PMC:5664696 / 20503-21359 JSONTXT

{"target":"https://pubannotation.org/docs/sourcedb/PMC/sourceid/5664696","sourcedb":"PMC","sourceid":"5664696","source_url":"https://www.ncbi.nlm.nih.gov/pmc/5664696","text":"There are also some limitations to this study: The reliability of census data in 2000 has been questioned, especially in Xinjiang [34]. Therefore, we deleted the data in south Xinjiang in 2000. There is no accurate per capita sea fish consumption index for each city in China. Consequently, we evaluated sea fish production based on survey results from only 139 cities. Meat and freshwater fish production were used as proxies for meat and freshwater fish consumption because there are no official per capita meat and freshwater fish consumption indexes. As previously discussed, assumptions about migration and pollution were also made which might mean key factors explaining the geographical distribution of longevity might be missing. What these limitations also point to is the overall limitation of carrying out retrospective cross-sectional analyses.","tracks":[]}