In the report, the second sentence of the Abstract should read as follows: “A population of 1080 healthy Japanese male local government employees without lifestyle-related diseases at baseline were studied.”
In addition, the third sentence of the Methods section should read as follows: “Among 2144 men with no missing data, 1045 men who had a history of heart disease, stroke, diabetes, hypertension, hyperlipidemia, or hyperuricemia in 1997 and 19 men who had a cancer history in 1997 or 2002 were excluded….”
The authors apologize for these errors.