Population ageing is considered to be challenging for most countries in the world, particularly for China – the country with the world’s largest aging population [1, 2]. In China, the elderly population is estimated to rise to 400 million, accounting for 26.9% of the total population by 2050 [3]. With the accelerated population aging, the incidence and burden of non-communicable diseases (NCDs) have continued to increase in China. The prevalence rate of NCD increased from 157.4‰ to 342.9‰ from 2008 to 2018 [4]. According to the World Health Organization (WHO), about 68.6% of the disease burden is attributable to NCD, China’s rapid population aging is expected to increase the NCD burden by at least 40% by 2030 if the NCD is not controlled effectively [5].

Cardiovascular disease (CVD) as the main NCD constitutes the leading cause of death globally, representing 32% of all global deaths as well as being the cause of more than 40% of deaths in the Chinese population [6–8]. Over the past decades, the prevalence and mortality of CVD have increased dramatically in China. According to recent research, the prevalence cases of CVD increased from 40.57 million in 1990 to 93.81 million in 2016, and the age-standardized prevalence of CVD rose from 5,266 to 6,037 per 100,000 in the same time period. Meanwhile, the mortality rate increased from 174 to 309.33 per 100,000 in rural areas and from 213 to 265.11 per 100,000 in urban areas [9]. As the major NCD in China, CVD has posed unprecedented challenges to Sustainable Development Goals (SDG target 3.4) as well as NCD prevention and control in China.

Usual source of care (USC) is conceptualized as a regular place (doctor’s office, clinic, health center, or other places) or medical professional that a person visits most often for health care when needed [10]. A growing body of literature suggests that having a USC is associated with improved access to health care, decreased medical expenditures, appropriate preventive and treatment services for chronic conditions, and lower rates of unmet health needs [11–14]. Within the current health care delivery system, Chinese patients have several choices of USC: public hospitals and primary care facilities, which are the main health care providers [15]. Most public hospitals are owned by the government, with revenues derived from government subsidy (8.97% of their total revenues) and health care fees [16]. They are perceived to have higher capacity (including adequate numbers of health care professionals, high level medical equipment and technology) at a higher price and provide both specialist and primary care [17]. Primary care facilities, including township health centers, community health centers, public clinics, and private clinics, are regarded as the health care system gatekeepers and are responsible for delivering primary care and public health services. They are predominantly subsidized by the government, with the exception of private clinics, and provide care at a lower cost, but have limited health care capacity [17].

The role of community-based primary care in the prevention and control of CVD has been emphasized in many countries. For example, the American Heart Association Guide for Improving Cardiovascular Health at the Community Level recommends implementing prevention and treatment at the community level [18]. The 2016 European Guidelines on CVD Prevention in Clinical Practice highlight the role of primary care in CVD prevention and management and suggest that the general practitioner should be at the core of long-term health care provision [19]. To reduce health disparities between different populations, the Korean government has proposed to implement a community-based health care services program for chronic disease patients [20]. From 2009 to date, the Chinese government has enacted a series of health policies to build primary care capacity, such as family-doctor-contract services system, to provide basic clinical care, public health services, and health management services for the residency, especially for chronic disease patients.

Primary care facilities are supposed to be the optimal USC for NCD patients in China. However, a recent study found that Chinese patients with chronic disease preferred hospitals as their health care providers [21, 22]. Studies have also revealed that health care utilization was different among different medical institutions. For example, one study of stroke patients indicated that the annual number of stroke-related outpatient visits (primary: 0.42, secondary: 0.59, tertiary: 0.41) and hospital admissions (primary: 0.13, secondary: 0.33, tertiary: 0.34) per patient in primary care facilities were fewer than that in hospitals [23]. In addition, variations in costs tended to be driven by the level of medical institutions [22]. Patients with stroke in hospitals incurred higher costs per outpatient visit (primary: 339 yuan, secondary: 421 yuan, tertiary: 533 yuan) and hospital admission (primary: 6,262 yuan, secondary: 9,250 yuan, tertiary: 18,374 yuan) than that in primary care facilities [23].

Although the benefit of the USC has been widely recognized internationally, few studies have been conducted in China, especially on CVD. Based on the data collected by the WHO from eight provinces in China, our study aimed to compare health care utilization in three types of USCs among the elderly with CVD. Our study complements existing research on the USC and fills the gap in knowledge regarding the difference in health care utilization in different USCs among the elderly living with CVD. This study has important implications for broader chronic disease management in China.

This study compared the health care utilization in three types of USCs among Chinese CVD patients aged 50 and over. We found that nearly 70% of older participants with CVD chose public hospitals as their USC rather than primary care facilities. When it comes to health care utilization, the elderly CVD patients using primary care facilities as their USC generally had fewer hospital admissions, but a higher probability of unmet health needs compared with those using public hospitals as their USC. At the primary care level, the elderly CVD patients identifying public clinics as their USC had more outpatient visits compared with those using private clinics as their USC. Using primary care facilities as the USC, particularly public primary care facilities, appeared to increase access to health care to a certain extent.

Our data indicated that nearly 70% of older CVD patients choose public hospitals as their USC rather than primary care facilities, which is consistent with a prior study conducted in China that 69.63% of heart failure patients choose hospitals to receive treatment in preference to primary care facilities [35]. Even China has made remarkable progress in strengthening primary health care, however, the quality of primary health care is still deemed poor, resulting in patients bypassing primary care facilities and seeking health care from hospitals for chronic conditions [36–38]. This preference for hospitals was contrary to international recommendations to prevent and control CVD at the level of primary care facilities [18–20]. The Chinese government still needs to struggle to ensure the quality of primary health care in NCD prevention and control.

Health care was largely underused by CVD patients, particularly outpatient services. In our study, more than one-third of patients had not been treated as outpatients in the past year, and less than a quarter of patients had been hospitalized over the past year. Meanwhile, the mean number of outpatient visits per year was 5.33 times, which was lower than that suggested in the Chinese guidelines for hypertension management, of once every two or four weeks, particularly for patients with poorly controlled blood pressure [39]. Other clinical guidelines also require regular monitoring of blood lipids and medications for patients with CVD in prevention [40, 41]. This underuse might lead to poorly adhering to treatment guidelines, then increase hospitalization and health care cost [42].

In the analysis of public hospitals and primary care facilities as USC, we found that patients using primary care facilities as their USC had fewer hospital admissions than those who adopted public hospitals as their USC, and the difference in outpatient visits was not significant. These results suggested that primary health care fulfilled certain functions in increasing access to health care in older patients with CVD, which is likely to reflect the achievement of China’s primary health care system reform with core responsibilities in preventing and managing chronic disease [38, 43]. This finding can support the utilization of primary care facilities as the USC to prevent and control CVD. However, among elderly people with CVD, those who used primary care facilities as their USC had a comparatively higher probability of unmet health needs than those using public hospitals as their USC. This difference in unmet health needs can be explained by the quality of health care provided in these medical institutions [44, 45]. Public hospitals, in terms of service capacity (including the availability of health professionals and equipment) and diagnoses of severe or complex diseases, have relatively high quality [17, 31]. In comparison, although some progress has been made with increased financial investments in primary care, primary care facilities frequently struggle with structural characteristics, incentives and policies, and other serious challenges, widespread gaps in the quality of primary health care still exist [36, 38, 46]. One study revealed that for patients with dysentery or unstable angina symptoms, village doctors completed only 18% of the suggested questions and correctly diagnosed only 26% of patients’ conditions [47]. Multimorbidity may also be associated with this difference in health care utilization between public hospitals and primary care facilities. Several studies have shown that multimorbidity has a significant influence on the likelihood of accessing health care, particularly for individuals living with CVD [48–50]. In our study, no matter whether patients choose public hospitals or primary care facilities as their USC, more than half of them suffered from multimorbidity. Due to the difference in quality of care, public hospitals could better meet their health care needs than primary care facilities.

At the primary health care level, we found that those who chose public clinics as their USC had more outpatient visits than those who used private clinics as their USC. This gap seems to be largely mediated by the quality of primary health care. A national study conducted in China, aiming to rate the care quality of public and private clinics based on patient-perceived quality, revealed that public clinics were rated more favorably in prompt attention, communication, autonomy, dignity, and confidentiality [31]. In the current health system reform, in addition to increasing substantial financial investment and infrastructure building in primary health care, the Chinese government also encourages partnerships between hospitals and primary care facilities, especially public primary care facilities, such as establishing integrated health care delivery system based on primary health care, to strengthen local primary health care capacity and hence prevent and manage chronic diseases [36, 38]. Previous research also demonstrated that private primary care facilities tended to be poorly regulated and suffered from low quality of prescribing, which was related to limited government subsidies, limited availability of essential medicines, and their for-profit orientation [31, 51, 52]. Moreover, higher public trust and stronger preference for public primary health care facilities might be considered a factor influencing health care utilization [31, 51]. A previous study in three Chinese provinces showed that only 8% of individuals stated that physicians in private clinics had better skills than that in public clinics, and just 29% of individuals reported they would prefer to see a private doctor than a public doctor [51]. Thus, compared with private clinics, public clinics have a higher quality of continuity and coordination in CVD management with a higher degree of trust, leading to greater utilization of outpatient services by patients who identified public clinics as USC. Notably, in China, progress has been made toward the prevention and control of NCD in public primary health care, but substantial opportunities for primary health care improvement remain.

Our findings have important policy implications for CVD prevention and control. Firstly, although primary care facilities, as USC, play a certain role in increasing access to health care in older patients with CVD, suboptimal quality persists. It is necessary to call upon the government to pay more attention to the improvement of primary care quality. More health care resources, especially highly qualified health care workers, should be allocated to primary care facilities to narrow the capacity gap between primary care facilities and hospitals [37, 38, 53]. Secondly, primary care facilities are at a disadvantage in the tiered health care delivery system, which rendered them difficult to function as gatekeepers. It is essential to strengthen the coordination between primary care facilities and hospitals to establish an integrated delivery system, such as improving the bidirectional referral mechanism, changing the provider payment mechanism to pay according to the size of the population served and the quality of care delivered, integrating and sharing electronic patients’ records [38]. Thirdly, to shift citizens’ health care demands toward primary health care, the Chinese government should consider organizing large-scale health education activities to improve nationwide understanding of primary health care function and its importance in the prevention and control of NCD, thereby guiding citizens to develop appropriate health care seeking behaviors [36].

This study has several limitations. First, the SAGE-China data did not cover all types of CVDs. Further studies analyzing the relationship between USC and health care utilization among patients with other CVD are needed. Second, our study used cross-sectional data in 2010 because the second round (2014–2015) of SAGE-China data is still in the process of collation and not available. We could not infer causality between USC and health care utilization. However, Although the data is relatively old, residents’ health care utilization pattern has not changed significantly in the past years [15, 36, 46]. The outpatient visits and hospital admissions at primary care facilities (outpatient visits: 61.72% in 2009, 50.2% in 2021; hospital admissions: 31.01% in 2009, 14.5% in 2021) kept decreasing relative to those at hospitals (outpatient visits: 34.97% in 2009, 45.8% in 2021; hospital admissions: 64.3% in 2009, 81.5% in 2021) from 2009 to 2021. Moreover, the multistage cluster sampling design and using IPTW with survey weighting could help us generalize the estimated effect to the elderly with CVD in China. Considering these reasons, we maintain that our analysis still could provide new insight into CVD prevention and control for the whole elderly population in China. Third, China has established a three-tier health care delivery system containing tertiary hospitals, secondary hospitals, and primary care facilities. In our study, USC was divided into two categories: public hospitals and primary care facilities. We could not clearly distinguish whether the public hospital is a secondary hospital or a tertiary hospital. Further research could compare the health care utilization among different health facility level.