It has been suggested that dairy products might have a preventive effect against type 2 diabetes. However, the results of observational studies looking at the correlation between the consumption of dairy and incidence of type 2 have been inconclusive. A new systematic review set out to combine the evidence from these different studies into an overall estimate, and explore the potential impact of different types and levels of consumption.
The reviewers searched MEDLINE, EMBASE and Scopus using the terms “diabetes mellitus”, “dairy” and “milk”. Two independent reviewers examined the studies they found and evaluated their quality.
From these studies they extracted data about populations, settings, type of dairy food consumption and diabetes status. Dairy consumption was standardised to grams per day. The risk of diabetes was expressed as a Relative Risk (RR). Where studies didn’t re, or a transformed odds ratio where RRs were not directly reported.
The reviewers carried out meta-analysis based on an assumption of a non-linear relationship between diary consumption and risk of type 2 diabetes.
Data from 13 studies were analysed, comprising 457,893 subjects and 27,095 cases of type 2 diabetes. Overall, the relative risk for having a high intake of dairy foods compared with low intake was 0.89 (95% CI 0.81–0.98). That is, more intake of dairy products was associated with a lower incidence of type 2 diabetes.
Using low-fat dairy products was even better. Eight studies compared low-fat versus full-fat dairy diets. Here, the RR was 0.81 (0.74–0.89).
Although this observational evidence cannot establish causation, the reviewers also looked for evidence of a dose-response relationship. Although these results were less clear, they concluded:
Our analysis of high- and low-fat dairy products revealed an inverse association of only low-fat dairy food intake and T2DM risk.
We should remember that this is observational evidence and cannot establish causation. Consumption of low fat dairy products may be acting as a marker of other healthy activities, for example. We’d really want to see prospective evidence from an RCT to be sure.
- This may be tricky, since we already know that there are other benefits of low-fat diets and it wouldn’t exactly be ethical to deny them to a study participant. So this might be the best evidence we are likely to get.
- There are some other caveats to this study. The evaluation criteria for study quality were not previously validated. Two of the studies included self-reports of diabetes status. Overall, methodological quality was “moderate”.
- There was substantial heterogeneity in the studies, including the population taking part, the types of diet being tested and compared and the baseline incidence of type 2 diabetes. However, the consistency of effect was encouraging.
- The review text mentions 15 studies but data are only presented for 13.
- Finally, the reviewers themselves highlight a potentially substantial risk of publication bias. Negative studies are less likely to be published, or indeed submitted for publication. They present an interesting and detailed exploration of the possible impact.
Dengfeng Gao, Ning Ning, Congxia Wang, Yuhuan Wang, Qing Li, Zhe Meng, Yang Liu, Qiang Li. Dairy Products Consumption and Risk of Type 2 Diabetes: Systematic Review and Dose-Response Meta-Analysis. PLoS ONE, 27 September 2013; 10.1371/journal.pone.0073965.