Difference in bypass for inpatient care and its determinants between rural and urban residents in China

In recent years, excessive demand on inpatient care has been continuously concerned by policy makers and health insurers [1]. This excessive demand on inpatient care is due to patients receiving unnecessary inpatient care or services that are beyond their capacity to pay [2]. It could occur on three occasions: inappropriate admission, where patients receive unnecessary inpatient care rather than outpatient one or services that are beyond their capacity to pay; inappropriate inpatient services, in which patients receive inappropriate services during necessary hospitalization; and bypass for inpatient care, in which patients unnecessarily receive inpatient care from a higher-level hospital located farther away than the health care facility closest to their residence [3]. This research mainly focuses on the last occasion. The concept of bypass for inpatient care emphasizes that the inpatient care received by patients is unnecessary. The bypass behavior for diseases or illnesses that could be treated at local healthcare facilities brings many disadvantages, as it undermines functions of local healthcare facilities, crowds out resources for patients in need of higher-level hospital care, increases economic burden of patients, and decreases the effectiveness of a healthcare system in the long run [4].

In China, healthcare facilities that provide clinical care could be roughly classified into two groups: primary care facilities and hospitals. Primary care facilities predominantly consist of village clinics and township health centers in rural areas and community health stations and community health centers in urban areas. Except for hospitals, primary care facilities, including township health centers and community health centers, also provides inpatient care to residents in each town or community. Standardized primary care inpatient services constitute services from department of internal medicine, pediatrics, surgery and obstetrics and gynecology, and the inpatient services are mainly provided by doctors, nurses, pharmacists, clinical laboratory technicians and imaging technicians who work at primary care facilities. Primary care inpatient services are funded by a combination of health insurance and out-of-pocket payments. Without a gate-keeping system, residents are free to choose their first-contact healthcare facility; they generally have little trust in primary care and usually bypass primary care facilities to seek healthcare in hospitals. The growth rate of outpatient visits and admissions at hospitals were much higher than that at township/community health centers [5], leading to “kan bing nan, kan bing gui” (“medical treatment is difficult to access and expensive”) for decades [6].

In response to the unbalanced distribution of medical resources and inappropriate patient flow, China issued policy documents on the Hierarchical Medical System in 2015 and family doctor signing service in 2016 [7, 8]. The Hierarchical Medical System indicates that different levels of healthcare facilities have a clear division of responsibilities for undertaking different health services. Family doctor signing service was introduced to promote the establishment of the Hierarchical Medical System through building a stable service relationship between primary care physicians (or teams) and contracted residents. Unlike the mandatory gatekeeping system by legislation in many countries [9], patients were voluntarily encouraged to sign a contract with primary care physicians (or team), visit primary care first, and utilize hospital care via referral from primary care physicians through economic incentives in China. This study focuses particularly on bypassing primary care for inpatient care at hospitals.

Studies from several countries have focused on bypass behavior, in which a patient travel past the local primary care facility to seek services further away [10‐12]. The rates of bypassing primary care ranged from 13.7% in Japan [6] to 67% in India [8], implying that bypass behavior is context dependent; thus, country-specific analysis is indispensable [4]. Various factors are associated with bypass behavior, and they could be categorized into two groups: patient factors, such as income, satisfaction, and disease severity [13, 14]; and supply-side factors, including geographic accessibility, quality of care, and cost of treatment [10, 15]. Evidence regarding bypass for inpatient care from China is limited. Few studies related to bypass have only investigated the bypass rate for outpatient care [16], the distribution of health spending between primary and higher-level care [17, 18], and the factors associated with the health-seeking behavior of patients through qualitative analysis [19, 20]. To date, the rate of bypass for inpatient care and its determinants are largely unknown, and little is known with regard to the difference in bypass for inpatient care between rural and urban areas in China, especially after the implementation of the Hierarchical Medical System and family doctor signing service. Given the differences in accessibility of hospitals and economic development, differences possibly exist in the bypass for inpatient care between urban and rural areas. Therefore, this study aimed to estimate the rate of bypass for inpatient care and explore the factors predictive of bypass in rural and urban areas, respectively. Rural-urban comparison may be conducive to understanding the differences in health-seeking behaviors and improving equity in health care between rural and urban residents. Bypass for inpatient care is a type of health services utilization behavior. Anderson’s Behavioral Model, which provides a theoretical framework for explaining health services utilization behavior and was validated on rural and urban population in China in many previous studies [21‐25], was used to identifying potential factors of bypass behavior.

This study showed that the rate of bypass for inpatient care (73.8%) was sizable, which is similar with that for outpatient care (85.83%), as indicated by a study among four counties in Central China [16]. In the present study, the rate of bypass for inpatient care in rural areas was 56.2%, lower than that in urban areas. However, this rate was still out of the range of 30–44% shown by previous studies on the proportion of patients who bypassed their local hospital for inpatient care services from the US [15, 32]. Such high rates of bypass for inpatient care may be due to the weakness of primary care, a lack of gate-keeping system, the increase in patients’ risk aversion, and the insufficient economic incentives from the health insurance in China. Though one important aim of the 2009 healthcare reform in China was strengthening the primary care system, a decreasing proportion of primary care providers to overall healthcare providers illustrated a relative strengthening of the hospital sector and a relative weakening of the primary care sector from 2009 to 2017 [33]. Residents generally have little trust in primary care due to both objective factors (e.g. the relatively lower staff capacity and less medical resources) and subjective factors (e.g. the fear of misdiagnosis or wrong treatment) [34, 35], and they could access any healthcare facilities without referral under China’s free-access system. To avoid misdiagnosis and delays in disease treatment, patients usually bypass primary care to seek healthcare in hospitals [34]. In addition, the compensation gap from health insurance for the same disease between primary care and hospitals is not large; thus, residents do not have sufficient financial incentives to visit primary care facilities [36]. The rate of visits to primary care facilities decreased from 62% (2009) to 54% (2017) [37], which is much lower than the value (no less than 80%) recommended by World Health Organization. Therefore, more actionable measures in strengthening primary care and increasing financial incentives to lead patients to visit primary care are needed. Gradually establishing a referral system on the basis of primary care capacity building and interacted information system between healthcare facilities is also recommended.

The rate of bypass for inpatient care among urban residents was higher than that among rural residents (91.3% vs. 56.2%). This finding was mainly due to the higher hospital accessibility in cities and the inability of some community health centers to carry out inpatient services. Urban residents are closer to higher-level health care facilities than rural residents, and the time and cost to reach higher-level health care facilities are often lower. Garnick believes that patients evaluate the time and cost of arrival when choosing a healthcare facility [38]. Thus, urban residents were more inclined to visit higher-level hospitals than rural residents. However, compared with cities, only one township health center and one village clinic, which are both primary care facilities, usually exist in each town and village, respectively. Therefore, the time and economic cost of reaching a hospital was much higher for rural residents. Although healthcare facilities with better medical quality are attractive, higher fees and longer distance limit the choice of rural residents [38, 39]. In cities, different from township health centers, which are required and able to provide inpatient care services in rural areas, only community health centers with strong health service capacities are encouraged to provide inpatient care services [26]. Therefore, the inability to provide inpatient care among community health centers was another reason for the much higher rate of bypass in urban areas.

In both areas, age and the type of nearest healthcare facility were associated with residents’ bypass behavior for inpatient care. Older residents were less likely to experience bypass, inconsistent with the result of a previous study [40]. This finding may be due to the inconvenience for the elderly to travel and be hospitalized far from home and their higher satisfaction with primary care facilities shown by previous surveys [41]. Patients who lived closest to clinics/community health stations (which are primary care facilities with smaller size and cannot provide inpatient care services) or hospitals had a higher risk of bypass for inpatient care than those who lived closest to township/community health centers. For majority of patients who lived closest to clinics, especially rural patients, township health centers may be their closest healthcare facility that is able to provide inpatient care. However, they preferred travelling out of the town for inpatient care at hospitals. This finding indicated that township/community health centers were only able to retain small areas of surrounding residents. According to Anderson’s Behavioral Model, the distribution of health care facilities and how they are structured to offer services have impacts on health service utilization of residents [42]. Therefore, it may need to be improved how different levels of health care facilities were structured to deliver services in both rural and urban areas. Surprisingly, having signed a family doctor had no significant effect on bypass for inpatient care for both patients according to logistic regression analysis. Although the family doctor contracting policy has been in place for several years, its overall effect was not satisfactory, and the restrictions on residents’ health-seeking behavior have not begun to appear because of the deep-rooted self-selection of doctors, as some research reported [43‐45]. In addition, bypass for inpatient care was identified in this study if the patient was hospitalized in a hospital for a certain disease covered by a catalog (which included a list of diseases that could be diagnosed and treated at primary care facilities and was established by the government). Therefore, patients who had signed a family doctor may experience bypass for inpatient care via referral on their own request or on family doctors’ advice as a result of relatively lower staff capacity and lower quality of inpatient care at primary care facilities [34, 35]. Among rural patients, those with circulatory, digestive, or skin and bone system diseases were more likely to show bypass behavior than those with respiratory diseases, possibly indicating that the inpatient care services for circulatory, digestive or skin and bone system diseases at primary care facilities could not meet the health care needs of patients and requires further improvement. However, the result must be interpreted with caution because other factors (e.g. severity of illness or patient satisfaction) may also lead to the higher possibility of bypass behavior among patients with circulatory, digestive or skin and bone system diseases. This needs to be further studied.

In this study, the rate of bypass for inpatient care was estimated, and the determinants of bypass for rural and urban residents were explored. The results showed that bypass for inpatient care was sizable, and urban residents had a higher bypass rate for inpatient care than rural residents. Age and the type of nearest healthcare facilities were associated with bypass for inpatient care. However, whether having signed a family doctor or not had no significant effect on the bypass behavior for both types of patients. Disease was related to bypass only in rural areas. More actionable measures in strengthening and leading patients to primary care are needed. Gradual establishment of a referral system is also recommended. Inpatient care services for circulatory, digestive, or skin and bone system diseases may be prioritized for improvement at primary care facilities to reduce bypass behavior in rural China.

Though some community health centers in urban areas were not able nor required by the government to provide inpatient care services (which may result in the high rate of bypass) before 2018, the analysis for urban residents was still included in this study. The reason is because community health centers were encouraged to provide inpatient care by the government by publishing the standards for capacity building in delivering services for primary care facilities through a campaign called Delivering Qualified Services at Primary Care Facilities, in 2018. Therefore, understanding the bypass rate for inpatient care and the contributing factors that may guide the implementation of the campaign is still worthwhile. In addition, a few potential contributing factors (e.g., patient satisfaction, severity and duration of illness, quality of care) were not included due to lack of data. Some contextual characteristics included in Anderson’s Behavioral Model (e.g. community values and cultural norms regarding where to obtain health care, financial incentives from health insurance to reduce bypass behavior) may also have impact on bypassing, which were not addressed in this study. Further studies on the other contributing factors of bypass for inpatient care in China are needed.