The study team noted about 25 percent less risk for Type 2 diabetes among men and women who reported eating one or more, as opposed to fewer, servings of white or oily fish each week.

Unexpectedly, however, they found that men and women who ate similar amounts of shellfish -- primarily prawns, crab, and mussels -- had about 36 percent increased risk of developing Type 2 diabetes.

But "it may not be the 'shellfish' per se which increased the risk for diabetes," Dr. Nita Forouhi, of Addenbrooke's Hospital, University of Cambridge, noted in an email to Reuters Health.

Rather, the cooking and preparation methods used in the UK, for example, oils used when frying or butter- and mayonnaise-based sauces served with shellfish, may increase cholesterol intake which, in turn, may raise diabetes risk.

Forouhi and colleagues assessed the weekly intake of shellfish plus white fish such as cod, haddock, sole, and halibut, or oily fish such as mackerel, kippers, tuna, and salmon, reported by 9,801 men and 12,183 women. The study participants were 40 to 79 years old at the time and had no history of diabetes.

Over an average of 10 years, 725 of these men and women developed Type 2 diabetes.

Both the lower risk linked with white and oily fish and the greater risk tied to shellfish intake remained when the investigators allowed for a range of diabetes risk factors including physical activity, obesity, alcohol use, and fruit and vegetable intake.

The investigators emphasize that the link between shellfish intake and diabetes risk requires further investigations in other populations. This observed link, Forouhi commented, "does not imply that one is the cause of the other."

The findings on white and oily fish "reinforce the public health message to consume fish regularly," the investigators conclude, while the shellfish findings should be studied further.