Effects of chronic diseases on health related quality of life is mediated by sleep difficulty in middle aged and older adults

Abstract

Middle-aged and older adults with chronic diseases are more likely to encounter sleep difficulty and have a reduced Health-Related Quality of Life (HRQoL), but there is little research on their possible mechanisms. Therefore, the main objective of this study was to explore how sleep difficulty mediates the impact of chronic diseases on the HRQoL of middle-aged and older adults. The survey data were from a cross-sectional study carried out in 2019 in Hangzhou, China. We used a multi-stage cluster random sampling method to recruit participants from seven districts in Hangzhou. Multiple regression was used to analyze the relationship between chronic diseases, sleep difficulty and HRQoL. And the mediate package in the R language was used to analyze the mediating effect. A total of 3,550 middle-aged and older adults were enrolled, including 2,273 patients with chronic diseases and 1,277 patients without them. Patients with chronic diseases had lower health utility values (β=-0.0084, P < 0.01) and were more likely to suffer from sleep difficulty (β = 0.5737, P < 0.001). After correcting for demographic and life characteristics, the mediation analysis results indicated that sleep difficulty mediated the relationship between chronic diseases and HRQoL (β=-0.0022, 95% Bca CI: -0.0034 -0.0014). Additionally, sleep difficulty influenced the association between chronic diseases and daily activities as well as pain (or discomfort) (β = 0.0083, 95% Bca CI: 0.0042–0.0111; β = 0.0162, 95% Bca CI: 0.0107–0.0225) in the analysis of the dimensions. Sleep difficulty partially mediated the relationship between chronic diseases and HRQoL, and primarily in the daily activities and pain (or discomfort) dimensions in middle-aged and older adults.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-86420-1.

Introduction

With the aging of the population, the disease burden caused by chronic diseases has become the primary public health problem facing the international community and an obstacle to achieving the goal of a “healthy China” in China. On the other hand, socialization, mental health and environmental adaptation needs are also aspects of old age that cannot be ignored. Therefore, taking positive measures such as physical activity and social networks to maintain or improve the quality of life of middle-aged and older chronic patients becomes essential to age actively and healthily^1,2. According to statistics, the global rate of multiple chronic diseases among the elderly aged ≥ 65 years is 40–60%. In 2013, the prevalence of chronic diseases in China was 23.54%, 38.90%, and 53.99% for people aged 45–54, 55–64, and ≥ 65 years, respectively³. In 2017, the number of deaths caused by cancer, cardiovascular disease, chronic respiratory issues, and diabetes in China reached 8.1 million⁴. Additionally, some studies suggest that chronic diseases such as hypertension and diabetes are included in the top five global mortality risks in 2019⁵.

Health-Related Quality of Life (HRQoL), a multi-level and multi-dimensional indicator, can comprehensively evaluate the impact of treatment on patients in terms of physiological, psychological, and social functioning⁶. It was widely applied to the study of chronic diseases and considered as a key indicator for the evaluation of health status. As a quantitative indicator of HRQoL, the health utility values reflect people’s preference for a given health state. There are several preference-based measurement tools for health utility values, such as the EuroQol five-dimensional questionnaire (EQ-5D), the 36-item Short Form Health Survey questionnaire (SF-36), and the Short Form-12 Health Survey (SF-12)^7,8. The EQ-5D instrument has been used in various contexts, including outcomes in clinical practice, evaluation in health economic research or population health^9–11. And the validity of the scale has been demonstrated¹². Moreover, the scale has a utility value conversion system based on the Chinese population. Therefore, it has been widely used to assess the HRQoL of Chinese people. According to research using the EQ-5D scale to measure health status, those with chronic illnesses had lower health utility values than healthy individuals¹³.

Sleep difficulty, such as difficulty falling asleep or maintaining sleep, is a widespread problem for older adults. Studies have found that older adults report sleep difficulty more commonly than younger adults^14,15. Sleep difficulty among older adults results from a combination of age and disease¹⁶. For middle-aged and older adults, chronic diseases are also an important factor contributing to the occurrence of sleep difficulty^17–19. Those with arthritis were 23% more likely than people without arthritis to experience sleep difficulty²⁰. Additionally, a study has found a strong link between cardiovascular diseases and sleep disorders²¹. Recently, evidence suggests that this link may be due to the fact that chronic disease itself affects the neuronal circuits that control sleep and wakefulness, in addition to the associated psychological stress²². It was shown that sleep difficulty is associated with decreased HRQoL in patients with muscle fibromyalgia²³. In addition, previous studies found sleep difficulty relevant to negative emotions, mood, and pain, all of which have been linked to HRQoL²⁴. But few studies have assessed the effect of sleep difficulty on HRQoL by affecting which specific dimension using the EQ-5D scale.

Sleep difficulty is key diagnostic criteria for insomnia disorder. However, the diagnosis of insomnia must be accompanied by daytime dysfunction or the experience of daytime distress in addition to the symptom of sleep difficulty, which can be assessed using standardized scales^25,26. The concept of sleep difficulty has also been frequently used in other studies^27,28. Assessing sleep difficulty through one or a few questions is relatively easy to accomplish for a large survey city and gives a true sense of the respondent’s sleep state^29,30.

Previous research has primarily focused on the effects of specific chronic diseases on HRQoL, such as diabetes and breast cancer^31–34. In addition, Krawczyk-Suszek et al. only analyzed the factors influencing HRQoL in people over 65³⁵. However, few studies have explored the possible mechanisms that conduct the effects of chronic diseases on HRQoL. Consequently, there is a lack of targeted measures to improve the quality of life of patients with chronic diseases.

This study aims to evaluate how sleep difficulty mediates the impact of chronic diseases on the HRQoL of middle-aged and older adults using the EQ-5D scale and assess the role that sleep difficulty plays in the effects of chronic illness on the dimensions of the EQ-5D scale.

Materials and methods

Participants and survey design

The data for this study were obtained from a cross-sectional survey of Chinese residents in Hangzhou. The survey was conducted from July to December 2019. Meanwhile, a multi-stage cluster random sampling method was applied. Stage 1: 8 districts (counties) of Hangzhou (4 in urban areas and 3 in rural areas) were selected by stratified random sampling method according to economic level and geographical distribution. Stage 2: 5 townships (streets) were randomly selected from each district (county). Stage 3: 5 villages or neighborhood committees were randomly selected from each township (street), and each village or neighborhood committee was divided into several groups of villagers or residents with 50 households as units. Stage 4: One of the villagers or resident groups is randomly selected. Residents aged ≥ 15 years whose date of birth was closest to the date of the survey were selected as respondents. Some individuals who were unable to participate in the survey due to health issues like dementia or deafness were excluded.

The survey information was collected through home visits and face-to-face questioning by trained and qualified staff of the community health service centers. A quality assessment team will be set up at each survey site, and at least 10 survey forms will be randomly selected from each administrative village for review. Our study was approved by the Medical Ethics Committee of the Hangzhou Center for Disease Control and Prevention (No. 2024-27) and was complied with the Declaration of Helsinki. All study subjects signed the informed consent form.

In the present study, participants aged ≥ 45 years were selected from the above sample. A total of 3,550 participants were included in the analysis after excluding participants with missing key variables.

Independent variables

Chronic diseases prevalence is defined as a patient’s self-reported chronic non-communicable disease that has been diagnosed by a medical professional before being surveyed. Fourteen chronic diseases were included in this study: hypertension, diabetes, dyslipidemia, heart disease (myocardial infarction, coronary heart disease, etc.), stroke, arthritis, chronic lung disease (chronic bronchitis and lung emphysema), malignant neoplasm, gout, osteoporosis, cataracts, liver disease, kidney disease, and gastrointestinal disease³⁶.

Dependent variable

There are two types of dependent variables in this study, one is the health utility values obtained by converting the EQ-5D-3 L scale according to the utility value point system, and the other is the five dimensions of the EQ-5D scale, which are mobility, self-care, daily activities, pain/discomfort, and anxiety/depression³⁷. Each question is answered with three options, 1 for no problem, 2 for a minor problem, and 3 for a serious problem. A total of 243 health states were included, and different health states corresponded to different health utility values.

The health utility values were converted using the 2018 Chinese population-based utility value point system³⁸. Utility value: × . The variables , , , , and represent the five questions on the scale. The subscripts =2, 3 represent the second option, the third option, and the selection of the first option means that the coefficient corresponding to that option is 0. The best health state is 11,111 with a health utility value of 1, and the worst health state is 33,333 with a health utility value of 0.1702.

Mediating variable

Sleep difficulty was the mediating variable in this study. Sleep difficulty was examined by asking respondents in a questionnaire: if you have had the following sleep problems at least 3 days a week for the past 30 days? The options are:

1. Difficulty falling asleep (taking 30 min or more to fall asleep),

2. Waking up twice or more in the middle of the night,

3. Early awakening with difficulty falling back to sleep.

Subjects were considered to have sleep difficulty if they select one or more of these options.

Control variable

The control variables in this study included basic demographic characteristics and lifestyle habits. The demographic variables included gender, residential area, age, and marital status. Residential areas were classified as urban and rural, age was categorized as 45–64 years old and 65 years old and above, and marital status was categorized as married/cohabiting and unmarried/divorced/widowed based on the availability of accompanied support. Lifestyle habit variables included smoking and alcohol consumption. Smoking status was obtained by asking the respondent: in the past 30 days, have you still been smoking? Drinking status was obtained by asking the respondent: in the past 12 months, have you been drinking alcohol?

Statistical analysis

Statistical analysis was performed using R version 4.3.2. Continuous information was expressed as ± and categorical information was expressed as N (%). For continuous variables, the t-test is used for the data that conforms to the normal distribution, and the nonparametric test is used for the data that not conforms to the normal distribution. Pearson’s chi-square test was used for the comparison of categorical variables. Mediated effects analysis of the impact of chronic diseases on health utility values was conducted using the mediate package. Set Bootstrap to repeat the sampling 5000 times with a 95% confidence level. It was considered statistically significant if the confidence interval did not include the value 0 for the indirect effect³⁹. The Bootstrap method is a nonparametric resampling procedure that does not require distributional assumptions and has high statistical power. Statistical significance was set at P < 0.05.

Results

As shown in Table 1, among the 3,550 samples, 1,766(49.75%) were male, and 1,784(50.25%) were female. In addition, 2,066(58.20%) respondents were from rural areas and 1,484(41.80%) were from urban areas. Married respondents accounted for 87.21% of all participants, and 1,277 (35.97%) of the respondents suffered from at least one chronic disease. 1560(43.94%) participants suffered from sleep difficulty. Regarding smoking and drinking status, 2,755(73.61%) respondents reported no smoking in the past 30 days, and 2,267(63.86%) respondents reported no drinking in the past 12 months. Health utility values were different in gender, residence, age, marital status, chronic diseases, sleep disorders, and drinking (P < 0.05).

Table 1.

Socio-demographic characteristics of participants.

Variable	N (%)		P value
Gender			< 0.001
Male	1766(49.75)	0.97 ± 0.08
Female	1784(50.25)	0.98 ± 0.07
Residence			< 0.001
Urban	2066(58.20)	0.98 ± 0.07
Rural	1484(41.80)	0.97 ± 0.08
Age			< 0.001
< 65	2320(65.35)	0.98 ± 0.06
≥ 65	1230(34.65)	0.96 ± 0.09
Marital status			< 0.001
Married	3096(87.21)	0.98 ± 0.07
Unmarried	454(12.79)	0.95 ± 0.11
chronic diseases			< 0.001
No	2273(64.03)	0.98 ± 0.07
Yes	1277(35.97)	0.97 ± 0.09
sleep difficulty			< 0.001
No	1990(56.06)	0.98 ± 0.06
Yes	1560(43.94)	0.96 ± 0.09
Smoking status			0.949
No	2755(77.61)	0.97 ± 0.08
Yes	795(22.39)	0.98 ± 0.07
Drinking status			< 0.001
No	2267(63.86)	0.97 ± 0.08
Yes	1283(36.14)	0.98 ± 0.07

All statistical tests between variables were performed using the wilcoxon rank sum test.

In addition to this, the demographic characteristics data by age (64 years old or less/65 years or more) were also explored in Supplementary Table (1) There were statistically significant differences in residence, sleep difficulty, and alcohol consumption between those with and without chronic diseases in both age-specific cohorts. The distribution and health utility values of chronic disease patients of different ages are shown in Supplementary Table (2) The chronic diseases in this study were predominantly hypertension and diabetes. The difference in health utility values between hypertension and diabetes was statistically significant in different age groups. The prevalence of sleep difficulty may vary between age groups, as well as the prevalence of sleep persistence and early awakening. Detailed data are presented in Supplementary Table 3.

Examining the mediating role of sleep difficulty on the association between chronic diseases and health utility values requires the fulfillment of the following four diseases.1) there is a significant association between chronic diseases and utility values; 2) there is a significant association between chronic diseases and sleep difficulty; 3) there is a significant association between sleep difficulty and health utility values; 4) there is a significant indirect effect of chronic diseases on utility values through sleep difficulty. Gender, residence, age, marital status, smoking status, and alcohol use were used as covariates in all analyses.

The settings of three regression models in the mediation analysis are shown in Table 2. The relative regression results showed that compared to non-chronic diseases, chronic diseases had a negative effect on utility value (β= -0.0084, P < 0.01) and could significantly predict sleep difficulty (β = 0.5737, P < 0.001). When the chronic disease was controlled for, sleep difficulty (β= -0.0168, P < 0.001) had an impact on utility value. These results show that sleep difficulty plays a mediating role in the influence of chronic diseases on utility value.

Table 2.

Regression coefficients for the mediator model among the study participants.

Predictors	Utility values	Sleep difficulty	Utility values
Chronic disease (Ref. non- chronic disease)	-0.0084**	0.5737***	-0.0062*
Sleep difficulty (Ref. non-sleep difficulty)			-0.0168***

*P < 0.05, **P < 0.01, ***P < 0.001.

Table 3 summarizes the total, direct, and indirect effects of chronic diseases on health utility values through sleep difficulty. The 95% CI for the total effect was [β= -0.0084 (-0.0142- -0.0031)], excluding 0. That indicates that chronic diseases were associated with poor health utility values. The bootstrap CI for the direct effect was [β= -0.0062(-0.0120- -0.0007)], excluding 0. That indicates that chronic diseases had a direct effect on health utility values after adjusting for other variables. The bootstrap CI for the indirect effect was [β= -0.0022 (-0.0034 - -0.0014)], excluding 0. That suggest that the effect of chronic diseases on health utility values was associated with the presence of sleep difficulty in patients, with a mediated effect of 26.19%.

Table 3.

Total, indirect, and direct effects of the mediator model.

Effect	Estimate	Boot (95% CI)		|aibi/c|
		Lower bounds	Upper bounds
Direct effect	-0.0062	-0.0120	-0.0007
Indirect effect	-0.0022	-0.0034	-0.0014	26.19
Total effect	-0.0084	-0.0142	-0.0031

Boot (95% CI): bootstrap 95% confidence interval; |a_ib_i/c|: proportion of relative mediation effect.

Each of the five dimensions of the EQ-5D scale was used as the dependent variable to explore the effect of chronic diseases through sleep difficulty in each dimension. The impact of chronic diseases through sleep difficulty on daily activities was valued at 0.0083, which accounted for 29.33% of the total effect. The impact of chronic diseases through sleep difficulty on pain (or discomfort) was valued at 0.0162, which accounted for 35.53% of the total effect. No effects of chronic diseases on mobility, self-care, and anxiety/depression dimensions through sleep difficulty were found. Detailed data are presented in Table 4.

Table 4.

Total, direct, and indirect effects of chronic diseases on the five dimensions of the EQ-5D scale, with sleep difficulty as a mediator.

Effect	Estimate	Boot (95% CI)
		Lower bounds	Upper bounds
Mobility
Direct effect	-0.0004	-0.0202	0.0197
Indirect effect	0.0039	0.0014	0.0075
Total effect	0.0035	-0.0159	0.0239
self-care
Direct effect	0.0105	-0.0033	0.0253
Indirect effect	0.0030	0.0013	0.0057
Total effect	0.0135	-0.0002	0.0286
Daily activities
Direct effect	0.0201	0.0023	0.0386
Indirect effect	0.0083	0.0042	0.0111
Total effect	0.0283	0.0092	0.0461
Pain/discomfort
Direct effect	0.0294	0.0044	0.0553
Indirect effect	0.0162	0.0107	0.0225
Total effect	0.0456	0.0202	0.0712
Anxiety/depression
Direct effect	0.0105	-0.0014	0.0231
Indirect effect	0.0025	0.0015	0.0034
Total effect	0.0130	0.0009	0.0256

Discussion

This study investigated whether sleep difficulty influences the association of chronic diseases with HRQoL using data from the health survey of middle-aged and older adults in Hangzhou. This study is the first to examine the possible intermediate variables of chronic diseases affecting HRQoL. This study identified that sleep difficulty partially mediates the relationship between chronic diseases and HRQoL, particularly in the dimensions of daily activities and pain/discomfort on the EQ-5D scale. The findings of this paper will provide more precise measures to improve the quality of life for middle-aged and older adults with chronic diseases.

We confirmed that the risk of sleep difficulty among middle-aged and older adults with chronic diseases is significantly higher than that of healthy individuals. In line with earlier research, patients with chronic diseases have more prone to experience sleep disorders^40–42. This is likely due to the increase in the number of diseases in the body, which disrupts the balance and leads to complications. These issues may affect the patient’s daily activities, which in turn can lead to sleep difficulty⁴³. For this reason, health professionals must aggressively create health education programs for middle-aged and older patients who have chronic illnesses. By lowering the likelihood of chronic disease consequences, sleep difficulty can be reduced. Meanwhile, we also confirmed that sleep difficulty has a negative effect on HRQoL, which is consistent with some previous studies^44,45. It may be because the occurrence of nocturnal sleep disorder leads to daytime drowsiness, which in turn leads to a decrease in HRQoL.

The mediating effect of sleep difficulty on the relationship between chronic diseases and HRQoL may be due to that sleep difficulty is one of the concomitant symptoms in middle-aged and older patients with chronic diseases. Additionally, sleep difficulty is associated with changes in patients’ health conditions. Chronic sleep difficulty may lead to decreased immune function, decreased insulin sensitivity, impaired glucose tolerance, and increased sympathetic nerve activity in middle-aged and older adults. These factors ultimately lead to an elevated risk of chronic diseases such as diabetes and hypertensive disease^46,47. Therefore, to improve the quality of life of people with chronic conditions, we can focus on measures to reduce sleep difficulty, including improving difficulty falling asleep and having a full night’s sleep. These measures include controlling the amount of time spent on electronic devices before bedtime, increasing physical activity, adopting the correct sleeping position, controlling diet, sleeping in bed when you feel like it, and reducing the amount of time spent resting during the day, among others. In terms of proper sleep positions, we generally recommend supine and right lateral positions. For example, patients with sleep apnea should be placed in the lateral position because the supine position increases upper airway collapse, which increases the frequency and duration of apneas⁴⁸.

Our study also noted that many chronic diseases result in physical dysfunction, which adversely affects the ability to perform daily activities and leads to a reduced quality of life. For example, patients with osteoporosis may have a limited range of motion due to susceptibility to fracture, which reduces daily activities^49,50. The link between chronic diseases and daily activities is mediated by sleep difficulty. The presence of sleep difficulty increases the probability of daily activity impairment in chronic patients. Sleep duration is shortened by chronic illnesses⁵¹. Whereas sleep contributes to neuronal recovery and remodeling. And, Chronic sleep difficulty impairs brain activity and raise the possibility of cognitive dysfunction, which subsequently impacts their daily activities.

We observed middle-aged and older adults with chronic diseases are more likely to experience pain and discomfort, which is consistent with the findings of Zheng et al.⁵². Chronic diseases, such as gastrointestinal and rheumatic diseases, are often associated with pain⁵³. Sleep difficulty acts as a mediating variable in the relationship between chronic diseases and pain, which is consistent with Écija et al.‘s study. This study suggested that sleep difficulty is frequently utilized as one of the mediating variables for pain⁵⁴. The presence of sleep difficulty can make patients with chronic diseases more susceptible to pain symptoms. Long-term illnesses like chronic diseases also result in sleep disorders that are permanent. By triggering the high expression of proteins that govern inflammation-related components, chronic sleep disorders might encourage the release of inflammatory signals, which in turn cause pain⁵⁵. Patients with multiple chronic diseases may experience more severe sleep disturbances and greater levels of pain. Meanwhile, chronic pain resulting from chronic diseases predisposes elderly patients to early awakening and reduced sleep duration.

Our study’s findings indicate that sleep difficulty is not related to how chronic illness affects mobility, self-care, anxiety, or depression. The differences in the total effects were not statistically significant in either the mobility or self-care dimensions. The results showed that the presence of chronic diseases did not affect the patient’s mobility and self-care, which is consistent with the results of previous studies⁵⁶. A possible reason is that the severity of chronic diseases in the study population selected for this study was mild and had not reached a level that would affect mobility and self-care. However, studies have also shown that adults with chronic illnesses tend to experience higher degrees of anxiety or depression⁵⁷. It may be due to the fact that the chronic diseases in this study were mainly hypertension and diabetes mellitus, which are widespread among middle-aged and elderly people, and the patients had a high level of knowledge about the diseases.

The basic demographic characteristics and lifestyle habits were selected as control variables for this study, which is consistent with most studies that have conducted HRQoL^57,58. Variables such as gender, age, smoking status, and alcohol consumption have an impact on HRQoL and EQ-5D dimensions. Therefore, all covariates in the mediation analysis process were kept consistent in this study.

Based on the above analysis, we should pay more attention to the sleep problems of elderly patients with chronic diseases and take a comprehensive approach to reduce the occurrence of sleep difficulty. For example, we can adopt cognitive behavioral therapy, which includes sleep hygiene education, stimulus control, sleep restriction, relaxation, and cognitive therapy⁵⁹. Besides, relevant study suggests that long-term CBT-I therapy is more cost-effective than drug therapy alone⁶⁰. Our findings provide precise measures to improve the HRQoL of middle-aged and older patients with chronic diseases and provide scientific evidence for developing public health policies. As a next step, we will carry out a comprehensive strategy to improve sleep difficulty in the study population to evaluate its effectiveness and provide experience for regions with demographics similar to Hangzhou.

According to the World Health Organization, 73.60% of deaths globally are caused by chronic diseases⁶¹. And these deaths are concentrated in developing countries⁶². Hangzhou is a representative city in East China, characterized by a large population base, high liquidity, and a large proportion of middle-aged and older population. The results of the study may be of some reference to other provinces in China with large middle-aged and older adults. In addition, China’s experience may also be beneficial to some less developed countries.

The current study has some limitations. Firstly, the researchers did not verify patients’ medical records for self-reported chronic diseases, which could lead to recall bias. Secondly, we used cross-sectional data, which did not allow us to predict a causal relationship between chronic diseases and HRQoL. Thirdly, the participants in our study come from seven districts of Hangzhou city in China, which may not accurately represent the true situation regarding the elderly population in China, and subsequent studies could replicate this study in other rural and urban areas for wider application. Finally, as mentioned previously, sleep difficulty was defined using three questions without evaluating the impact on patients’ daytime functioning, and we suggest that future research could fill this gap.

Conclusion

This study highlights that sleep difficulty in middle-aged and older populations mediates the impact of chronic diseases on HRQoL, especially in daily activities and pain (or discomfort) dimensions. Therefore, in the future management of chronic diseases, more emphasis should be placed on reducing sleep difficulty to enhance the quality of life of patients. At the same time, future research could explore targeted sleep interventions to assess their direct impact on HRQoL and EQ-5D dimensions.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

JX contributed to the conception and design of the study. YW conducted the data analyses and wrote the manuscript. ZC, ZC, KQ, ZY and CJ reviewed the manuscript. All authors read and approved the manuscript before submission.

Funding

The research was supported by the Hangzhou Biomedicine and Health Industry Development Supporting Science and Technology Special Project (Phase IX) (No.2023WJC071). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The data that support the findings of this study are available from Hangzhou Center for Disease Control and Prevention (Hangzhou Health Supervision Institution), but restrictions apply to the availability of these data, which were used under licence for the current study and so are not publicly available. The data are, however, available from the corresponding authors upon reasonable request and with the permission of Hangzhou Center for Disease Control and Prevention (Hangzhou Health Supervision Institution).

Declarations

Competing interests

The authors declare no competing interests.