Introduction

Shidu parents means parents who have lost their only child as a result of an accident, illness or other unfortunate circumstance. In China, due to the impact of family planning, the nuclear family consisting of two spouses and a single child became the dominant family pattern, resulting in an increase in families with only one child1. Data from the National Bureau of Statistics and the Ministry of Health indicate that the number of Shidu families in China is increasing. It is estimated that approximately 76,000 only children between the ages of 15–30 die annually, thereby impacting an equivalent number of Shidu families2. Though the “Comprehensive Two-Child” policy and “Comprehensive Three-Child” policy were introduced in 2015 and 2021 respectively, the number of only children in Chinese society remains significant3. In 2012, Shidu families were estimated to be around one million, with an annual growth rate of 76,000. Scholars have utilized population surveys and computer simulations to predict that by 2050, the cumulative number of only child deaths could exceed 4.5 million4.

According to the traditional Chinese saying, 'Raise a child for old age’, the only child is usually responsible for the care of the elderly parent, and in the event of the loss of an only child, missing the childbearing age, being unable or unwilling to have another child, or adopting a child, Shidu parents may be faced with a series of problems in the decades to come. According to Fei Xiaotong's triangle family structure theory, the only child's role is pivotal to maintaining family stability and enhancing overall family functionality5. The sudden death of the only child significantly reduces the intensity of social support for Shidu parents. In addition, Shidu parents experience a gradual decline in physiological functions as they age, presenting physical and psychological health problems. Shidu parents often lose confidence and courage in their future life after the death of the only child and suffer deterioration of their mental health, in which symptoms of anxiety, depression, post-traumatic stress disorder (PTSD), and complicated grief are common6,7. Shidu parents in Denmark experienced long-term depression after losing the only child8. The theory of mind–body interaction suggests that human psychology and physiology interact with each other, and long-term psychological problems can lead to the development of somatic diseases, which seriously affects the quality of their lives. It has been shown that the quality of life scores of Shidu parents are lower than those with healthy children9.

Quality of life (QOL) is a subjective, multidimensional concept that includes physical state, psychological well-being, social interactions, and overall personal condition influenced by socio-economic, cultural background and values of an individual10. A higher QOL score indicates that the person has a better quality of life, and conversely, a lower QOL score indicates that the person has a worse quality of life. The QOL of older people have garnered significant attention11,12,13,14, but few scholars have studied the QOL of Shidu parents. There is a need for better monitoring of Shidu parents’ QOL and investigate the factors that affect the QOL of Shidu parents, helping them to regain a happy life.

Several instruments measure QOL, such as EQ-5D, the short-form six-dimension survey (SF-6D) and the health utilities index (HUI). Among these instruments, EQ-5D has the characteristics of simplicity, ease of operation, wide application range, and high reliability, making it one of the most widely used universal scales internationally15,16.

Therefore, in this cross-sectional study, we evaluate the QOL of Shidu parents with the EuroQol five-dimension (EQ-5D) and use the Tobit regression model to explore the factors associated with the QOL of Shidu parents. The primary goal of these analyses is to provide a foundation for enhancing the QOL of Shidu parents, thereby promoting the ‘Active Ageing’ policy and the ‘Healthy China’ strategy.

Methods

Sampling

China is divided into three major economic belts based on differences in natural conditions and economic development levels. These belts include the eastern, central, and western regions17. For this study, a multi-stage, stratified, and cluster sampling method was used based on these economic belts. First, one province was selected from each region, including Zhejiang Province, Anhui Province, and Guizhou Province. Second, one city was selected from each province, including Hangzhou, Wuhu, and Anshun. Specific sampling methods were chosen based on the representativeness and feasibility of the survey in each city. Stratified random sampling was used in Hangzhou and Anshun due to their significant development gaps, while Wuhu adopted cluster sampling. Trained investigators conducted face-to-face interviews and completed the questionnaire item by item. The investigators checked and retrieved the completed questionnaires immediately after the interviews. A total of 651 Shidu parents were recruited and analyzed in this study (Supplementary Information).

The data collection process in Hangzhou, Zhejiang Province involved the use of stratified random sampling from July to August 2017. Based on Gross Domestic Product (GDP), the city was divided into three levels: high, medium, and low. Two districts were chosen from each level for a total of six districts. Additionally, two communities with the highest number of Shidu families were selected from each district using data provided by the District Health and Family Planning Bureau.

In Wuhu, Anhui Province, the data collection process took place from April to July 2018 and involved the use of cluster sampling. The city was divided into four districts: Jinghu District, Yijiang District, Jiujiang District, and Sanshan District. Jinghu District was chosen because of the presence of a high number of Shidu parents. All Shidu parents in the area were surveyed.

Anshun, Guizhou Province utilized a two-step sampling method. First, the stratified sampling method was used to divide the city's four districts into two levels based on GDP. One district was chosen from each level for a total of two districts. Second, the cluster sampling method was used to conduct a full field study in the communities with a high number of Shidu parents in the two selected districts.

The criteria for inclusion in the study were: (i) fathers or mothers who have lost their only child; (ii) aged 45 years or above, considering that most Shidu parents aged 45 and above would not choose to have children again; (iii) ability to communicate or access assistance for completing investigations. Exclusion criteria: (i) People who have moved elsewhere; (ii) People who refused various condolences or investigations.

The survey was conducted by the trained professors, undergraduate students, and local family planning officials who acted as interviewers and assisted the participants in completing the questionnaire. The questionnaire comprised four parts, including socio-demographic characteristics, the Geriatric Depression Scale (GDS-15), the Social Support Rating Scale (SSRS), and the three-level EuroQol five-dimensional questionnaire (EQ-5D-3L).

Measurement

Dependent variables

EQ-5D is a generic measure of health status developed by an international research group, the EruoQol Group. It consists of two parts: the EQ-5D descriptive system and the EQ visual analogue scale (EQ-VAS) evaluation. The EQ-5D descriptive system is composed of five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension is subdivided into three levels (no, moderate, severe), represented by values of 1–3. As a result, a total of 243 (35) health conditions in this system. Each health condition can be converted to a specific score according to the Chinese EQ-5D-3L value set18. The range of EQ-5D utility score is from − 0.149 to 0.961. In the EQ-VAS part, the respondents evaluate their own health status using a visual analogue scale where 0 was used to represent the worst health condition, 100 was used to represent the best health condition. In the above two parts, the higher the score, the higher the QOL.

Independent variables

The research about the QOL of the elderly was widely conducted in other countries19,20,21,22. Combining the findings of scholars, the factors influencing the QOL of older people can be summarized in three areas: demographic characteristics, health status, and psychosocial factors. From these three areas independent variables were selected and included variables that are relevant to Shidu parents.

The demographic characteristics included gender, age, city, educational level, marital status, monthly income, having grandchildren. Considering that the city, educational level, monthly income are unordered multi-categorical variables, dummy variable coding is used.

Chronic disorders or diseases were listed for reference, including hypertension, diabetes, malignant tumour, hyperlipidemia, cerebral infarction (stroke), coronary heart disease, chronic liver disease, cerebrovascular disease, senile dementia, gout, asthma, tuberculosis, arthritis, gynaecological disease, haematology, chronic low back pain, osteoporosis and cataract. The numbers of chronic diseases were categorized into four groups (0, 1, 2, and ≥ 3). Meanwhile, self-reported on health status are used to provide on overall assessment of the health status of the participants.

The psychosocial factors are categorized as social support and depression. Social support was expressed by Social Support Rating Scale (SSRS) which includes 10 items from 3 dimensions (objective support, subjective support and utility of support). Higher SSRS scores suggest a higher level of social support. A total score < 20 means a low level of social support, 20–30 means a fair level of social support while > 30 shows satisfactory social support. The Cronbach's α coefficient for the 10 items and the total score of the questionnaire were analysed to be 0.825–0.896, indicating that the SSRS has good reliability in China23. And depression was expressed by Geriatric Depression Scale-15 (GDS-15) which is used to assess depressive symptoms in older adults. There are 15 questions on GDS-15, and the answer “yes” to each question will get 1 point, with a total score of 15, in which 0–3 is classified as non-depression, 4–8 as minor depression, and 8–15 as severe depression. The Cronbach’s α coefficient of GDS-15 was 0.793, and the item-total correlation was 0.196–0.534. Therefore, the GDS-15 has acceptable reliability and validity for the study of depression symptoms in China24.

Statistical analysis

The characteristics and QOL of the participants assessed by the EQ-5D-3L are summarized using descriptive statistics. The median and interquartile range (M(QR)) are used for continuous variables, whereas the frequencies and percentages are used for categorical variables.

Due to the EQ-5D-3L utility scores are non-normal distributed, non-parametric test analyses were performed to explore the differences in the utility scores of patients with different characteristics, including the Mann–Whitney U rank sum test for comparison between two groups and the Nemenyi method for multiple groups.

Based on the previous introduction, the EQ-5D consists of two main components: the EQ-5D Descriptive System and the EQ-5D Visual Analogue Scale (VAS), and there are differences in the quality of life obtained due to the differences in the way they are measured. China guidelines for pharmacoeconomic evaluations recommended the EQ-5D utility score as the preferred method25. Since the EQ-5D scale has a ceiling effect, this study suggested that combining the two scores and conducting Tobit regression model analysis with each of them as a dependent variable would allow for a more comprehensive and specific study of the factors influencing the quality of life of Shidu parents. The independent variables were chosen from those with statistically significant (P < 0.05) in the non-parametric test analyses. The multiple collinearity of the independent variables was tested using the variance inflation factor (VIF) before the empirical analyses. SPSS 22.0 software was used for the main statistical analyses and Stata 14.0 software was used for the Tobit regression. Differences were considered statistically significant at P < 0.05.

Ethics approval and consent to participate

Ethical review and approval was not required for the study in accordance with the local legislation and institutional requirements. The Medical Ethics Committee of Zhejiang Chinese Medical University decided that a full review was not necessary, and decided to grant ethical approval waiver. The participants provided their written informed consent to participate in this study. All participants were assured of their right to refuse to participate or to withdraw from this study at any time. Confidentiality and anonymity of the participants were also assured.

Results

EQ-5D index score and EQ-VAS score

As shown in Table 1, among the 651 Shidu parents who completed the questionnaire, Shidu parents exhibited a higher proportion of “no problems” in EQ-5D self-care and usual activities, with percentages of 96.16% and 91.86% respectively. Meanwhile, 59.9% Shidu parents reported no problems in all five dimensions, whereas 3.7% reported severe problems in at least one dimension. Pain and discomfort were the most common problems among Shidu parents. Problems with anxiety and depression accounted for 20.89%, while the problem of self-care only accounted for 3.84%.

Table 1 Problems reported by Shidu parents in the EQ-5D dimensions[n (%)].
Full size table

The EQ-5D index score had a median of 0.961, an interquartile range of 0.114, a mean of 0.887, and a standard deviation of 0.119, with scores ranging from − 0.149 to 0.961. The EQ-VAS scores ranged from 10 to 100, with a median of 78, an interquartile range of 15, a mean of 72.66, and a standard deviation of 13.62. The EQ-5D-3L and EQ-VAS scores distributions tended towards the right, as shown in Figs. 1 and 2. Therefore, some traditional statistical methods are no longer applicable.

Figure 1
figure 1

Distribution of the EQ-5D-3L utility scores of Shidu parents.

Full size image
Figure 2
figure 2

Distribution of the EQ-VAS scores of Shidu parents.

Full size image

Characteristics of the participants

Among the 651 participants, the majority were aged 60 or older (464, 71.3%). Females comprised over half of the sample (352, 54.1%)..Participants with a junior high school education or lower accounted for 71.7%, whereas only 9.1% had attained a college education or above. The majority were married and those with a monthly income of more than 3000 accounted for 42.7%. About 79.9% of the participants reported having grandchildren. Meanwhile, almost 71.8% participants suffered from at least one chronic disease in which 60.4% were found to be suffering from depression.

Non-parametric test analysis results of EQ-5D scores and EQ-VAS scores of Shidu parents

As shown in Table 2, those Shidu parents who were older than 60, lived in Anshun, had a junior high school education or below, divorced or bereaved of one’s spouse, did not have grandchildren, had poor self-reported health status, had more than or equal to three chronic diseases reported lower EQ-5D-3L scores (P < 0.05). Meanwhile, there were no significant differences in the EQ-5D-3L scores found in gender, monthly income (P > 0.10).

Table 2 Comparison of the EQ-5D and EQ-VAS scores with different characteristics.
Full size table

Those Shidu parents who were older than 60, lived in Wuhu, had a junior high school education or below, had monthly income less than 2000, did not have grandchildren, had poor self-reported health status, had two chronic diseases reported lower EQ-VAS scores (P < 0.05). Meanwhile, no significant differences in the EQ-VAS scores were found in gender, marital status (P > 0.10).

Tobit regression model analysis results of EQ-5D scores and EQ-VAS scores of Shidu parents

The covariance test results showed that the tolerance values of all control variables were > 0.3 and the VIF values were < 5, indicating no covariance between the control variables.

The Tobit regression model confirmed that GDS-15 scores, marriage, education, self-reported health status were significantly associated with the EQ-5D index scores (P < 0.05) of Shidu parents, whereas city and third generation were not significantly related to the EQ-5D index score. Furthermore, the Tobit regression model revealed that, compared to the reference group, Shidu parents who were married, had lower GDS-15 scores, had higher education levels, good self-reported health status, and no chronic diseases had higher EQ-5D scores (P < 0.05), as shown in Table 3.

Table 3 Tobit regression analysis of the factors affecting EQ-5D index scores of Shidu parents.
Full size table

By conducting a Tobit regression model analysis with the EQ-VAS score as the dependent variable and demographic characteristics, GDS score, and SSRS score as independent variables, the result of Tobit regression model analysis as shown in Table 4 confirmed that the EQ-VAS scores of Shidu parents were significantly associated with SSRS scores, GDS-15 scores, self-reported health (P < 0.05). According to the coefficients of the variables, Shidu parents who had good self-reported health status and did not have any chronic disease were remarkable higher than others on EQ-VAS scores.

Table 4 Tobit regression analysis of the factors affecting EQ-VAS scores of Shidu parents.
Full size table

Discussion

To the best of our knowledge, only a few studies have been conducted on the QOL of Shidu parents. Chen et al. measured the QOL of elderly people in Nanjing, with a mean EQ-VAS score of 77.2226. Moreover, Zhang et al. assessed the QOL of residents in Jiangxi Province using the Chinese EQ-5D-3L scale and found an average health utility value of 0.91427. Similarly, Guan et al. conducted a survey in urban and rural areas of Beijing, Shenyang, Chengdu, and Nanjing to measure the QOL of Shidu parents, revealing mean scores for EQ-5D-3L and EQ-VAS of 0.951 and 88, respectively28. In comparison, the average EQ-5D and EQ-VAS scores of Shidu parents in this study were 0.887 and 72.66, respectively, indicating a lower QOL compared to the general population. This suggests that the QOL of Shidu parents is lower than that of the general elderly population. Therefore, it is necessary to focus on the QOL of Shidu parents to improve the overall QOL in society.

Our results demonstrated that pain/discomfort was the most frequently reported problem among Shidu parents (23.35%) in the five dimensions. Shidu parents are generally older, and the elderly tend to have weakened immunity and metabolic functions, making them more prone to chronic diseases such as cardiovascular diseases, diabetes, and hypertension, which may cause them pain/discomfort and seriously impact their physical health. Anxiety/depression was the second most frequently reported problem among Shidu parents (20.89%) in the five dimensions. Chinese Shidu parents have a higher prevalence of depression compared to other middle-aged and elderly individuals29. Therefore, it is vital to pay more attention to the psychological burden experienced by Shidu parents and find strategies to relieve anxiety and improve depressed mood.

Among the demographic characteristics, we found that age is one of the influencing factors on the EQ-5D index scores of Shidu parents. This has been consistently reported in previous studies30,31. Older people usually have reduced immunity and metabolic function, and multiple illnesses often require the use of multiple medications, leading to higher chances of adverse drug reactions and further health damage, ultimately reducing their QOL32. Moreover, married Shidu parents tend to have higher EQ-5D index scores than those who are divorced or bereaved of their spouse. This aligns with previous research that found marriage to be a significant factor influencing the well-being of older adults, both in terms of physical and emotional health. The absence of a spouse due to death or divorce can lead to a diminished QOL as the individual may experience loneliness and a lack of support33,34. Furthermore, the Tobit regression model showed that Shidu parents with higher education tend to have higher EQ-5D index scores and EQ-VAS scores. This is likely because individuals with higher education are more willing to access better medical services and possess greater personal health awareness, leading to an improved QOL35,36.

In addition, the study found that having grandchildren was positively correlated with the EQ-VAS scores of Chinese Shidu parents, indicating that involvement in the lives of the younger generation can prevent these individuals from becoming fully immersed in their grief37. The community needs to address marital problems and provide support for those who are struggling. Additionally, under the three-child policy, Shidu parents who are able and willing should be encouraged to have another child, while social security systems should be established to provide essential services such as medical insurance reimbursement for assisted reproductive technology, free prenatal testing, and door-to-door testing services. For those who cannot have another child, adoption through social welfare institutions should be encouraged. By taking proactive steps to address the needs of Shidu parents, we can help improve their overall QOL. Furthermore, the study also revealed that the QOL of Shidu parents in Hangzhou was better than those in Anshun and Wuhu, likely due to the lower level of economic assistance and emotional support provided by the government and society in western and central cities. To meet the demand for health services, individuals must be willing and able to pay for them38. According to health economics theory, economic factors play a significant role in promoting health. Therefore, it is necessary for government to increase the living allowance for Shidu parents, particularly those who suffer from major illnesses and have low or no retirement wages. The government can also provide direct purchasing of services or preferential policies to offer further protection.

Health status is a good indicator of QOL, as higher ratings correlate with better QOL for Shidu parents. Shidu parents with better self-reported health status and fewer chronic diseases had better QOL according to the Tobit regression model. For this factor, we should pay attention to the health status of Shidu parents. For example, communities can prioritize Shidu parents for family doctor appointments, maintain long-term health records, and have general practitioners provide regular health checks, education, and care to maintain and improve their physical health. If community health services are not effective, a higher-level hospital should be referred to with special consideration for the unique needs of the recipients.

In the Tobit regression model, the GDS-15 scores showed a significant association with both the EQ-5D index scores and the EQ-VAS scores. There is no doubt that psychological problems are fundamental for Shidu parents39,40,41. Therefore, we highlight the importance of paying more attention to psychological problems of Shidu parents. The community can provide psychological hotlines, counselors, or therapists to offer personalized and timely support for those in need of psychological treatment to help them overcome the psychological effects of losing their only child. Meanwhile, the results of the Tobit regression model of the EQ-VAS scores were consistent with previous studies, revealing that the QOL of Shidu parents is positively correlated with their level of social support, as measured by the SSRS score42. Social support can help those who have lost their only child by providing a sense of normalcy and reducing feelings of sadness and isolation. Informal support, in particular, can help compensate for emotional deficiencies and aid in overcoming psychological difficulties43. To seek support, Shidu parents should actively engage in communication and seek help from family, friends, and community services when needed. Despite community family planning cadres being available to support Shidu parents, they may face challenges due to limited time and resources. To address this, the government can purchase social organization services and provide specialized support in the form of door-to-door condolences, medical treatment, outdoor activities, and domestic assistance. Shidu parents are more likely to seek help from familiar, professional, and timely social workers44. We can derive inspiration from developed countries such as the United Kingdom, the United States, Japan, and Singapore, who have established mature social care systems for the elderly. A similar model of community care for Shidu families can be established, and informal social support entities such as neighbours, college students, and the general public can be encouraged to volunteer their assistance.

Limitations

There were several limitations in the study that deserve mention. First, the EQ-5D-3L was used as a quality of life measurement tool in this study, and the newer EQ-5D-5L will be considered for future research to obtain more accurate results. Second, the differences in sampling methods in the survey of Shidu parents increase the risk of bias. Third, this study solely focused on Shidu parents and did not include any investigation of non-Shidu parents, resulting in the inability to perform comparative analyses. Future research should incorporate comparative studies to enhance the study's findings. Last but not least, as a cross-sectional survey, our results cannot establish a causal relationship between Shidu parents and QOL.

Conclusion

In summary, this research measured the QOL of Shidu parents using the EQ-5D-3L and explored the influencing factors using a Tobit regression model. Worse QOL for Shidu parents was associated with high depression, lack of marriage, grandchildren, education, and poor self-reported health status. This study highlights that both personal and social factors impact the QOL of Shidu parents. Therefore, it is important to not only focus on individuals, but also to provide strong and professional social support and generous social welfare to improve their QOL. Recommendations include improving the national welfare system with a focus on Shidu families, and creating a group composed of street community leaders, community doctors, psychological counsellors, and volunteers to help Shidu parents improve their health, expand their social networks, and overcome their psychological challenges.