McKeon believes that the concept of QOL can be traced back to Aristotle’s relevant argument that happiness represents a good life (Zhan, 1992). But it was not until after the second world war that the name “quality of life” was widely used. In the initial stage, the QOL only refers to good “material life,” after which scholars add the meaning of “psychological” or “social level” (Farquhar, 1995). The healthcare field began to pay attention to the importance of QOL, which can be traced back to the World Health Organization proposing in 1947 that the definition of health is not only the absence of disease, but also the physical, psychological, and social well-being. However, it was not until the 1960s that the concept of QOL was widely used in health research (Renzaho et al., 2016). In addition, early studies often used the concept of QOL with terms such as happiness, morale, or life satisfaction (Kuyken et al., 1995). However, this practice was later questioned by scholars, who pointed out that these terms are not equivalent to the concept of QOL (Lau and Mckenna, 2001).

In order to clarify the essence of the concept of QOL, scholars have tried to define QOL from subjective, objective, or multi-level perspectives. For example, Sarvimäki and Stenbock-Hult (2000) define the QOL as the subjective feeling of an individual’s health, value, or meaning. Lawton (1991) thinks that QOL is a multi-faceted concept, including objective environment, behavioral ability, QOL understood by individuals, and mental health. Zhan (1992) defined the QOL as: the degree of personal satisfaction with life experience, which includes four factors such as life satisfaction, self-concept, health and function, and socio-economic status. According to the 1995 WHO definition, QOL refers to individual culture and value systems, and experience of survival status related to people’s goals, expectations, standards, and concerns. It is a subjective assessment (Harper and Power, 1998). QOL contains multidimensional and subjective components, so individual, family, social, and other environmental factors will affect the QOL (Dijkers, 1999).

The bio-psycho-social model posits that in addition to biological factors such as bacteria and viruses, some psychological factors can also affect the health status of the human body to a certain extent (Liang et al., 2006; Du, 2009). Personality is one of the important psychological factors affecting the QOL (Diener et al., 1999; Jaiswal et al., 2018). The “Big Five” personality model believes that human personality traits consist of five traits: neuroticism, extraversion, agreeableness, conscientiousness, and openness. The diverse strength of each personality trait can cause different effects on various aspects of the human body (Sutin et al., 2013). Previous studies show that personality traits are closely related to diseases (Cao and Liu, 2017), and personality traits can predict human health outcomes (Yao et al., 2018). Many studies have confirmed that agreeableness (Rassart et al., 2013; Fuente-Arias et al., 2020), conscientiousness (Warrian et al., 2009), and openness (Goodwin and Engstrom, 2002) positively predicted the QOL. Neuroticism can negatively predict the QOL (Harandi et al., 2020; Dedova et al., 2022). Few studies showed that extraversion negatively predicted the QOL (Warrian et al., 2009; Harandi et al., 2020). The vast majority of studies showed that extraversion positively predicted the QOL (Goodwin and Engstrom, 2002; Masthoff et al., 2007; Poppe et al., 2013; Ibrahim et al., 2015). Based on the existing studies, the present study proposed the following hypotheses:

Social support refers to the individual’s ability to adapt to and face difficulties by interacting with the social environment to obtain different levels of support (Lyrakos, 2012). Social support usually comes from important people, such as family members, friends, and partners (Attar-Schwartz et al., 2019). Important people provide personal substantive or emotional help to enable them to face pressure and negative events and complete tasks. When an individual has a good social support and a close social support network, it means that when he is facing difficulties, he can realize that he can get a lot of resources and assistance to tide over the difficulties smoothly (Guo et al., 2020). When an individual has a healthy physical and mental state and a good sense of well-being, his satisfaction with life is also high. Social support can also help individuals buffer against different pressures in work and life and help them adapt. People with high social support have stable and good interpersonal relations and social interaction (Belayneh et al., 2018). Even if they encounter bottlenecks, they can be sure that they are loved and cared for. When individuals face stressful events in life with such a belief, they can play a buffer role, slow down the impact on individuals, and have the power to face challenges and solve problems.

Ecological systems theory suggests that social factors in the macro-system affect the development of people (Bronfenbrenner, 1979). As one of the most significant macro-system in the outermost system of the social environment, social support can significantly impact the QOL (Wang, 2004). The large body of research on social support and physical and mental health shows that social support is closely related to the QOL (Gao et al., 2022). Good social support is beneficial to health, while malignant social relations damage physical and mental health. Social support has a buffering effect on stress and is important for maintaining a good mood. Some studies show that social support affects the QOL of patients with cognitive disorders (Afunugo and Rafael, 2019), and improved social support may enhance the life quality of lung cancer patients (Hofman et al., 2021). The higher the social support level of the elderly, the better the QOL (Moghadam et al., 2020), patients with breast cancer (Zhang et al., 2017), and hemodialysis patients (Alexopoulou et al., 2016).

The social support theory proposes that the influential factors for social support include personal, improvement, and environmental factors. The personality traits in personal factors influenced the degree of social support (Schoch et al., 2018). Empirical studies have found that personality can affect individuals’ perception of social support to a certain extent (Li et al., 2005). Individuals with different personality traits have different perceived levels of social support (Liu and Huang, 2010). Research results based on diverse objects suggest that higher agreeableness can promote the social support perceived by the individual (Wang and Gen, 2016; Yu et al., 2021). Conscientiousness (Barańczuk, 2019), openness (Bao et al., 2019), and extraversion (Olawa and Idemudia, 2020) have a significant positive correlation with social support. Neuroticism has a negative correlation with social support (Olawa and Idemudia, 2020).

From the above, social support can affect the QOL of PWD, but personality directly affects social support and the QOL of PWD. Therefore, this study hypothesized that personality traits indirectly influence the QOL of PWD through social support. The present study proposed the following hypotheses:

The sample size was determined, using Raosoft sample size calculator,¹ based on a margin of error of 5%, confidence level of 90%, population size of 20,000, and a response distribution of 50%. The calculated sample size was 267. The people with disabilities refer to those with visual, hearing, speech, or physical disabilities; mental retardation; mental disorder; multiple disabilities; and/or other disabilities. The questionnaire was completed by the personnel who meet all of the following inclusion criteria:

Recruit participants from individuals living with various types of disability.

This study used the random sampling method to select subjects from 29 provinces and cities of China. In total, 389 questionnaires were distributed, and 358 valid questionnaires were recovered, with an effective rate of 92.03%. Among the 358 respondents, 12 (3.35%) participants were under 19 years old, 218 (60.89%) were between 19 and 59, and 128 (35.76%) were over 59; 165 (46.09%) were male and 193 (53.91%) were female; 51 (14.25%) were single, 218 (60.89%) were married, 43 (12.01%) were divorced, and 46 (12.85%) were widowed; and 61 (17.04%) were visual disabilities, 53 (14.80%) were hearing disabilities, 19 (5.31%) were speech disabilities, 78 (21.79%) were physical disabilities, 13 (3.63%) were mental retardation, 49 (13.69%) were mental disorder, 74 (20.67%) were multiple disabilities, and 11 (3.07%) were other disabilities (Table 1).

Normality of data was tested using D’Agostino and Pearson’s normality test and Shapiro–Wilk normality test. SPSS 24.0 was used for the descriptive statistical analysis and correlation analysis. Mplus version 7.2 was used to examine the mediating effect of social support on the relationship between neuroticism, extraversion, agreeableness, conscientiousness, openness, and QOL of PWD. Missing data were handled using full-information maximum likelihood estimation, and the significance of paths was verified by bootstrapping analysis with 5,000 replicates.

Using self-reporting methods to collect data is likely to produce common method bias. According to the recommendations from Zhou Hao and Long Lirong, we took control measures in the test, such as using reverse presentation for some entries. Before analyzing data, the common method bias was evaluated by using the Harman single-factor test. As a result, there were 6 factors with eigenvalues greater than 1, which explained 66.87% of the variance. The first factor accounted for 35.50% of the total variance, which was less than the critical value of 40% (Podsakoff et al., 2003). Thus, there was no serious problem of common method bias in this study.

Pearson’s correlation analysis (Table 2) showed that neuroticism was negatively correlated with extraversion (r = −0.185, p < 0.001), agreeableness (r = −0.181, p < 0.001), conscientiousness (r = −0.175, p < 0.001), openness (r = −0.114, p <0.05), social support (r = −0.120, p < 0.05), and QOL (r = −0.186, p < 0.001). QOL was positively correlated with social support (r = 0.402, p <0.001), extraversion (r = 0.324, p < 0.001), agreeableness (r = 0.474, p < 0.001), conscientiousness (r = 0.482, p < 0.001), and openness (r = 0.498, p < 0.001). Social support was positively correlated with extraversion (r = 0.251, p < 0.001), agreeableness (r = 0.307, p < 0.001), conscientiousness (r = 0.370, p < 0.001), and openness (r = 0.304, p < 0.001). Extraversion was positively correlated with agreeableness (r = 0.113, p < 0.05), conscientiousness (r = 0.345, p < 0.001), and openness (r = 0.238, p < 0.001). Agreeableness was positively correlated with conscientiousness (r = 0.508, p < 0.001) and openness (r = 0.361, p < 0.001). And a significant positive relationship was observed between conscientiousness and openness (r = 0.379, p < 0.001).

Structural equation modeling was used to test the study hypotheses. The model estimation method was the maximum likelihood estimation method. The structural equation analysis was performed with neuroticism, extraversion, agreeableness, conscientiousness, and openness as the predictors, employment and permanent residence as control variables, social support as the mediator, and QOL as the outcome variable. The significance test was estimated using the bias-corrected bootstrap method and repeated 5,000 times after putting it back.

The results of the hypothesized model showed good overall model fit indices: χ2 = 6.299, df = 5, RMSEA = 0.047 [90% CI: 0.001, 0.093], CFI = 0.995, TLI = 0.980, SRMR = 0.013 (Figure 1).

Further examination of the model used a bias-corrected bootstrap method. Results (see Table 3) showed that the 95% confidence interval of other mediation effects did not contain 0, except for the mediating effect of neuroticism → social support → quality of life. It means other mediation effects were significant, except for the mediating effect of neuroticism → social support → quality of life. Therefore, social support mediated the relationship between extraversion, agreeableness, conscientiousness, openness, and quality of life. And the mediation effects of social support were 0.025, 0.026, 0.039, and 0.028. H2b, H2c, H2d, and H2e are all supported.

The advantages of the study are as follows. Firstly, this is a China-wide cross-sectional investigation, and the selected sample is representative to a certain extent, which makes the research results more credible. Secondly, this study verified the relationship between big five personality and the QOL of PWD. Neuroticism negatively influenced the QOL, while the other four personalities could positively affect the QOL. Thirdly, positive personality traits may help PWD perceive social support. They received help that they perceived, thereby improving their QOL. Therefore, we will take measures to increase social support in the future to increase the QOL of PWD. Fourth, the study found that the mediating influence of social support on neuroticism and the QOL was not significant. It shows that social support alone cannot improve the QOL of disabled people with neuroticism. Thus, we should pay more attention to such populations with disabilities, and other approaches are required to improve their QOL. Finally, the mediation model indicated that considering not only individual but also environmental factors is required to propose systematic and comprehensive programs which can improve the QOL of PWD.

The present study had some limitations, and future studies should address them. First, this study was a cross-sectional study. Although it is advantageous for analyzing the specific relationship between different variables, the lack of longitudinal data is not conducive to inspecting variables in this study, especially testing the direction between them. In subsequent studies, we can investigate whether the relationship between the big five personality, QOL of PWD, and social support will change over time. Second, the data in the present study which collected from self-reports of PWD. The self-report can reflect individuals’ physical and mental conditions, but the self-report is prone to measurement error. In the follow-up study, we can consider obtaining information from all sides to reduce measurement error. Third, openness is a western abstract concept and may be hard to be realized for Chinese, especially elderly people with low income and low academic status. The participants may not express their actual openness status correctly.