Background
High-cost (HC) populations have been widely studied worldwide. Previous studies in the United States (US) and Canada have found that the top 5% of the HC population accounts for approximately 50% of total health care expenditures [
1‐
3]. The HC population was also defined as the top 10% or top 1% of patients by medical expenditure, who are generally most affected by chronic diseases in the last years of their lives [
4]. The HC population is also often associated with high need, which means that these populations often suffer from three or more chronic diseases [
1,
5]. These HC populations often require extensive attention and consume a disproportionate share of resources [
6]. If their needs for health care services are unmet, they will experience a greater risk of poor health outcomes and increased health care utilization and health care expenditures [
7]. Previous studies showed that the HC population status for over half of certain patients would persist for many years [
8]. In China, previous studies have found that the top 5% of HC families in Hubei Province spent approximately 44.9% of the total medical expenditures on frequent emergency department (ED) visits and hospital stays [
9], and approximately 68.0% of total medical expenditures of families in Jiangsu Province were attributed to the top 5% of HC patients [
10].
Many studies have proven that health care costs incurred by the HC population are partially preventable. Of the 4.8% of potentially preventable Medicare spending, 73.8% was used by the HC population [
11]. In 2010, 32.9% of ED-related costs were incurred by the HC population, and 41% of these costs were preventable [
12]. A population-based study of HC patients with cancer in Ontario, Canada, revealed that 9% of their medical spending stemmed from potentially preventable or avoidable acute care [
2]. The overuse of costly therapies and services that are disconnected from patients’ multiple needs places a heavy burden both on the patients and the health care system [
13]. Previous studies have also shown that HC patients with chronic conditions rely heavily on health care services delivered by primary care providers [
14,
15]. Moreover, simple interventions among patients with heart failure in outpatient settings, such as home-based physical therapy programs, may reduce unnecessary spending [
11], and intensive outpatient health care programs were found to be useful in reducing ED visits, hospitalization and costs [
16].
A high-quality primary care delivery system usually requires the enhanced coordination of care and a multidisciplinary team [
17]. One of the representative practices is to establish specialized clinics that provide intensive medical, social and psychosocial services with a wide array of providers (nurses, medical assistants and social workers). This model has been credited with a 20% reduction in monthly health care spending and a 55% reduction in ED visits [
17]. Another study found that emphasizing interdisciplinary patient engagement led to a 7% reduction in the use of hospitalization and a 31% reduction in the rate of 30-day readmission [
18]. For patients with cardiovascular disease, teams with good working relationships among members were also associated with fewer hospital stays and lower costs for patients [
19]. Moreover, team-based primary care services could also improve the continuity of care, and a greater continuity of care is related to lower health care expenditures [
20], thus leading to a reduced prevalence of the HC population [
21]. Additionally, compared to beneficiaries with fewer regular primary care visits, subgroups with more frequent primary care visits had less utilization of ED visits and hospitalization services and fewer overall Medicare costs [
22].
Complex health care needs require more comprehensive and high-quality services by primary care providers [
23]. However, current primary care providers in China are ill-equipped and not sufficiently trained to manage these high-need or HC populations. Many studies have shown mixed results. Compared to the non-HC group, the HC population had a lower percentage of preventable ED visits and hospitalization costs, and regions with a greater density of primary care physicians reported greater levels of preventable spending for the HC population [
12]. Another study also suggested that the non-HC population had over three times the rate of preventable utilization than HC patients [
24]. No consistent association was found between changes in the capacity of patient-centred medical homes and the utilization of health care services among HC patients [
24]. Additionally, clinics that provide complex care management did not reduce overall expenditures after discounting upfront costs [
6]. The reasons for these different results are complex. First, these interventions need substantial long-term investment, including providing the necessary resources and training courses for these providers. For example, team-based care might only provide longitudinal continuity of care rather than interpersonal continuity of care [
21]. Second, the quality of these services varies widely and often cannot reach the places that need them most [
5]. Third, although most studies overemphasized the utilization of acute care services, such as ED visits and hospitalization services for the HC population, nearly 70% of this spending was used in long-term care services [
8], and the largest driver of inpatient spending was catastrophic events. Therefore, the factors associated with the prevalence of the HC population remain to be examined.
In China, although rural residents account for nearly half of the population, medical resources are still concentrated in urban areas [
25]. To improve access to health care services and equity in the utilization of health care services, the 2009 New Health Care Reform has provided massive financial resources to provide public health care services [
26], and the central government has committed to supporting innovative pilots of home or community-based care models, such as the Sanming Model and Xiamen Model [
27,
28]. Although the Chinese government has made great progress in improving equal access to basic health care and financial risk protection, gaps remain in the efforts to control unreasonable increases in health care expenditures [
29]. More importantly, the service scope of primary care facilities (PCFs) is narrowing, so patients tend to seek health care services at higher-level facilities [
26]. Many PCFs only provide services regarding chronic disease management and close their surgical and obstetric services [
30,
31]. A previous study indicated that a wider PCF service scope could reduce the utilization of health care services outside of PCFs and thus reduce overall per capita spending [
26]. Therefore, it is vital to explore ways to strengthen the primary care delivery system to reduce the prevalence of the HC population. However, no studies have focused on the association between PCF service scope and the prevalence of the HC population, and the extent of the association between the service scope of PCFs and the prevalence of HC patients remains unclear. Therefore, this study aimed to investigate the association between PCF service scope and the prevalence of HC populations among corresponding communities.
Discussion
To the best of our knowledge, this study is the first to examine the association between PCF service scope and the prevalence of the HC population in China. Our results not only provide evidence for potential ways to expand the service scope of primary care providers but also offer strategies for reducing the prevalence of the HC population. We found that a greater PCF service scope was associated with a lower prevalence of the HC population, which suggests that a greater PCF service scope can help avoid certain preventable costs, thus reducing the occurrence of the HC population. Differences in the proportion of HC patients by age group and sex once they had been enrolled in a poverty reduction scheme and regardless of their referral status were also observed.
First, patients living in communities with greater PCF service scopes were less likely to be in the HC population. This result may be related to the fact that an improved comprehensiveness of care can promote the use of health management services by those potential HC populations, especially those suffering from several chronic diseases [
1,
5]. On the one hand, a wider PCF service scope suggests that more services are provided by primary care physicians; thus, patients’ needs would be better satisfied by primary care providers than by medical staff in high-level hospitals. On the other hand, better health management and follow-up services also lead to more utilization of preventive services that improve patients’ health status with less consumption of health care resources [
39,
40]. As a result, a greater service scope can reduce the occurrence of avoidable complications, thus reducing some preventable costs. Comprehensive care could also promote interpersonal continuity of care among primary care physicians, which has been shown to be associated with a modest survival improvement without increasing the intensity of end-of-life care [
41]. Given that approximately 30% of HC patients are in their last year of life [
4], a greater PCF service scope might mean better care for patients during their final phase of life.
Second, we found that the marginal differences detected in facility-level service scope between the top 5% and top 10% HC populations, between the service scopes of quantile 3 vs. quantile 1 and between those of quantile 5 and quantile 1 were slightly larger than the results of comparisons between the other subgroups and quantile 1. One previous study pointed out that a greater PCF service scope could lead to a smaller rate of 30-day readmission [
26], which would promote cost savings through a reduction in preventable readmissions and the unnecessary utilization of related services. We also found that older adults are more likely to be in the HC population, which is consistent with a US study that found that the HC frail and elderly group accounts for over 40% of total potentially preventable spending [
4]. Moreover, a broader service scope of PCFs, combined with home-based health care services, could generate cost savings for both patients and health care providers [
42]. Evidence suggests that meeting their housing, nutritional and personal care needs can also reduce patients’ health care costs [
43]. However, the greatest PCF service scope was not associated with a reduction in the likelihood of the top 1% HC population, which may be related to the fact that the top 1% HC population might go to high-level hospitals directly since PCFs cannot deal with these patients’ health care needs. This result raises concerns about the effectiveness of tailored interventions that are aimed both at the population and community levels in meeting the health care needs of these HC patients, especially those in the top 1% of the HC population [
44,
45].
Third, although expanding the service scope of PCFs is a promising way to meet the health care needs of the HC population and to reduce the incidence of the HC population, the presence of most intensive primary health care services did not lead to a reduced use of hospitalization services or to a reduction in their associated costs during the half-year follow-up [
46]. The validation of these results through the use of a larger-scale dataset and the inclusion of a greater number of covariates, such as the travel distance between the centre of different communities and the PCFs that service them, is urgent. Additionally, policy efforts may also increase health disparities if these interventions disproportionately benefit advantaged groups while leaving vulnerable populations behind. Therefore, a more detailed assessment of the utilization of practitioner-level and facility-level services among different subgroups that could inform the capacity-building of the primary care system in China is also warranted.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.