Rational and Objective Analysis of China's Tiered Medical System Policy

Article reposted fromCenter for Public Policy, Chinese Academy of Social Sciences (WeChat Official Account)

There is virtually no disagreement across society regarding the role and significance of tiered diagnosis and treatment, namely that it optimizes the allocation of medical resources and yields the best cost-effectiveness. In light of the current predicament facing China’s healthcare system reform, there is a consensus that “tiered diagnosis and treatment” is an effective remedy for the healthcare system. It should be established as a formal “institution” under government leadership. In addition to relevant authorities issuing regulations to define the scope of practice and functional roles of medical institutions at all levels, emphasis is placed on the “tri-partite coordination” among healthcare services, health insurance, and pharmaceuticals. This requires health insurance departments to widen the disparity in reimbursement rates for insured patients seeking care at medical institutions of different tiers. Some scholars even propose that health insurance authorities should issue administrative regulations mandating patients to seek initial consultation at primary care facilities before being referred elsewhere; otherwise, expenses will not be reimbursed. The rationale cited is that this reflects advanced international experience, with the United Kingdom often referenced as an example of such a policy.

It is undeniable that in the United Kingdom, Germany, and Taiwan, primary care services provided by general practitioners (GPs) constitute the majority of healthcare services. Patients typically consult GPs first and access specialist care or hospitalization only through referrals (rather than being exclusively restricted to this pathway). This orderly service model is indeed advanced and efficient, representing a system we eagerly aspire to emulate. However, establishing such a well-structured tiered healthcare order is far from as simple as reverse-engineering imported products; one cannot simply adopt foreign blueprints to produce identical outcomes. A frequent error in learning from international experiences during China’s healthcare reforms in recent years has been understanding “what” works without grasping “why” it works. This is evident in initiatives such as the Essential Medicines System, the GP contract-based service model, the national drug price negotiation mechanism, and universal free healthcare schemes. The tale of “Dong Shi imitating Xi Shi’s frown”—a Chinese idiom denoting blind and ineffective imitation—has become a recurring narrative in China’s healthcare reform over many years. Currently, similar tendencies appear to be emerging in the promotion of tiered diagnosis and treatment.

What Exactly Is Tiered Diagnosis and Treatment?

“Initial consultation at the community level and referral to specialized inpatient care” is a procedural description of tiered diagnosis and treatment, not its nature or inherent attribute. Is tiered diagnosis and treatment a “system,” as some claim? Is it a “management framework”? Did China truly establish a “tiered diagnosis and treatment system” before the reform and opening-up period, as asserted by certain individuals? The author believes:

First, tiered diagnosis and treatment is not a standalone system, but rather the outcome of a well-functioning healthcare system, representing a healthy state of the medical service delivery system.

Neither the United Kingdom nor Germany has enacted specific legislation to regulate tiered diagnosis and treatment as a standalone system. Instead, tiered diagnosis and treatment is regarded as an objective pursued within the frameworks of healthcare service delivery management and health insurance systems. In other words, tiered diagnosis and treatment is an institutional outcome resulting from the operation of comprehensive medical and health insurance systems, rather than a system in its own right. The emergence of tiered diagnosis and treatment in the UK and Germany is underpinned by regulatory frameworks that feature open and competitive systems for physicians’ independent practice and outpatient care provision dominated by private clinics. Government regional healthcare planning does not mandate choices regarding ownership structures. Rather, the functional stratification of healthcare institutions and patients’ healthcare-seeking behaviors have naturally evolved over time under these regulatory regimes.

The capitation payment for clinic physicians in the UK and the point-based fee-for-service system for clinic physicians in Germany are not supporting policies designed to establish a “tiered diagnosis and treatment system,” nor are they regulatory mandates requiring patients to seek initial care at the primary level. Rather, these payment models were adopted because the healthcare systems naturally yielded the beneficial outcome of tiered diagnosis and treatment; health insurance funds then implemented appropriate payment methods based on this positive result to achieve optimal cost-effectiveness. We must clearly understand this causal relationship and underlying logic.

Second, tiered diagnosis and treatment is not about the geographical separation of medical institutions or mandatory functional stratification, but rather the continuity of healthcare services in terms of function.

Medical institutions at all levels and of all types are inevitably located in geographically distinct areas, which is a natural state. Some argue that China has failed to establish a tiered diagnosis and treatment system because tertiary medical institutions still operate general outpatient departments. It has been suggested that regulations should be introduced to require tertiary medical institutions to abolish their general outpatient services (it is rumored that Shanghai will adopt this approach), thereby creating a spatially structured tiered diagnosis and treatment system. In Germany and the United Kingdom, large hospitals provide inpatient, emergency, and specialized outpatient services, but indeed do not offer general practice services. Nevertheless, there are numerous general practice clinics within the vicinity of these hospitals, with some even renting premises from the hospitals themselves.

The spatial configuration of such medical institutions is not the result of artificial administrative intervention (such as China’s healthcare resource allocation based on governmental administrative hierarchies), but rather the outcome of collaboration among various types of medical institutions to achieve functional integration in healthcare delivery. General practitioners in clinics are located as close as possible to residents, enabling them to address the majority of primary care needs while establishing close collaborative relationships with specialists and hospitals (as seen in the numerous cooperation contracts among different healthcare providers in the UK and Germany) to manage complex medical cases requiring advanced technical expertise. This functional integration constitutes the essence of tiered diagnosis and treatment, which is entirely independent of the physical distance between clinics and hospitals or the hierarchical classification of hospitals.

Third, tiered diagnosis and treatment is not a simple spatial transfer of patient visits, but rather a substantive continuity in the healthcare services received by patients.

Many research reports argue that the failure to establish a tiered diagnosis and treatment system is due to Chinese patients’ poor healthcare-seeking habits, or even attribute it to issues of personal quality. This argument turns black into white and confuses right with wrong. Although there is a saying that “prolonged illness makes a skilled physician,” even if the vast majority of a country’s population held university degrees, they could not all attain the medical expertise of professional healthcare providers. Nor could they accurately assess whether their specific conditions match the functional capabilities of various medical institutions, even in the internet age. Therefore, patients’ needs extend beyond merely hoping that the initial consulting physician can accurately diagnose and effectively treat their condition. They also expect that, when the initial physician is unable to provide an accurate diagnosis or effective treatment, they will receive professional advice on where to seek further care, along with referral services (referral should be viewed as a service, not a means of control). Furthermore, physicians should proactively communicate patient conditions to specialists and inpatient teams when referring upward (and similarly when referring downward). If patients could receive such seamless, continuous care through collaboration among various types of physicians, while also saving time and other indirect medical costs, who would still be labeled as having “poor healthcare-seeking habits”? Under our current medical service system, forcibly directing patients from primary care institutions to tertiary medical facilities, without ensuring continuity in the substance of care, inevitably leads to increased inconvenience, redundant tests, higher expenditures, delayed treatment, and even life-threatening consequences.

Fourth, the tiered diagnosis and treatment system during the planned economy era was a result of comprehensive economic and social controls, rather than an optimal state of medical care sequencing.

The orderly state of tiered diagnosis and treatment exhibits strong procedural characteristics. Two approaches can establish such order, akin to queuing: one involves forming lines through military-style command-and-control; the other emerges naturally as a social contract aimed at maximizing efficiency through competition, as seen when people queue for buses or shopping. The former is simple and rapid, suitable for emergencies; the latter is gradual and incremental, appropriate for routine socioeconomic life.

For the first three decades, we adopted a highly centralized and regulated planned economy system, in which all socioeconomic activities were controlled according to administrative hierarchies. People’s movements were restricted to specific areas, and the allocation of medical resources followed the same pattern. Consequently, healthcare-seeking behavior was also subject to regulation (for instance, referrals or proof of authorization were required for seeking medical care outside one’s locality or for other reasons). This constituted the institutional foundation of tiered diagnosis and treatment at that time. The outcomes of such planned-economy controls are well known; could the control-based tiered diagnosis and treatment system have been an exception? Certainly not. The planned economy system led to inequitable allocation and access to pharmaceutical and medical resources under conditions of severe scarcity. The extent of one’s power determined the degree of freedom in seeking medical care, as well as the quantity and quality of healthcare services obtained.

The core of the reforms over the past three decades has been the deregulation of the economy and society. With the establishment of a market economy system and the gradual relaxation of social controls, the free mobility of people has removed the economic and social institutional foundations that gave rise to the tiered diagnosis and treatment model under the planned economy. Consequently, this control-based tiered diagnosis and treatment model is inevitably destined to disappear. Do we expect the emergence of an optimal tiered diagnosis and treatment order to be achieved by acting against the overarching trend of reform and opening-up and persisting with mandatory regulatory measures? Does the socioeconomic institutional basis for such an approach still exist? Therefore, under a market economy system, a tiered diagnosis and treatment model must be established through healthcare management systems adapted to market economy principles.

Answering this question requires adherence to a fundamental principle: patients’ healthcare-seeking behavior is not determined by whether they have medical insurance, but by the distribution of physicians capable of meeting their needs. Simply put, patients follow doctors, not reimbursement rates. This is a critical, life-and-death common sense that brooks no elaborate theoretical justification.Therefore, the prerequisite for establishing a tiered diagnosis and treatment system is the rational hierarchical allocation of physician resources. The institutional foundation for achieving this state lies in sound physician management regulations and medical institution licensing systems.

The period from 2010 to 2012 was marked by reforms in primary healthcare institutions aimed at strengthening grassroots services and establishing operational mechanisms. During this time, various regions introduced general outpatient coverage under medical insurance schemes and widened the reimbursement gap between primary care facilities and tertiary hospitals. But what were the outcomes? According to statistical communiqués issued by health authorities, the number of licensed (assistant) physicians in primary hospitals increased from 92,887 to 101,952, a net increase of 9,065; meanwhile, the number in tertiary medical institutions rose from 406,743 to 529,762, a net increase of 123,019. Over the same period, annual patient visits at primary hospitals grew from 145.74 million to 167.67 million, a net increase of 21.93 million; whereas patient visits at tertiary hospitals surged from 760.46 million to 1.08671 billion, a net increase of 326.24 million. Additionally, the average annual number of patient visits per physician in primary hospitals increased from 1,569 to 1,645, a net rise of 76 visits; while in tertiary hospitals, the figure rose from 1,870 to 2,051, a net increase of 181 visits.

What do these three sets of data indicate? First, the imbalance in physician allocation is becoming increasingly severe, and policies aimed at strengthening primary care have failed to yield positive results. Second, patients tend to follow physicians, meaning that reimbursement policies have limited effectiveness in guiding patient healthcare-seeking behavior. Third, the workload of physicians in tertiary hospitals has increased significantly compared to that in primary hospitals, further incentivizing tertiary hospitals to expand their physician workforce. This demonstrates that the healthcare system has fallen into a vicious cycle that is entirely contrary to the establishment of a tiered diagnosis and treatment model. Why? The root cause lies in the fact that healthcare system reforms in recent years have continued to adhere to and further strengthen administrative controls, including mandatory restrictions on medication formularies for primary care physicians, mandatory implementation of separate management of revenue and expenditure, mandatory centralized bidding for price reductions, mandatory zero-markup policies, and mandatory stipulations on the functional roles of hospitals at different levels.

The tiered diagnosis and treatment system is a product of sound medical management frameworks. Since practice has demonstrated that mandatory, directive-based medical management systems fail to achieve the goals of tiered healthcare, it is necessary to shift our approach and direct reforms toward deregulating the healthcare sector.

When we learn from advanced international experiences (such as healthcare insurance management systems), we do not emphasize “Chinese characteristics” (even though the management system precisely requires such emphasis). Why, then, must we invoke “Chinese characteristics” when returning to internationally recognized best practices in healthcare management—such as allowing physicians to practice freely, encouraging privately run healthcare institutions (particularly non-profit ones), and reducing the number of government-owned public medical institutions (including those that existed in China’s pre-1956 healthcare system)?

While the healthcare industry possesses unique characteristics, these specificities do not inherently justify government-led, mandatory command-and-control regulations. The uniqueness of the healthcare sector lies first in its restricted entry: not everyone is permitted to practice; one must acquire state-recognized qualifications through professional education, necessitating legislation to establish a national physician licensure examination system. Secondly, the inherent imbalance in bargaining power between patients and providers makes fair market transactions difficult, requiring the state to establish a health insurance system that creates a mechanism for equitable transactions with healthcare service providers. Given that universal health coverage has already addressed these two fundamental issues, it is no longer necessary for the healthcare regulatory framework to maintain strict controls over physician practice and the operation of medical institutions.

Without establishing a physician management system centered on free practice, and without opening up a competitive medical service system dominated by social non-profit medical institutions, tiered diagnosis and treatment cannot be realized under the government-monopolized public healthcare system and the “unit-affiliated” physician model. In today’s era of increasingly specialized medical divisions, physicians have diverse career aspirations: some prefer general practice, others specialize in specific fields, and some choose to work as hospitalists. General practitioners earn substantial patient volumes by delivering high-quality care for the vast majority of primary healthcare issues, thereby gaining higher social status and respectable income. Specialists, meanwhile, achieve professional prestige and decent earnings through their expertise in advanced, cutting-edge technologies and techniques. Only with a medical management system that allows free practice can physicians freely choose their career development paths, seize opportunities for upward mobility along their chosen trajectories, and thus lay the institutional foundation for tiered diagnosis and treatment. Furthermore, only with a diversified medical service system featuring multiple forms of ownership can competition among peers enhance service quality, while collaboration across different types of physicians improve service efficiency.

Some argue that to achieve tiered diagnosis and treatment, health insurance should guide patient flow through payment policies; some even suggest that health insurance should mandate initial consultations at community-level facilities, with reimbursement denied otherwise. Such views from health authorities are understandable, as they seek a scapegoat for the chaotic medical service order resulting from their failure to reform the healthcare system. What is baffling, however, is that individuals within the health insurance system also endorse and actively implement such measures. For instance, the so-called “general outpatient pooling” was established on the grounds that expanding outpatient coverage for minor illnesses would steer patients toward community facilities, thereby optimizing the allocation of medical resources. Yet the outcome has been an increase not only in visits to primary care institutions but also, and more significantly, in visits to large hospitals. The number of physicians at primary care facilities has risen slightly, while the number at large hospitals has grown substantially. Furthermore, measures that widen the disparity in reimbursement rates between primary care institutions and large hospitals violate the fundamental principles of basic medical insurance. These measures have resulted in wasteful spending, eroded public confidence in the function of basic medical insurance, and given rise to an anomalous offshoot—the critical illness insurance system.

There is no causal relationship between the medical insurance system and tiered diagnosis and treatment. As a payer, medical insurance maintains a market-based relationship with healthcare providers. The goal of medical insurance is to leverage the outcomes of tiered diagnosis and treatment to maximize cost-effectiveness, much like how store layouts are the result of market competition, allowing consumers to make choices that benefit them. In contrast, there is a true causal relationship between the healthcare management system and tiered diagnosis and treatment; tiered diagnosis and treatment is a favorable outcome of an effectively operating healthcare management system.However, as long as mandatory and directive medical management systems remain in place, tiered diagnosis and treatment cannot be realized, and health insurance, as the payer, will be unable to maximize cost-effectiveness through such a tiered system.Measures that appear highly rational and aim to promote tiered diagnosis and treatment through medical insurance are ultimately thankless tasks under the current healthcare management system.

If we wish to ensure the stable operation of medical insurance in the current period and establish a well-functioning tiered diagnosis and treatment system, our immediate priority should not be to introduce additional payment or coverage policies aimed at actively guiding patient care-seeking behavior. Instead, we must strengthen management to plug loopholes inherent in the current medical administration system, thereby minimizing losses to both the insurance fund and patients’ out-of-pocket expenses. Most urgently, we must vigorously steer healthcare management reform back onto the correct path. The longstanding situation in which medical insurance has been unfairly blamed for systemic shortcomings and incurred substantial unnecessary costs during healthcare reforms can no longer be tolerated or sustained.The core of the next step in healthcare reform lies in resolutely implementing the spirit of the Second Plenary Session of the 18th CPC Central Committee, streamlining administration and delegating power, reforming the mandatory medical regulatory system, and establishing a medical management system that allows physicians to practice and open clinics freely.