and King, 1997 (30) Opinion piece NA NA NA No 
Langman et al, 1966 (31) Experimental Humans grouped according to ABO blood group system Fatty breakfast followed by a small bar of milk chocolate2Serum concentrations of intestinal alkaline phosphatase3No 
Langman et al, 1967 (32) Narrative review NA NA NA No 
Larhammar, 2005 (33) Opinion piece NA NA NA No 
Mäkivuokko et al, 2012 (34) Observational Humans grouped according to ABO blood group system No intervention Determination of microbiota profiles No 
McGowan, 2001 (35) Letter NA NA NA No 
Meltzer et al, 2002 (36) Narrative review NA NA NA No 
Moen, 2001 (37) Narrative review NA NA NA No 
Mysterud, 2002 (38) Letter NA NA NA No 
Nylund et al, 2004 (39) Narrative review NA NA NA No 
Poleszynski, 2001 (40) Narrative review NA NA NA No 
Power, 2007 (41) Experimental Humans grouped according to ABO blood group system Food-allergy tests (mRAST and ELISA/ACT LRA tests) Quantitative measurements of IgG, IgE, and T cells No 
Schroder, 2005 (42) Narrative review NA NA NA No 

View Large

From within the 16 articles, 4 articles met the inclusion criteria of both study design and population; 2 of these articles fulfilled the outcome inclusion criteria, and only one of the articles met the inclusion criteria for the intervention (ie, adherence to a diet that was designed to improve health). A flowchart of the study selection process is shown in Figure 1.

Ultimately, only one of the 16 identified papers fulfilled all of the inclusion criteria; Birley et al (29) studied the effects of a low-fat diet on LDL-cholesterol concentrations of study participants grouped according to MNS blood type. None of the studies showed an association between ABO blood type diets and health-related outcomes

The identified study analyzed a total of 254 participants [127 subjects within experimental groups (MN genotype: n = 67; MM genotype: n = 38; NN genotype: n = 22) and 127 subjects within control groups (MN genotype: n = 61; MM genotype: n = 40; NN genotype: n = 26)]. Intervention groups were given the objective of a 25% reduction in dietary fat intake, and their diet was supplemented with 25 g unprocessed wheat bran/d. Control groups did not change their diets. LDL cholesterol values of both control and experimental groups within each MNS blood type were measured at baseline and 6, 12, and 18 mo. These values are shown within the evidence summary presented in Table 2. A difference in responses to the low-fat diet was shown between the MN blood group and the combined MM and NN blood groups. In the ANOVA, this difference was statistically significant at both 6 and 18 mo (P = 0.0031 and P = 0.0002, respectively) but not at 12 mo (P = 0.0936). It was concluded that an association exists between MNS blood groups and the variability of responses to a low-fat diet, with MN individuals responding the least. Of interest, LDL-cholesterol concentrations within the MN control group (ie, without adherence to a low-fat diet) had decreased by the end of the 18-mo study from 4.6217 to 4.3847 mmol/L (a change of −0.237 mmol/L) compared with LDL-cholesterol concentrations within the MN intervention group, which had slightly increased from 4.2469 to 4.2516 mmol/L (a change of 0.0047 mmol/L).

TABLE 2

Summary of findings

Outcome Comparison/risk factor Effect size No. of participants (experimental, control) Reference 
Change in mean values of LDL cholesterol (mmol/L), according to MNS status at 6 mo from baseline Low-fat diet compared with normal dietary regimen MN genotype Total: 254 (127, 127) Birley et al, 1997 (29) 
 Intervention: 4.1955 − 4.2469 = −0.0514 MN genotype: 128 (67, 61) 
 Control: 4.4949 − 4.6217 = −0.1268 MM genotype: 78 (38, 40) 
MM genotype NN genotype: 48 (22, 26) 
 Intervention: 4.1316 − 4.4026 = −0.271 
 Control: 4.6256 − 4.5923 = 0.0333 
NN genotype 
 Intervention: 3.8909 − 4.2364 = −0.3455 
 Control: 4.1261 − 4.2280 = −0.1019 
Significant difference between MN and MM/NN intervention groups in ANOVA; P = 0.0031 
Change in mean values of LDL cholesterol (mmol/L) according to MNS status at 12 mo from baseline MN genotype 
 Intervention: 4.1508 − 4.2469 = −0.0961 
 Control: 4.3915 − 4.6217 = −0.2302 
MM genotype 
 Intervention: 4.2056 − 4.4026 = −0.1970 
 Control: 4.4538 − 4.5923 = −0.1385 
NN genotype 
 Intervention: 3.8752 − 4.2364 = −0.3612 
 Control: 4.3615 − 4.2280 = −0.1335 
Difference between MN and MM/NN intervention groups in ANOVA; P = 0.0936 
Change in mean values of LDL cholesterol (mmol/L) according to MNS status at 18 mo from baseline MN genotype 
 Intervention: 4.2516 − 4.2469 = 0.0047 
 Control: 4.3847 − 4.6217 = −0.2370 
MM genotype 
 Intervention: 4.0389 − 4.4026 = −0.3637 
 Control: 4.4897 − 4.5923 = −0.1026 
NN genotype 
 Intervention: 3.8045 − 4.2364 = −0.4319 
 Control: 4.1200 − 4.2280 = −0.1280 
Significant difference between MN and MM/NN intervention groups in ANOVA, P = 0.0002 
Outcome Comparison/risk factor Effect size No. of participants (experimental, control) Reference 
Change in mean values of LDL cholesterol (mmol/L), according to MNS status at 6 mo from baseline Low-fat diet compared with normal dietary regimen MN genotype Total: 254 (127, 127) Birley et al, 1997 (29) 
 Intervention: 4.1955 − 4.2469 = −0.0514 MN genotype: 128 (67, 61) 
 Control: 4.4949 − 4.6217 = −0.1268 MM genotype: 78 (38, 40) 
MM genotype NN genotype: 48 (22, 26) 
 Intervention: 4.1316 − 4.4026 = −0.271 
 Control: 4.6256 − 4.5923 = 0.0333 
NN genotype 
 Intervention: 3.8909 − 4.2364 = −0.3455 
 Control: 4.1261 − 4.2280 = −0.1019 
Significant difference between MN and MM/NN intervention groups in ANOVA; P = 0.0031 
Change in mean values of LDL cholesterol (mmol/L) according to MNS status at 12 mo from baseline MN genotype 
 Intervention: 4.1508 − 4.2469 = −0.0961 
 Control: 4.3915 − 4.6217 = −0.2302 
MM genotype 
 Intervention: 4.2056 − 4.4026 = −0.1970 
 Control: 4.4538 − 4.5923 = −0.1385 
NN genotype 
 Intervention: 3.8752 − 4.2364 = −0.3612 
 Control: 4.3615 − 4.2280 = −0.1335 
Difference between MN and MM/NN intervention groups in ANOVA; P = 0.0936 
Change in mean values of LDL cholesterol (mmol/L) according to MNS status at 18 mo from baseline MN genotype 
 Intervention: 4.2516 − 4.2469 = 0.0047 
 Control: 4.3847 − 4.6217 = −0.2370 
MM genotype 
 Intervention: 4.0389 − 4.4026 = −0.3637 
 Control: 4.4897 − 4.5923 = −0.1026 
NN genotype 
 Intervention: 3.8045 − 4.2364 = −0.4319 
 Control: 4.1200 − 4.2280 = −0.1280 
Significant difference between MN and MM/NN intervention groups in ANOVA, P = 0.0002 

View Large

The study was a controlled interrupted time series, with participants drawn from another study (43). The process of random assignment was unclear, and there was a lack of participant blinding (which is somewhat unavoidable in dietary interventions). A total of 315 patients were randomly assigned (intervention: n = 157; control: n = 158); however, only 254 patients were analyzed, which indicated an incomplete accounting of patients and outcome events.

Because the study was a randomized experimental trial, the initial grade for the quality of evidence was high (A). However, this level was downgraded because of study limitations (a lack of adequate random assignment and allocation concealment along with an incomplete accounting of patients), imprecision (a small study size), and indirectness (the analysis provided an indirect outcome because it did not directly assess the current PICO question). Overall, the strength of the evidence, according to the GRADE approach, was determined as low (C) to very low (D).

DISCUSSION

Blood type diets have maintained significant popularity for more than a decade. They have been heavily endorsed on claims of health improvement, but to our knowledge, no studies have been performed to specifically examine resulting health effects. This situation is despite it being mentioned within D’Adamo’s books (published in 1996 and 2004) that ABO blood type diet trials were expected to be completed within 2 y and 12 wk, respectively (44, 45).

This current systematic review was performed by using sensitive search strategies within well-established life sciences and biomedical information databases. Of the 16 articles identified on the basis of titles and abstracts (Table 1), only 4 articles presented experimental or observational studies. Of these 4 articles, only one article met all of the inclusion and exclusion criteria; Birley et al (29

Much of Dr. McMahon's criticism suffers from the fallacy of argument from incredulity (sometimes called "argument from ignorance") and the fallacy of argument by appeal to fear and emotions.

Both Henry Lindlahr and Benedict Lust died in their early 60's. A  recent study of 1200 people who reached the century mark between 1932 and 1952 showed only four were vegetarians. At some point in time, naturopathic medicine, and in some peoples' minds health itself, became strongly associated with a vegetarian diet.  Some individuals within my own profession suggest that I had abandoned the core of naturopathic medicine by advocating good quality meat for individuals of certain blood types.

Getting stuck in a belief system can be a tough place to dwell. Often a good question to ask is, "what does the evidence show?"  Naturopathic medicine developed from the water cure movement of Europe. Theodor Hahn is credited as being the first of the pioneers of this water cure movement to integrate vegetarian dietetic principles. He was convinced that a meat-free diet would prolong life. In fact he was so convinced of the value of a vegetarian diet that he spent a great deal of his professional life writing books and pamphlets on the subject and was the editor of a magazine called The Vegetarian

He died of colon cancer at the age of 59. 

Perhaps his diet some how extended his life and he would have died at an even younger age had he not been a vegetarian. This at least is a quite common argument I have heard repeated by supporters of a vegetarian diet when one of their proponents dies an early death.  Ultimately there is no answer, but it is ironic that the person responsible for integrating a vegetarian diet into what would become naturopathic medicine died so young of colon cancer.

According to Drs. McMahon and Williams, Henry Lindlahr, the  founder of "Scientific Naturopathy" in the United States, "was completely committed to vegetarian diet."  Actually, the opposite is true. In fact it has been stated that Lindlahr often incurred the wrath of militant vegetarians by suggesting that "properly prepared and combined vegetables and meats could be more wholesome than certain bad vegetarian combinations." He wrote that it was not his intention "to make a fetish of vegetarianism." Clearly he was not the strict proponent of this diet that these critics would like to believe.

Dr. D'Adamo is far from the only naturopath advocating a hunter gatherer-type diet for some people. The work of Dr. Ron Schmid, ND in his books Native Nutrition and Traditional Foods Are Your Best Medicine constitute some of the best writings by an ND on the subject. 

There may be a repair mechanism that helps heal the intestinal lining from lectin damage, but that does not constitute an effective argument for their wholesale consumption.  Put another way,  our skin would also eventually heal if we sliced it open with a kitchen knife, but that is not a good reason to cut yourself.

The lectin ricin  is from the castor bean, Ricinis communis, not botulism toxin (as McMahon and Williams state above). Botulism is actually a bacterial toxin, not a lectin, from the Clostridium botuliinum. The rest of the review is a hodgepodge of assertions, with many technical errors (including an incorrect depiction of the rational behind the use of the urinary indican test to assess an individual's success with the diet) and  leaves us no purposeful way to respond. 

Is it too much to expect a critic with an agenda to be possess a rudimentary level of knowledge in the field they purport to criticize?

 


0 thoughts on “Blood Type Diet Research Paper”

    -->

Leave a Comment

Your email address will not be published. Required fields are marked *