Dear Hongyun,
Well, shrimp is one of the typical 'not very recommended' items to eat in cases of 'elevated' cholesterol though I am definitely not competent to discuss the significance of the numbers.
Regarding the mercury issue, there is a detailed look in this link -
http://www.fao.org/d...2e09.htm#bm09.2( this link also illustrates the substantial effort required to carry out MRAs)
I think your general question about the value of eating more or less seafood is interesting and deserved a bit more response. It is rather difficult to give answers to such a broad query, ie specific objectives could be to avoid being ill (or worse) from food poisoning in general, or to avoid salmonellosis, or to not suffer specific health problems as in yr link or … It also obviously depends on the consumer vulnerability as usual (eg pregnant ladies).
Either way, clearly understandable (to me anyway) quantitative data seems not so easy to find.
A rather negative overall view is given here -
http://www.fao.org/d...3e06.htm#bm06.1however a more positive interpretation (though earlier) is here -
http://www.seafoodla...m... to eat.pdfsome of the discrepancy between the two links is possibly commented here -
http://www.findartic...v25/ai_10942749A more recent (positive but without many explicit numbers) link is -
http://www.seafoodpl...tation_plan.pdf.
Maybe somebody else can find a better or clearer answer.
Rgds / Charles.C