Increase Font Size Decrease Font Size View as PDF Print

What is the relationship between adherence to dietary guidelines/recommendations or specific dietary patterns (assessed using methods other than index/score, cluster or factor, or reduced rank regression analyses) and risk of type 2 diabetes?

Conclusion

There is insufficient evidence on a relationship between adherence to a Mediterranean-style or vegetarian diet pattern and incidence of type 2 diabetes. There is limited, inconsistent evidence that adherence to a Mediterranean-style, DASH or modified DASH, or Nordic dietary pattern results in improved glucose tolerance and insulin resistance.
 

Grade

IV-Not Assignable – Incidence of type 2 diabetes; III-Limited-Glucose tolerance and insulin resistance

 

Key Findings

  • Four types of dietary patterns were identified using other methods of assessing dietary exposure related to type 2 diabetes risk: (1) a Mediterranean-style pattern, (2) a DASH or modified DASH pattern, (3) a vegetarian pattern, and (4) a Nordic pattern.
  • Overall, there were too few articles and the dietary patterns and study characteristics were too varied to compare across studies.
  • A favorable association was found in a Mediterranean-style diet combined with olive oil and/or nuts, and in a vegetarian diet compared to a non-vegetarian diet with incidence of type 2 diabetes higher in Black versus non-Blacks.
  • Five out of eight studies were conducted outside of the United States with only three out of eight articles reported race/ethnicity and, of those, only one study reported results based on race/ethnicity.
  • Limitations of the studies include:
  • All of the randomized controlled trials (RCTs) included different at-risk populations.
  • Too few articles examined a relationship between dietary patterns and the endpoint outcome of incident type 2 diabetes to draw a conclusion, although the two patterns studied (one Mediterranean-style and one vegetarian) showed a favorable effect.
  • Too few articles assessed the intermediate outcomes of impaired glucose tolerance and/or insulin resistance. The results related to impaired glucose tolerance and/or insulin resistance were too mixed to identify a consistent pattern.
  • It is difficult to assess food components, as there were too few studies across several different patterns that were operationalized differently.
     

Evidence Summary Overview

Description of the Evidence
A total of eight articles met the inclusion criteria for this systematic review on dietary patterns and incident type 2 diabetes outcomes assessed using methods other than index/score, cluster or factor, or reduced rank regression analyses. The body of evidence examined seven studies (two articles on PREDIMED) consisting of six RCTs (RCTs) (Adamsson, 2011; Blumenthal, 2010; Esposito, 2004; Gadgil, 2013; Rallidis, 2009; Salas-Salvado, 2008 and 2001) and one PCS (PCS) (Tonstad, 2013). In terms of study quality, seven of the eight articles received a positive quality rating and one was rated neutral (Tonstad, 2013). The studies were carried out between 2004 and 2013. Two studies were conducted in the United States; one in the United States and Canada; one in Spain (2 PREDIMED articles); and one each in Greece, Italy, and Sweden. The sample sizes of the RCTs ranged from 82 to 1,224 participants and the PCS had a sample size of 41,387 participants (in total, 2 studies <100; 3 studies >100; 1 study >400; 1 study >1,000; and 1 study >40,000). All eight studies were conducted in adults. RCT duration ranged from 6 weeks to a median of 4 years and the PCS duration was 2 years (in total, 5 studies <1 year, 2 studies <2 years, and 1 study >3 years).
 
Population:
The RCTs were primary prevention studies of at-risk subjects. Baseline health status in these subjects included subjects with mild hypercholesterolemia (Adamsson, 2011); overweight or obese subjects (Blumenthal, 2010; Gadgil, 2013); subjects with metabolic syndrome (Esposito, 2004); subjects with abdominal obesity (Rallidis, 2009); and subjects with three or more CVD risk factors, including metabolic syndrome (Salas-Salvado, 2008 and 2011). The PCS subjects were non-diabetic individuals in the Adventist Health Study (Tonstad, 2013). The mean age in six of the studies was between 44 and 67 years and the age of participants in one study ranged from 25 to 65 years (Adamsson, 2011). Female participation in the seven studies was between 45 and 67 percent. Race or ethnicity was not reported in the three studies that looked at a Mediterranean-style diet (Esposito, 2004; Rallidis, 2009, Salas-Salvado, 2008 and 2011) and one study examining the Nordic diet (Adamsson, 2011). Three studies that looked at either at the DASH diet or variation of the DASH diet (Blumenthal, 2010; Gadgil, 2013) or a vegetarian diet (Tonstad, 2013) had a representative Black subgroup which accounted for 17 to 55 percent of the study population.
 
Dietary Exposure:
Four of the seven studies (all RCTs) examined a Mediterranean-style dietary pattern or variation of one. One study looked at the Mediterranean-style diet with addition of either olive oil or nuts compared to a control, low-fat diet in the PREDIMED trial of subjects at risk for CVD (Salas-Salvado, 2008 and 2011). One study looked at the Mediterranean-style diet versus a “Prudent” diet as control in subjects with metabolic syndrome (Esposito, 2004) and one study looked at a Greek Mediterranean diet in subjects with abdominal obesity (Rallidis, 2009). The other three RCTs looked at either a variation of the DASH dietary pattern (Blumenthal, 2010 [ENCORE]; Gadgil, 2013 [OmniHeart]); and the Nordic diet (Adamsson, 2011). The PCS looked at vegetarian patterns, including vegan, lacto-ovo vegetarian, pesco vegetarian, and semi-vegetarian (Tonstad, 2013). Table 1 provides an overview of the study characteristics, a description of the dietary patterns examined, and the results as intermediate and endpoint outcomes.
 
Dietary Assessment:
Dietary intake in this review was assessed using a variety of methods, including food frequency questionnaires (FFQs), diet adherence questionnaires, dietary history interviews, 24-hour recalls, and food records, diaries, and checklists.
 
Four of the seven studies used a FFQ to assess dietary intake in addition to other assessment methods. In the PREDIMED trial, a 137-item validated FFQ and a 14-item validated questionnaire on adherence to the traditional Mediterranean diet were used (Salas-Salvado, 2008 and 2011). In the ENCORE trial, a self-reported FFQ (4-week recall) and 4-day food diary were used (Blumenthal, 2010). The only PCS in this review, Tonstad (2013), used a validated FFQ of 130 commonly consumed food/food groups and included a write-in option. The FFQ was developed specifically for the study population.
 
Other methods of assessing dietary intake included weekly diet diaries (Esposito, 2004); 3-day food diaries, 24-hour recalls, check list of foods consumed daily, and the return of empty food packages (Rallidis, 2009); and diet history interviews by trained dieticians before baseline and after 6 weeks with self-reported daily checklists to record deviations from the menu (Adamsson, 2011). Gadgil (2013) provided all meals and snacks to subjects who were also asked to limit alcohol consumption to usual patterns.
 

Qualitative Synthesis of the Collected Evidence

Themes and Key Findings
 
Health Outcomes:
The eight articles in this review examined (1) type 2 diabetes incidence or (2) impaired glucose tolerance and/or insulin resistance.
 
Incidence of Type 2 Diabetes
Studies that examined incidence of type 2 diabetes as the primary outcome were included in this category (table 1). Subjects who met the American Diabetes Association or World Health Organization criteria for fasting blood glucose or oral glucose tolerance, or were taking hypoglycemic medication, were considered having incident type 2 diabetes. Two studies examined the association between adherence to a dietary pattern and incidence of type 2 diabetes (Salas-Salvados, 2011; Tonstad, 2013).
 
Although the results of both studies showed a favorable association between either a Mediterranean-style or a vegetarian dietary pattern and incidence of type 2 diabetes, the studies differed in design and dietary pattern used to assess diet exposure. In the PREDIMED trial, Salas-Salvado (2011) compared adherence to a Mediterranean-style diet with a low-fat diet (<35 percent fat) in subjects with three or more CVD risk factors in Spain. The Mediterranean-style diet included the provision of olive oil or nuts to the two treatment groups, and both groups showed decreased incidence of type 2 diabetes compared to the control low-fat group. In a PCS, Tonstad (2013) compared a vegetarian diet with a non-vegetarian diet in non-diabetic subjects in the United States and Canada (Adventist Health Study-2). In addition to a favorable association among vegetarians compared to non-vegetarians, Tonstad also stratified results by race/ethnicity (Blacks and non-Blacks) and vegetarian category (vegan, lacto-ovo, pesco, and semi-vegetarians). For incident type 2 diabetes, Blacks had an increased risk compared to non-Blacks. In Whites, vegan, lacto-ovo vegetarian, and semi-vegetarian diets were protective against type 2 diabetes; whereas, in Blacks, only the vegan and lacto-ovo vegetarian diets showed a decreased risk of type 2 diabetes, compared to a non-vegetarian diet.
 
Impaired Glucose Tolerance and/or Insulin Resistance
This category included studies that assessed fasting blood glucose, oral glucose tolerance, fasting blood insulin, insulin resistance using the Homeostasis Model Assessment-Insulin Resistance (HOMA-IR) equation, or the quantitative insulin sensitivity check index (QUICKI) (table 4-C-IV-1). These outcomes were measured by standard clinical and laboratory methods. Six articles (all RCTs) looked at the effect of adherence to a dietary pattern and intermediate outcomes related to glucose tolerance and/or insulin resistance (Adamsson, 2011; Blumenthal, 2010; Esposito, 2004; Gadgil, 2013; Rallidis, 2009; Salas-Salvado, 2008). Salas-Salvado (2008) did not examine insulin resistance. The results from these studies on glucose tolerance and/or insulin resistance are mixed.
 
The three RCTs examining a Mediterranean-style diet pattern reported inconsistent results on glucose tolerance and/or insulin resistance. The PREDIMED trial in Spain found no effect of a Mediterranean-style diet on fasting glucose in subjects with three or more CVD risk factors (Salas-Salvado 2008). Instruction to adhere to a Mediterranean-style diet resulted in improved glucose tolerance and insulin resistance, compared to a Prudent diet, in subjects with metabolic syndrome in Italy (Esposite 2004); however, this had no effect on insulin resistance among subjects with abdominal obesity in Greece (Rallidis, 2009). 
 
The DASH diet and modified DASH diet were examined in two RCTs (Blumenthal, 2010; Gadgil, 2013). In the ENCORE trial, results showed that the DASH diet alone had no effect on fasting glucose or fasting insulin and HOMA-IR among overweight or obese subjects with high blood pressure (Blumenthal, 2010). In the OmniHeart trial, results showed that a DASH diet modified to be high in unsaturated fat (primarily monounsaturated fat), when compared to a carbohydrate-rich diet (similar to DASH), had a favorable effect on HOMA-IR, but no effect on fasting blood glucose or insulin levels in overweight or obese subjects with high blood pressure (Gadgil, 2013). The high carbohydrate/DASH-style diet and the high protein diet had no effect on HOMA-IR, fasting glucose or fasting insulin.
 
Adamsson (2011) (RCT) found that a Nordic diet in Sweden among subjects with mild hypercholesterolemia resulted in a decrease in fasting insulin resistance and HOMA-IR, but no effect on fasting glucose.
 

Qualitative Assessment of the Collected Evidence

Quality and Quantity
Quality assessment for the studies included in this systematic review involved determining the validity of each study. Validity was assessed by examining the scientific soundness of study design and execution to avoid potential bias in the findings related to outcomes. The majority of evidence for this body of evidence consisted of positive quality studies (7 out of 8 articles). This was a limited number of studies with variation in the outcomes measured.
 
Consistency
There were too few articles related to the clinical endpoint outcome, incident type 2 diabetes, with different study designs, countries, and dietary patterns to make them comparable even though the studies found a favorable outcome associated with either a Mediterranean-style diet (PREDIMED) or a vegetarian diet. Glucose tolerance and/or insulin resistance were assessed in six large RCTs that examined three different dietary patterns (Mediterranean-style, DASH-style, and Nordic) with mixed findings. There were differences within the Mediterranean-style and DASH-style patterns that made comparison difficult.
 
Impact
The body of evidence directly addressed the exposures and health outcomes of interest for this systematic review, but only two studies measured the endpoint outcome, incident type 2 diabetes. When there were associations between a dietary pattern and incidence of type 2 diabetes, these were clinically meaningful. However the two studies cannot be compared since they looked at different dietary patterns.
 
Generalizability/External Validity
Five out of the eight studies were conducted outside of the United States/Canada so those dietary patterns examined may not be representative of a U.S. diet. Also, five of the studies did not report race/ethnicity and only one study stratified their results based on race/ethnicity (Blacks versus non-Blacks). However, the European studies were likely conducted with a predominantly Caucasian population; therefore, it may be representative of a U.S. Caucasian population. Overall, there were too few studies on either incident type 2 diabetes or glucose tolerance and/or insulin resistance to draw a conclusion based on the findings.
 

Limitations of the Evidence

It is difficult to synthesize the results from the studies in this review because there were too few studies and they examined different dietary patterns or patterns that were operationalized differently. The studies included a predominantly Caucasian population with varied baseline health status.
 

Research Recommendations

Overall, there is a need for additional research RCTs and observational studies conducted in the United States on risk of type 2 diabetes that address the key dietary patterns in a standardized way. In addition, more analysis of key subpopulation groups would further inform policy in this area.
 

Abbreviations

Dietary Approaches to Stop Hypertension (DASH), Exercise & Nutrition interventions for Cardiovascular Health (ENCORE), Nordic Diet (NORDIET), Optimal Macronutrient Intake trial to Prevent Heart Disease (OmniHeart), Prevention with Mediterranean Diet (PREDIMED).

Table 4-C-IV-1Summary of Findings Impaired glucose tolerance, insulin resistance, and incident type 2 diabetes
Table 4-C-IV-2 Overview Table: Type 2 Diabetes
 

REFERENCES

  1. Adamsson V, Reumark A, Fredriksson IB, Hammarström E, Vessby B, Johansson G, Risérus U. Effects of a healthy Nordic diet on cardiovascular risk factors in hypercholesterolaemic subjects: a randomized controlled trial (NORDIET). J Intern Med. 2011 Feb;269(2):150-9. doi: 10.1111/j.1365-2796.2010.02290.x. Epub 2010 Oct 22. PubMed PMID: 20964740. (Also in DP-CVD Other Methods)
  2. Blumenthal JA, Babyak MA, Sherwood A, Craighead L, Lin PH, Johnson J, Watkins LL, Wang JT, Kuhn C, Feinglos M, Hinderliter A. Effects of the dietary approaches to stop hypertension diet alone and in combination with exercise and caloric restriction on insulin sensitivity and lipids. Hypertension. 2010 May;55(5):1199-205. Epub 2010 Mar 8. PMID: 20212264. (Also in DP-BW+CVD- Other Methods)
  3. Esposito K, Marfella R, Ciotola M, Di Palo C, Giugliano F, Giugliano G, D'Armiento M, D'Andrea F, Giugliano D. Effect of a mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA. 2004 Sep 22;292(12):1440-6. PubMed PMID: 15383514.
  4. Gadgil MD, Appel LJ, Yeung E, Anderson CA, Sacks FM, Miller ER 3rd. The effects of carbohydrate, unsaturated fat, and protein intake on measures of insulin sensitivity: results from the OmniHeart trial. Diabetes Care. 2013 May;36(5):1132-7. doi: 10.2337/dc12-0869. Epub 2012 Dec 5. PMID: 23223345.
  5. Rallidis LS, Lekakis J, Kolomvotsou A, Zampelas A, Vamvakou G, Efstathiou S, Dimitriadis G, Raptis SA, Kremastinos DT. Close adherence to a Mediterranean diet improves endothelial function in subjects with abdominal obesity. Am J Clin Nutr. 2009 Aug;90(2):263-8. Epub 2009 Jun 10. PubMed PMID: 19515732.
  6. Salas-Salvadó J, Bulló M, Babio N, Martínez-González MÁ, Ibarrola-Jurado N, Basora J, Estruch R, Covas MI, Corella D, Arós F, Ruiz-Gutiérrez V, Ros E; PREDIMED Study Investigators. Reduction in the incidence of type 2 diabetes with the Mediterranean diet: results of the PREDIMED-Reus nutrition intervention randomized trial. Diabetes Care. 2011 Jan;34(1):14-9. Epub 2010 Oct 7. PubMed PMID: 20929998; PubMed Central PMCID: PMC3005482.
  7. Salas-Salvadó J, Fernández-Ballart J, Ros E, Martínez-González MA, Fitó M, Estruch R, Corella D, Fiol M, Gómez-Gracia E, Arós F, Flores G, Lapetra J, Lamuela-Raventós R, Ruiz-Gutiérrez V, Bulló M, Basora J, Covas MI; PREDIMED Study Investigators. Effect of a Mediterranean diet supplemented with nuts on metabolic syndrome status: one-year results of the PREDIMED randomized trial. Arch Intern Med. 2008 Dec 8;168(22):2449-58. PubMed PMID: 19064829.
  8. Tonstad S, Stewart K, Oda K, Batech M, Herring RP, Fraser GE. Vegetarian diets and incidence of diabetes in the Adventist Health Study-2. Nutr Metab Cardiovasc Dis. 2013 Oct 7. [Epub ahead of print] PubMed PMID: 21983060.



Research Design and Implementation
For a summary of the Research Design and Implementation results, click here.
Worksheets
Adamsson V, Reumark A, Fredriksson IB, Hammarström E, Vessby B, Johansson G, Risérus U. Effects of a healthy Nordic diet on cardiovascular risk factors in hypercholesterolaemic subjects: A randomized controlled trial (NORDIET). J Intern Med. 2011 Feb; 269(2): 150-159.

Blumenthal JA, Babyak JA, Sherwood A, Craighead L, Pao-HWa L, Johnson J, Watkins LL, Wang JT, Kuhn C, Feinglos M, Hinderliter A. The effects of the dash diet alone and in combination with exercise and caloric restriction on insulin sensitivity and lipids. Hypertension. 2010; 55(5): 1,199-1,205.  

Esposito K, Marfella R, Ciotola M, Di Palo C, Giugliano F, Giugliano G, D'Armiento M, D'Andrea F, Giugliano D. Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome. JAMA. 2004; 292 (12): 1,440-1,446.

Gadgil MD, Appel LJ, Yeung E, Anderson CA, Sacks FM, Miller ER 3rd. The effects of carbohydrate, unsaturated fat, and protein intake on measures of insulin sensitivity: Results from the OmniHeart trial. Diabetes Care. 2013 May; 36(5): 1,132-1,127.

Rallidis LS, Lekakis J, Kolomvotsou A, Zampelas A, Vamvakou G, Efstathiou S, Dimitriadis G, Raptis SA, Kremastinos DT. Close adherence to a Mediterranean diet improves endothelial function in subjects with abdominal obesity. Am J Clin Nutr. 2009 Aug; 90(2): 263-268.

Salas-Salvado J, Bullo M, Babio N, Martinez-Gonzalez M, Ibarrola-Jurado N, et al. Reduction in the incidence of type 2 diabetes with the Mediterranean diet: results of the PREDIMED-Reus nutrition intervention randomized trial. Diabetes Care. 2011 Jan; 34(1): 14-19.

Salas-Salvado J , Fernandez-Ballart J, Ros E, Martinez-Gonzalez MA, Fito M, Estruch R ,Corella D, Fiol D, Gomez-Gracia E, Aros F, Flores G, Lapetra J, Lamuela-Raventos R, Ruiz-Gutierrez V, Bullo M, Basora J, Covas M. Effect of a Mediterranean diet supplemented with nuts on metabolic syndrome status; one-year results of the PREDIMED randomized trial. Arch Intern Med. 2008; 168 (22): 2,449-2,458. 

Tonstad S, Stewart K, Oda K, Batech M, Herring RP, Fraser GE. Vegetarian diets and incidence of diabetes in the Adventist Health Study-2. Nutr Met & Cardiovasc Dis.  2013; 23: 292-299.