Depression is a major public health problem that can create a variety of emotional and physical problems. Using the large-scale nationally representative survey data, this study measured the prevalence of depressive symptoms among the elderly in China. The result of this study found that the proportion of older adults with depressive symptoms in China had reached 32.55% in 2013, much higher than a previously reported 14.81% from a meta-analysis on the prevalence of depressive symptoms in the 1980s and early 1990s [26]. Two reasons could partially explain the dramatically increased prevalence of depressive symptoms in the past few decades. First, with the phenomenal economic development in China in the past three decades, the accelerated pace of modern life and increasing pressures of living, coupled with unhealthy life-styles increased the risk of suffering from depressive symptoms. Second, although many health promotion interventions, such as free physical check-ups for elderly people and free operations for poor cataract patients, have been provided by the Chinese government to improve the health status of the elderly since 2009, measures aimed at alleviating depressive disorders were seldom conducted [27]. When left untreated, depression can have serious consequences, affecting every aspect of elderly people’s lives. Therefore, policy interventions aimed at relieving the symptoms of depression in the elderly should be of interest to the Chinese government.
Cross-country comparison showed that the prevalence of depressive symptoms among the elderly in China was also higher than that reported in most studies carried out in other countries and areas [19, 20, 28–30]. However, due to methodological differences, including variation in samples studied, the screening scale used and the cutoff point adopted, the cross-country comparison should be interpreted with caution.
The present study explored the association between a variety of socio-demographic variables and depressive symptoms. Consistent with most previous studies, our study found that increasing income was negatively associated with suffering from depressive symptoms, whereas lower education attainment was positively associated with depressive symptoms. Although a few studies found insignificant association between sex and depressive symptoms, in agreement with most previous studies, we found that women had a higher risk of suffering from depressive symptoms than men [21, 31]. This may in part be owing to the disadvantaged sociocultural roles and psychological attributes related to women’s greater vulnerability to life events and depressive symptoms [32, 33]. The relationship between age and depressive symptoms are inconclusive. Most studies pointed out that age is a risk factor for depressive symptoms. However, unlike most studies, our study found that age was inversely associated with depressive symptoms. Some published studies support our findings. Blazer et al. found that increased age was positively associated with depressive symptoms among the elderly over 65 years old in bivariate analysis; however, after the confounding variables were simultaneously controlled, this relationship was reversed [34]. Geographically, residence in the rural-west, rural-central, rural-east, urban-west and urban-central areas were all associated with increased risk of suffering from depressive symptoms, compared to living in urban-east areas. That may be because living in economically undeveloped areas implies less access to depression treatments and lower social participation. Our findings showed that more than 9% of older adults in China are living alone. Although one previous study conducted in China found that living alone was positively associated with the health status of the elderly, our study found that those living alone had a higher risk of having depressive symptoms [35]. In line with previous studies, our study found that social activity was negatively associated with an increased risk of suffering from depressive symptoms, whereas physical disabilities, brain damage and vision problems elevated the risk of suffering from depressive symptoms [29, 30]. The reason why such disabilities increased the risk of depressive symptoms is that these long-term, severe body disabilities can result in chronic strain, which was demonstrated as being associated with depression. A poor elderly person with residence in rural-west areas, living alone, illiterate, not participating in any social activity, and having physical disabilities, brain damage and vision problems had the highest probability of having symptoms of depression.
A small number of studies have been published on income-related inequality in depressive symptoms internationally [36, 37]. These existing studies retrieved the data from different countries, aimed at different populations, and using different measures of depressive symptoms, so straight comparison should be cautioned. However, these studies, as well as our study, consistently revealed that depressive symptoms are unequally distributed among the income spectrum, with the poor having the higher probability of suffering from depressive symptoms [37].
As expected, most demographic, socioeconomic and geographic characteristic variables showed a positive percentage of contribution to the overall inequality, and the probability of depressive symptoms was more concentrated among the poor. Among them, our findings are in agreement with previous studies, that income made the largest percentage of contribution to pro-rich inequality [36]. Despite income, residence in rural areas (rural-west, rural-central, and rural-east) and being illiterate also exhibited a substantial percentage of contribution to pro-rich inequality. Living in rural areas and being illiterate was more concentrated among the poor elderly. Although being a woman had a higher probability of having depressive symptoms, the contribution of gender to the total inequality was not large. That is because this variable was almost evenly distributed between the advantaged and disadvantaged. The contribution of age, living alone, and working status (i.e. still working) were to reduce the magnitude to which depressive symptoms were concentrated among the worse off. Our findings showed that although statutory retirement age for ordinary workers in China is 60 for men and 55 for women, more than 54% of the elderly still engaged in work due to a variety of reasons, and those people engaged in work are more concentrated among the poor. Regarding life-style variables investigated in this study, social activity also made a substantial contribution to the overall pro-rich inequality in depressive symptoms, and deserves more attention.
Appropriate policy and intervention strategies should be implemented to reduce the income-related inequalities in depressive symptoms. Possible strategies are as follows: firstly, disadvantaged elderly people (lower income, illiterate, etc.) and vulnerable residential areas should be identified. Secondly, narrowing the gap between the rich and the poor through redistributing income measures is a feasible way to reduce pro-rich inequality. Since the compulsory health insurance in China has the redistributive features, optimal health insurance is a feasible way to decrease income inequality. China’s basic health insurance mainly consists of the Urban Employee Basic Medical Insurance (UEBMI) which is designed for urban residents, the Urban Resident Basic Medical Insurance (URBMI) which is designed for unemployed urban residents, and the New Rural Cooperative Medical Insurance (NRCMI) which is designed for the rural population. Elderly residents covered by UEBMI no longer need to pay a premium after they have retired, and have a higher level of insurance protection, while conversely, insured elderly covered by URBMI and NRCMI still need to pay fixed premiums, and have relatively lower levels of insurance protection. Therefore, establishing and improving the urban-rural integrated medical insurance system should be explored. In addition, enhancing the medical assistance system, which renders assistance to depressed people in lower income families or vulnerable areas, can also help reduce pro-rich inequalities. Thirdly, since lower education not only increases the risk of depressive symptoms, but also makes a substantial contribution to income-related inequality, lower education needs to be addressed and given larger investment. This is especially true for elderly women, in that 48.30% of them are illiterate. Education programs on depression for illiterate elderly people should be emphasized. Fourthly, considering large contributions of social activity to pro-rich inequality in depressive symptoms, encouraging poor elderly to engage in all sorts of social activities is of importance to reduce this contribution.
There are some limitations in our study. Firstly, because of the cross-sectional design of our study, causal conclusions cannot be reached. Secondly, all of the data employed in our study were self-reported. Thirdly, recall bias, which can affect the study results, may exist in our study. To limit recall bias, different recall periods were used in CHARLS, i.e. one week was used for food expenditure, one month for communication, local transportation, entertainment fee, and one year for clothing, heating, education and training, and automobiles.