Assessing the impact of ICT adoption on residents' self-rated health: evidence from China

Purpose Utilizing data from the 2017 Chinese General Social Survey (CGSS2017), the paper aims to investigate the impact of information and communication technology (ICT) adoption on residents' self-rated health and reveals the mechanisms behind ICT. Design/methodology/approach In the study, ICT adoption is defined as a dummy variable, which takes the value of one if respondents adopt the computers or mobile phone. Meanwhile, respondents' perceptions on five categories of self-rated health are used to construct the dependent variable. Then, based on a fixed-effects regression model, the ordinary least squares (OLS) and ordered probit approaches are applied to estimate their association. Moreover, the two-stage least squares (2SLS) and instrumental variable (IV)-oprobit methods are used to tackle the potential endogeneity of ICT adoption. Finally, the heterogeneity across individuals and regions as well as the underlying mechanisms are discussed. Findings The results indicate that ICT adoption significantly improves residents' self-rated health, which confirms the health utility model with ICT adoption. The conclusion is robust after overcoming the endogeneity issues with IV. In addition, heterogeneity analysis shows that ICT adoption is more beneficial to the health of residents who are male, young, better educated and those who live in the rural areas and in central and western China. Furthermore, the study demonstrates that ICT adoption for searching health-related information and improving social capital are two crucial mechanisms underlying its health effects. Practical implications The findings of this research can help Chinese Government improve population health by issuing corresponding digital and health policies at the regional and individual level. Originality/value First, the study provides fresh microscopic evidence on health outcomes of ICT adoption based on data from the latest wave of CGSS2017. Second, individual and regional heterogeneity is extensively discussed in contrast to most related macro studies that consider average effects. Third, the study addresses underlying mechanisms that have not been thoroughly tested or studied primarily on a theoretical level.