Survey Design

The baseline survey was conducted in 1998, with follow-up surveys with replacements for deceased elders were conducted in 2000, 2002, 2005, 2008/09, 2011/12, and 2014 in a randomly selected half of the total number of counties and cities in the 22 provinces. The survey areas covered 1.1 billion people, 85 percent of the total population in China. An enumerator and a nurse or a medical school student conducted the interview and performed a basic health examination at each interviewee’s home.

The CLHLS attempted to interview all centenarians who voluntarily agreed to participate in the study in the randomly selected half of the counties and cities of the 22 provinces. For each centenarian, one near-by octogenarian (aged 80-89) and one near-by nonagenarian (aged 90-99) of pre-designated age and sex was interviewed. “Near-by” is loosely defined – it could be in the same village or street if available, or in the same town or in the same sampled county or city. The predefined age and sex, used to identify the approximately equal numbers of male and female nonagenarians and octogenarians, are randomly determined, based on the code numbers of the centenarians. The goal was to have comparable numbers of male and female octogenarians and nonagenarians at each age from 80 to 99.

We interviewed 9,093 oldest-old persons aged 80+ (including 2,418 centenarians) in our 1998 base line survey, and 11,098 oldest old aged 80+ (including 2,431 centenarians) in our 2000 follow-up survey. Among 11,098, 4,761 elders were survivors who were interviewed in 1998 and re-interviewed in 2000; and 6,337 elders were newly recruited in 2000. In addition, 3,338 valid questionnaire interviews of those who died after the 1998 survey were conducted from a close family member of those deceased elders.

With additional support from UNFPA, the China National Social Science Foundation, and other Chinese resources, the 2002 wave of our survey expanded from covering ages 80+ to ages 65+, with a total sample size of 16,057 elders (including 3,321 centenarians, 3,632 nonagenarians, 4,210 octogenarians, and 4,894 younger elders aged 65-79) plus interviews from the next-of kin interviews for 3,348 respondents who were interviewed in the 2000 wave but died before the 2002 wave.

With supports from the Taiwan Academia Sinica and the China National Foundation for Social Sciences, we have added a sub-sample of 4,478 elderly interviewees’ adult children aged 35-65 in eight provinces in eastern coast areas among the 22 provinces of our healthy longevity survey. The main goal is to comparatively study intergenerational relationships in the context of rapid aging and healthy longevity between China Mainland and Taiwan. The total sample size of our 2002 survey is therefore nearly 21,000 persons aged 35-110.

The 2005 wave interviewed 4,955 young elderly aged 65-79 and 10,658 oldest-old aged 80+ (including 2,797 centenarians, 3,952 nonagenarians and 3,909 octogenarians), with a total sample size of 15,613 (another 25 elders are younger than age 65). Information on before dying was collected from the next-of-kin for nearly 5,900 deceased respondents who were interviewed in 2002 but died before the 2005 wave.

The 2008-2009 wave interviewed 4,286 young elderly aged 65-79 and 12,277 oldest-old aged 80+ (including 3,378 centenarians, 4,621 nonagenarians and 4,278 octogenarians), with a total sample size of 16,563 elderly aged 65+ plus 391 elders who are younger than age 65. In addition, a middle-aged peers’ survey which contains 3,412 respondents aged 40-59 was performed in the 2008 wave, 697 of which were interviewed in the 7 longevity areas and 2,715 of which were interviewed in non-longevity areas. Due to the effects of the serious earthquake which occurred in China in 2008 and the delay of the NIH competitive supplement grant caused by the international financial crisis, 5.1% of the previously planned 2008 survey interviews were delayed and completed in the first half of 2009, while the majority (94.9%) of the interviews were completed in 2008. Information on before dying was collected from the next-of-kin for 5,209 deceased respondents who were interviewed in 2005 but died before the 2008-2009 wave.

The 2011-2012 wave interviewed 3,149 young elderly aged 65-79 and 6,530 oldest-old aged 80+ (including 1,457 centenarians, 2,433 nonagenarians and 2,640 octogenarians), with a total sample size of 9,679 elderly aged 65+ plus 86 elders who are younger than age 65. In addition, a middle-aged peers’ survey which contains 423 respondents aged 40-59 was performed in the 8 longevity areas in 2012. All the respondents in non-longevity areas were interviewed in 2011 while those respondents, including the middle-aged peers, in the 8 longevity areas were interviewed in 2012. Information on before dying was collected from the next-of-kin for 5,642 deceased respondents who were interviewed in 2008-2009 but died before the 2011-2012 wave.

The 2014 wave interviewed 2,369 young elderly aged 65-79 and 4,738 oldest-old aged 80+ (including 877 centenarians, 1,654 nonagenarians and 2,207 octogenarians), with a total sample size of 7,107 elderly aged 65+ plus 85 elders who are younger than age 65. Information on before dying was collected from the next-of-kin for 2,879 deceased respondents who were interviewed in 2011-2012 but died before the 2014 wave.

Among the total number of 96,843 interviews with the elderly in the seven waves (1998, 2000, 2002, 2005, 2008-2009, 2011-2012, and 2014), 16,547 interviews were for centenarians, 23,232 for nonagenarians, 25,719 for octogenarians, and 19,884 for younger elderly aged 65-79. We also collected data on date/cause of death and disability and suffering before dying through interviews with a close family member for 26,242 deceased elderly. These data combined provide the largest study of the oldest-old ever conducted in the world.

Notes:
The 2008-2009 and the 2011-2012 waves also interviewed and collected DNA samples from 3,412 and 423 middle-age respondents aged 40-59, respectively. These peer respondents have no any blood-relation with any elderly respondents and who are neighboring controls of the centenarians interviewees. We used very short questionnaire for interviewing the middle-age controls to collect data of their basic demographic and socioeconomic conditions. Although the grant total sample size of our CLHLS 2008-2009 wave includes these middle-age controls, it makes no sense to include the middle-age sub-samples in the 20008-2009 elderly dataset, as we did not collect any health and behavior data from the middle-age respondents and their participation in CLHLS is only for the future case-control genetic analysis (the genotyping is not yet funded).