A systematic review of observational studies has been published in Endocrine Practice that concludes diabetes is associated with a small but signficant increased risk of cancer.
In people with diabetes, what is the incidence of cancer compared to people without diabetes?
The reviewers searched Medline and the Cochrane Library for relevant studies up to June 2010. They found 12 studies, comprising 257,222 patients in total.
When they combined all of the data from all of the studies, they found that people with diabetes had a Relative Risk of 1.10 (1.04 to 1.17) of getting any type of cancer, compared to people without diabetes.
They cite an overall incidence of 7% among these studies. Assuming this figure as a baseline risk, diabetes would raise the risk to 7.7%. Translating this to absolute risk, diabetes could be associated with one extra case of cancer for every 142 diabetics.
- As they only searched Medline and the Cochrane Library, it seems likely that the reviewers will have missed relevant studies published elsewhere.
- Did the reviewers do enough to assess the quality of the studies they included in their review?
- Does it make sense to combine all these different studies – carried out in different places and at different times – into one overall assessment?
- It seems likely that any impact of diabetes would be different among different groups of patients. Did the reviewers carry out appropriate sub-group analyses that might be more directly relevant to you?
- Analysis by type, duration, and treatment of diabetes would be of relevance.
- This observational research does not establish causation. There may be a third unknown factor (e.g. diet, lifestyle) that causes both diabetes and cancer.
Noto H, Tsujimoto T, Sasazuki T, Noda M. Significantly increased risk of cancer in patients with diabetes mellitus: a systematic review and meta-analysis. Endocr Pract. 2011 Jul-Aug;17(4):616-28.