Long-term Follow-up

How often are the participants re-assessed?

Periodic resurveys are conducted every 4-5 years, involving about 5% randomly selected surviving participants. Following completion of the baseline survey in July 2008, a resurvey was undertaken immediately in 10 study regions during August and October 2008. It used a cluster random sampling approach, with the administrative unit (i.e. rural village or urban residential committee) as the basic sampling unit. Apart from a few additional questions (e.g. about recent hospitalizations), the data collection and survey procedures were much the same as in the baseline survey. The second resurvey was conducted during 2013-14. The data from the first and subsequent periodic resurveys will allow, after controlling for “regression dilution” bias, unbiased estimation of associations between long-term “usual” levels of particular risk exposures and disease in the whole population. These resurveys also provide an opportunity to add new and more detailed data collection to the study (e.g. better measures of physical activity or dietary patterns), so that information collected at the baseline survey can be calibrated.

What long-term follow-up data are collected?

A range of health related outcome data are collected, including cause-specific mortality, morbidity for a few major diseases and any episode of hospitalization. The vital status of study participants is being monitored regularly through official residential records and death certificates reported to the Regional CDC, where each study RCC office is based. Any deaths occurring among participants are coded (using the 10th International Classification of Diseases, ICD-10) by trained staff "blinded" to baseline information. Causes of death from official death certificates are being supplemented, if necessary, by review of medical records (which are usually available). Information on disease incidence for stroke, IHD, cancers, and diabetes is also being collected through linkage with established disease registries that are currently available in 8 out of the 10 study areas. Electronic linkage with the newly established national health insurance system has now been acheived for ~98% of participants, which records details of all hospitalised events including disease description, diagnositc procedures and ICD-10 code.