This article was published in Medical Observer.
The article is also available on the Medical Observer website (may need registration).
I recently diagnosed Ji-hoon, a 40-year old Korean man, with type 2 diabetes. As we were discussing lifestyle changes that would help him manage the diabetes, Ji-hoon mentioned he had come across the “Mediterranean diet” in his reading.  I realised that although I was aware of some of the cardiovascular data for this diet, I wasn’t sure about its specific utility in diabetes management.
What is the effect of the Mediterranean diet on glycaemic control in people with type 2 diabetes?
What does the research evidence say?
Step 1: The Cochrane Library
The Cochrane Library contains a review on the Mediterranean diet for the primary prevention of cardiovascular disease, which is important but not the question we are asking. 
Step 2: Trip Database
I used the PICO search tool – “P: type 2 diabetes, I: Mediterranean diet, C: [blank], O: glycemic control”.
A number of very promising references come up! First on the list is a recent systematic review and meta-analysis by Ajala et al. (2013) published in the American Journal of Clinical Nutrition.  This paper examines many different diets (including low carb, high protein, low GI, and others), but I’ll focus exclusively on the parts of it which address the Mediterranean diet.
I will use the appropriate appraisal sheet from the Centre for Evidence Based Medicine. 
What PICO question does the systematic review ask?
In adult patients with type 2 diabetes (Participants); what is the effect of the Mediterranean diet (and a number of other diets) (Intervention); compared with a control diet (Comparator); on glycaemic control as measured by HbA1c (as well as changes in weight, and serum lipids) (Outcome).
Is it clearly stated?
Is it unlikely that important studies were missed?
Unclear. The authors described their search strategy briefly. They searched three major bibliographic databases, and manually searched reference lists of relevant reviews. It is not clear whether unpublished reports were sought.
Were the criteria used to select articles for inclusion appropriate?
Yes. The authors included all randomised trials of adult patients with type 2 diabetes that compared the diets of interest with a control diet, where the intervention lasted at least 6 months.
Were the included studies sufficiently valid for the question asked?
Unclear/Unknown. The authors formally assessed the risk of bias of the included studies, and the risk was “unclear” for most (Figure 1, p. 507).  The risk of bias of the three specific studies in the Mediterranean diet meta-analyses are unknown as the authors did not report at this level of detail.
Were the results similar between studies?
No. There was substantial heterogeneity (I2 = 82%) between the studies.
What were the results?
The authors identified 20 eligible trials – 3 provided data for the meta-analyses on the Mediterranean diet. Participants on this diet, as compared to the control diets, had:
- a mean lower HbA1c: -0.47% (95% CI -0.64 to -0.30), P < 0.00001
At best, this systematic review and meta-analysis provides moderate quality evidence on the effect that the Mediterranean diet has on glycaemic control. The substantial heterogeneity is problematic and possibly represents the large differences in participant characteristics and trial design between the studies. There is some question as to whether it was reasonable to statistically pool the results together given these differences. 
That said, we can still make some cautious conclusions. The comparator that was used in two of the included studies was a “control diet” – the American Diabetic Association diet in one,  and a low-fat diet in the other.  Although we shouldn’t definitively conclude that the Mediterranean diet is superior, it is nonetheless very likely at least as effective as these other diets in the management of diabetes. That is, it results in clinically meaningful reductions in HbA1c,  and may reduce or delay the need for pharmacotherapy. 
I advised Ji-hoon that dietary changes are an important part of the management of diabetes, and that the Mediterranean diet is as good, and possibly better, than the other recommended diets.
Study heterogeneity is expected in a meta-analysis, but too much inconsistency may render the meta-analysis invalid. The I2 statistic quantifies the heterogeneity – it is the percentage of the variability between studies that is due to heterogeneity rather than chance. 
- Johnson C. 10 commandments of the real Mediterranean diet [website]. ABC Health & Wellbeing. Last updated: 2013 Apr 15 (Retrieved: 2014 Jan 28). http://www.abc.net.au/health/thepulse/stories/2013/04/15/3737114.htm
- Rees K, Hartley L, Flowers N, Clarke A, Hooper L, Thorogood M, Stranges S. ‘Mediterranean’ dietary pattern for the primary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews 2013, Issue 8. Art. No.: CD009825. DOI: 10.1002/14651858.CD009825.pub2
- Ajala O, English P, Pinkney J. Systematic review and meta-analysis of different dietary approaches to the management of type 2 diabetes. Am J Clin Nutr 2013; 97(3): 505-516
- Systematic Review: Are the results of the review valid [MS Word document]? Centre for Evidence Based Medicine, University of Oxford. Last updated: 2008 Oct 1 (Retrieved: 2014 Jan 28) http://www.cebm.net/index.aspx?o=1157
- Mann JI, Morenga LT. Diet and diabetes revisited, yet again. Am J Clin Nutr 2013; 97(3): 453-454
- Elhayany A, Lustman A, Abel R, Attal-Singer J, Vinker S. A low carbohydrate Mediterranean diet improves cardiovascular risk factors and diabetes control among overweight patients with type 2 diabetes mellitus: a 1-year prospective randomized intervention study. Diabetes Obes Metab 2010; 12(3): 204-209. DOI: 10.1111/j.1463-1326.2009.01151.x
- Esposito K, Maiorino MI, Ciotola M, Di Palo C, Scognamiglio P, Gicchino M, et al. Effects of a Mediterranean-style diet on the need for antihyperglycemic drug therapy in patients with newly diagnosis type 2 diabetes: a randomized trial. Ann Intern Med 2009; 151(5): 306-314
- Higgins JPT, Green S [editors]. “Identifying and measuring heterogeneity” (Section 9.5.2) in Cochrane handbook for systematic reviews of interventions (version 5.1.0) [website]. The Cochrane Collaboration. Last updated: 2011 Mar (Retrieved: 2014 Jan 28). http://handbook.cochrane.org/