The China Health and Retirement Longitudinal Study (CHARLS) is a landmark survey of the middle-aged and elderly in China. Spearheaded by Peking University’s National School of Development, the study aims to collect multidisciplinary data, ranging from socio-economics status to health conditions, to be used to support the scientific analysis of China’s ageing issues. The Charls baseline study polls a nationally representative sample of more than 17,700 individuals from more than 10,000 households, in 150 counties/districts in 28 of China’s 30 provinces (excluding Tibet). The individuals will be followed up every two years.
Vast study seeks to tackle China's age challenge
Unprecedented survey of 17,000 people seeks to assess impact of social reforms on the lives of people in the world's largest ageing society
A landmark study on ageing population on the mainland hopes to provide answers to the raging debate if China will grow rich before it grows old and if the country’s nascent social safety net is equal to the tasks ahead.
The China Health and Retirement Longitudinal Study (CHARLS) polled 17,708 individuals from 10,287 households in 28 Chinese provinces between May 2011 and March 2012. The sheer scope and detail of the data collected, ranging from socio-economic aspect to physical and psychological health, is unprecedented in China.
The first major report from this baseline survey will be released today by Peking University in Beijing, which conducted the research together with other foreign institutes.
“China is going through a period where there is enormous [social] reform going on. Health insurance is being provided to people, government pensions are just starting to come into play,” says Professor James P. Smith, a member of CHARLS’ international research team and the Distinguished Chair in Labour Markets and Demographic Studies at RAND Corporation.
“We don’t know yet what the real impact of all that is on people’s lives – and this is why doing CHARLS now is so important.”
The study measures the existence and impact of these social safety nets at both the household and community levels. It interviewed one person per household aged above 45 and their spouses to track the participants into retirement. The researchers will conduct follow-up study every two years, with the second one scheduled to begin this summer.
All raw survey data has been made public on the CHARLS website, which Smith says breaks the tradition in China. “It’s so unusual for a high quality dataset on China to be publicly available. It’s going to have enormous impact,” says Albert Park, one of the study’s principal investigators and an economics professor from the Hong Kong University of Science and Technology.
CHARLS joins a family of 30 other well-established international ageing studies that are the most influential studies in the world on ageing issues.
The first of these studies was the Health and Retirement Study (HRS) in the US, which started in 1992 and tracks its participants every two years. There have been more than 2,000 papers written using the HRS and published in leading journals in health, economics and demographics.
HRS sister surveys, which share key principles and features, are also in Mexico, England, Ireland, Costa Rica, New Zealand, Indonesia, South Korea, India, Thailand, Japan, and Europe (as a 19-country network survey).
Smith was involved in getting the HRS started, and in 2005 initiated the setting up of CHARLS. “China was an obvious country for such a study because it is ageing at a very rapid rate,” he says.
China has the largest ageing population in the world: about 185 million people aged 60 and over at the end of 2011, according to the National Bureau of Statistics.
The UN projects the share of this elderly population will increase from 12 per cent in 2010 to 34 per cent in 2050. The median age of the population will be 49 years in 2050, up from 24 in 1950. In 2000, there were more than 12 Chinese potential workers for every person 65 years and older. By 2050, the ratio will be about two to one.
Chinese women had 6.11 children each on average back in the early 50s. In 2011, the rate was 1.6 – and it’s projected to stay that way. Life expectancy has risen from about 45 in 1950 to 76 in 2011.
The mainland’s dramatic demographic transition to an ageing society has taken only about 40 years, compared to more than 100 years in such developed countries as Britain, the US, and the Nordic countries.
Because of its family planning policies in the past, China faces rapid population ageing at a much earlier stage of development than other countries, Park says.
“People are coming of age now at a time before pension systems are very mature and well-managed,” he says. “It’s not clear whether the elderly in the future are going to have enough resources to manage their needs in retirement. They are going to have less support from children too, because there are just fewer children,” Park says.
But more than just influence policy issues, CHARLS will also shed light on the health status, medical care, material well-being, family support, and work and retirement behaviours of China’s elderly – and how these interdependent life domains affect each other.
It will also allow for cross-national comparisons, and help countries learn from each other’s experiences. Park says there’s research in progress that compares data from England’s ageing study with CHARLS. “Preliminary results show that across almost every measure of physical and psychosocial well-being, the Chinese don’t do as well,” he says.
Apart from self-rated measures of health, the CHARLS survey also included biomarkers such as physical measurements, performance tests and blood sample analysis. This is particularly critical in China due to widespread undiagnosed diseases, including hypertension and diabetes, and low awareness of health problems.
“The goal of CHARLS is to create something very valuable for the long term,” says Park. “So every step of the survey was administered at an extremely high level of quality, care and professionalism, with many layers of checks on data quality and real efforts to maximise response rates.”
The overall response rate was over 80 per cent – 94 per cent in rural areas and 69 per cent in urban areas. That’s “very high from an international perspective”, Park says; many surveys in Hong Kong would have a rate lower than 50 per cent.
Getting people to agree to do the survey – which took on average 94 minutes to complete (excluding biomarker measurements) – was probably the greatest of many challenges faced in the study. “We worked really hard for those rates,” says Park. “The richer people get, the more valuable their time is, the less they’re willing to sit down and do a survey.”
Some 150 mappers/listers and 300 interviewers, mostly university students, were recruited to conduct the survey. Working with community leaders, posters and letters were used to inform people of the survey and its purpose. Personal introductions helped get a better response.
The project is led by an international team of first-rate academic economists and epidemiologists, with guidance provided by domestic and international advisory boards that include leading ageing experts from both inside and outside China.
It was financed by major grants from Peking University, the National Natural Science Foundation of China, the US National Institutes of Health, and the World Bank. The researchers declined to reveal the total funding amount.
Given the complexity of doing a survey at the national level in China, a pilot panel survey was done in 2008 (and followed up in 2012) in two provinces, Gansu and Zhejiang. The pilot data, from 2,685 individuals living in 1,570 households, has been available since April 2009. More than 2,600 users, including about 25 per cent from outside China, have downloaded it.
About 40 papers have been written based on the pilot data. Park has co-authored a few, including one which revealed that 15 per cent of the elderly consume at levels below the World Bank’s US$1.25 per day international poverty line. Poverty rates were higher and consumption significantly lower for the elderly compared to those aged 45-60.