Gordon G. Liu: community clinics, rational drug use, and surgical robots could help China achieve accessible healthcare

China's leading health economist says the status quo leaves doctors undervalued, patients underserved, and public budgets overstretched.

Gordon G. Liu is BOYA Distinguished Professor of Economics, Dean of the Institute for Global Health and Development, and Director of the China Centre for Health Economic Research at Peking University.

He delivered a TED-style talk on China’s healthcare challenges and proposed reforms at TAIXUE, a programme of the New Economist think tank. The transcript was published on the New Economist’s official WeChat blog on 29 August, and the video recording remains available on TAIXUE’s YouTube channel.

刘国恩：人人如何能看得起病？

Gordon G. Liu: How to Make Healthcare Accessible to Everyone

It is a great honor to speak here today at TAIXUE on how to make healthcare accessible to everyone. I believe this topic will be of great interest to all of you, as well as of close concern.

From the perspective of economics, the question is how the provision of medical services can better meet each person’s demand for medical services, so the state of supply and demand determines whether everyone can access medical care.

Because the resources required for accessing healthcare—hospitals, beds, doctors, equipment, and medicines—are limited, the supply–demand contradiction can be approached from four aspects to see whether it can be better resolved; in other words, there is considerable room to improve the medical-service model so that more patients can receive better care.

One of the most distinctive features of medical services in China is that when they fall ill, people tend to go straight to large hospitals—a hospital-centered model that delivers comprehensive services. This is a healthcare system with strong Chinese characteristics, and it is not the case in many other countries. In many developed countries, hospitals are not like those in China: although they provide comprehensive services with inpatient and emergency departments, they rarely have large general outpatient departments. That is because outpatient services are largely diverted to the community, so when inpatient and emergency care are needed, the difficulty of securing an appointment or a bed does not arise—why is that?

In any Chinese hospital, especially large ones, about 70–80% of patients are outpatients, while inpatients, emergency cases, and the critically ill account for 20–30%. If roughly 80% of cases could be redirected to community care, the benefits would be substantial.

For most illnesses, outpatient services in China are sufficient, making it possible to receive care conveniently in the community. This saves both time and money. At the same time, large hospitals can then be reserved for emergencies, inpatient treatment, critical cases, and complex conditions.

Therefore, improvements to China’s medical service model should aim to enable more people—and ultimately everyone—to access care. A key step came in 2009 with the national healthcare reform plan, which set several major goals; a primary goal was to implement tiered diagnosis and treatment, guiding patients to seek care at medical institutions that match the level and severity of their conditions. The goal was to relieve pressure on large, crowded hospitals by channeling many ordinary patients who do not require tertiary care into community services.

At the same time, if a significant share of routine outpatient cases is kept within the community, a question arises: can the community provide a warm, convenient, and good-value platform to deliver such care? This, of course, is an issue that must be addressed.

Imagine that across wide residential areas—right on people’s doorsteps—numerous easily accessible clinics are run by general practitioners, rather than today’s situation of relatively few community services or predominantly private institutions. If many highly trusted physicians practice within residential communities, the benefits are obvious.

Since 2009, many years have passed under healthcare reform. To assess whether the tiered diagnosis-and-treatment goal of diverting ordinary outpatients from large hospitals has achieved the expected progress, one should examine the national macro data released each year—specifically, the proportion of ordinary patients treated by tertiary hospitals. These figures, as observed, are not particularly encouraging.

In other words, over more than a decade, large hospitals, because of their strong foundations in human resources, infrastructure platforms, and a wide range of technologies, have exerted a powerful siphon effect. This is why, after more than ten years, the work of tiered diagnosis and treatment still faces significant challenges and pressure: the siphon effect is simply too strong. Nevertheless, at least the root of the problem is clear—the siphon effect of large hospitals is too great. Once the root is identified, the question becomes how to carry forward future efforts so that progress can be smoother than in the past.

There are, of course, many measures to change the care-seeking pattern. Foremost, in my view, is to free more than four million physicians, especially general practitioners, internists, and pediatricians, from the ivory towers of tertiary hospitals and enable them to practice widely in communities. That would put clinics staffed by trusted doctors right on people’ doorsteps. Over time, as habits shift, for common illnesses, community general practitioners would naturally serve as gatekeepers for every citizen.

With gatekeepers in place, why spend hours queuing at top Class-A tertiary hospitals [the highest grade among China’s tertiary hospitals] losing an entire day, only to see a doctor for a few minutes if the condition is not serious? Ordinary citizens do not want that either. Yet when a broad, trustworthy cadre of ordinary physicians and general practitioners is absent in local communities, rational citizens will naturally choose to register and queue at large hospitals. As a result, the shortage of appointments and beds becomes a hard nut to crack.

Therefore, having physicians step out of large hospitals and into communities and primary care is a crucial step. To make this happen, however, related institutional reforms will also need to advance in parallel.

For example, when physicians leave large hospitals to practice independently in society at large, question arise: Can their benefits, compensation, reputation, and career development stay comparable—or become better and more flexible—than in hospital settings? This is a comprehensive issue that relevant authorities will need to address in coordination.

If these issues are resolved gradually, demand across communities for general, primary-level services will remain—and it accounts for the majority. Then, for an individual, roughly 80–90% of annual medical needs can be handled on an outpatient basis, right? Accordingly, the career prospects of physicians in community practice would rest on a very broad public foundation.

In fact, take a look at the United States, Japan, or even the United Kingdom with a strong role of government for inpatient care and financing: their physicians are largely independent professionals. During medical school, they invest substantial funds: their own money, support from their parents, and student loans. After finally completing medical school and finding a job, they are left with the challenge of paying back substantial debt. Imagine the burden if all of them stayed in state hospitals, trying to service large loans on a fixed salary.

Thus, after medical school, most physicians set up community clinics to serve the public, gaining greater income, flexibility, and control over schedules than in fixed hospital posts, and in doing so realise their aspiration to practice medicine with dignity.

China still has a long road ahead on this front. On the one hand, physicians want the freedom to develop their careers; on the other, it is hard to give up the various benefits attached to positions in large state hospitals. As a result, many face difficult dilemmas and, at times, are forced into very painful choices.

For example, some spend many years in medical school, secure a job, then find their aspirations unmet and their circumstances difficult—so they abandon medicine altogether for other fields. Pharmaceutical companies, for instance, employ many well-trained physicians.

Some may remain in their posts at large hospitals, but then face a pressing question: how to earn a decent income and live with the same dignity as classmates from university or graduate school? Pressures like these have led to problems that no one wants to see.

Therefore, enabling more than four million physicians, either already in general practice or prepared to enter it, to move beyond large hospitals is essential. I am convinced that with policies that prevent heavy exit costs for leaving the public system and address their concerns, this transition is achievable.

Given such conditions, Chinese physicians, like their counterparts in developed countries, would have broad community settings to choose from. They would earn dignified incomes and hold respected positions, while the public could address routine outpatient needs at clinics close to home. Only then would the broader objective—access to care for everyone—be within reach.

If the transformation and upgrading of the medical service model stalls, and outpatients—even those seeking routine checkups—continue to flock to large hospitals alongside severe cases and inpatients, leaving them jammed while primary facilities sit empty, the goal of accessible care for all may remain foever out of reach.

Therefore, reforming and transforming the current medical model is crucial. Supply and demand are in tension, and the question is whether nonessential, nonurgent demand can be reduced or even eliminated so that the existing supply can better meet essential, urgent needs.

Healthcare demand arises when the body goes wrong. To trim the nonessential part of that demand, the issue is whether there is room to reduce how often such problems occur, or how severe they become.

What room exists to lower the risk of illness and cut preventable disease? The first example that comes to mind is smoking. Lung cancer incidence is high in China; it is also high worldwide, but not as high as in China. The greatest risk factor for lung cancer is smoking, and the choice to smoke or not rests with each individual. If smoking can be further reduced, disease risks attributable to it, especially tumor-related risks, will decline.

Are there other areas like smoking where improvements would reduce demand for medical services? Certainly: diet. Dietary structure may have an even greater impact and a wider reach. Consider dietary composition and, for example, obesity. Obesity is the greatest risk factor for many metabolic diseases; its control, to a large extent, is something individuals can influence decisively and proactively.

There is also pollution—PM2.5 and related pollutants—which are often viewed as sources of respiratory disease. In fact, pollution has a substantial impact on cardiovascular disease as well, because fine particles enter the body and do not remain only in the respiratory system; they also reach the vascular system and cause damage. Leading medical journal The Lancet has published several major papers showing that indoor and outdoor pollution are major contributors to cardiovascular disease. Pollution control, of course, is a task for government, the public, and society as a whole to tackle together.

Therefore, stronger governance and control of pollution would significantly lower the risk of discomfort, illness, or even serious disease caused by pollution, bringing this category of disease under control and preserving resources for patients with essential needs.

These are three entry points tied to everyday habits and behaviors that can reduce demand for medical services and free up resources for patients with essential needs, making it possible for more people to access care.

Beyond behavioral factors, certain medical measures can also reduce disease risk. For example, numerous studies show that for women over 50, regular breast cancer screening significantly lowers preventable incidence. Routine breast cancer screening is therefore a highly cost-effective measure for women over 50.

There are additional screenings for both men and women, such as regular colorectal cancer screening. For people over 50, regardless of sex, colorectal cancer poses a risk; routine screening lowers that risk, sparing patients unnecessary suffering and preserving limited resources for those with essential needs. Some diseases are genetic; patients with such conditions—the group with essential needs—would then have their medical needs better met.

In addition, antihypertensive drugs can lower demand for medical services. Hypertension is a root cause of many illnesses, and cancer is often called the emperor of all maladies. Modern medicine offers many ways to reduce high blood pressure; antihypertensive drugs are inexpensive, common, and highly accessible.

If affordable, effective antihypertensive and cholesterol lowering medications are made available, disease risks associated with the “three highs” can be reduced for many individuals. Demand for medical services would then decline further, reserving more of that capacity for patients with essential needs.

In short, behavior, screening, and affordable “little pills” can all help reduce preventable demand for medical services, allowing limited resources to be directed to those who cannot lower their risks through such measures. This is how demand can be reduced so that medical supply can better meet people’s needs.

From the supply side, are there measures that could improve medical service conditions so that more people, and ultimately everyone, can obtain necessary care? Yes. A distincitve feature of China’s healthcare system is that pharmaceuticals account for an unusually high share of total medical costs.

Let me give you an comparison. In most countries, especially developed ones, drug spending per visit typically makes up about 10% to 30% of total medical costs. In the United States, for instance, it is roughly 10% to 15%. Asian countries are somewhat higher, around 20% to 30%. It is rare for drug costs to reach 30% or 50% of total medical expenses.

By contrast, in the years before healthcare reform, pharmaceuticals accounted for as much as 50% to 60% of total medical costs in Chinese medical institutions.

For sure, after years of national healthcare reform, the share of pharmaceuticals in total medical costs has declined to a little over 30% today, but it remains high compared with other countries. Curbing inappropriate or excessive drug use requires scientific approaches. Rational drug use is a technical field, with many methods available for scholars to study how to reduce unnecessary and unreasonable medication use. Pharmacoeconomics, for example, is an effective tool for assessing whether a medicine offers good value and is worth using in clinical practice.

Pharmacoeconomics primarily evaluates drugs with the same indication by first comparing clinical effectiveness and then comparing prices, recommending affordable, high-value options that are readily accessible. Since its establishment in 2018, the National Healthcare Security Administration of China has convened many experts to systematically assess medicines’ value and cost—that is, their cost-effectiveness. In these evaluations, three groups of experts play especially important roles.

One group is clinical experts, who judge from a clinical standpoint whether a drug is more effective than alternatives. A second group is health insurance experts, who consider affordability: if a drug is recommended and widely used, could its cost overwhelm a city’s insurance budget? They run the numbers from an accounting perspective. A third group is economists, who compare one drug with another to determine whether it offers good value for money and provide cost-effectiveness recommendations. All of these tools can be applied more broadly so that, when medications are used in clinical practice, both money and time are well spent and outcomes are improved.

Most patients are not fully aware of the relative value and effectiveness of different drugs. Scientific evaluation can therefore be used to guide choices, helping patients reduce unnecessary expenses and enabling more people to obtain necessary medical services.

In fact, other areas of evaluation can also be undertaken. After a disease occurs, should treatment proceed with medication, with surgery, or with non-surgical, non-pharmacological approaches such as psychological counseling or convalescent care? These options within medical services can likewise be assessed systematically using pharmacoeconomic methods, providing evidence-based recommendations for patients and health insurance authorities, so that unnecessary medical expenses are reduced and limited resources are secured for those who truly need care.

Therefore, from the perspective of pharmaceuticals, there is considerable room for savings. With 30% of total medical expenses in China still devoted to drugs, the potential for efficiency gains is substantial.

Of course, one caveat is needed. For certain medicines, such as newly launched innovative drugs that have been on the market for only six months to two years, assessing overall effectiveness and cost-effectiveness may require broader use and longer follow-up. Only then can experts issue recommendations grounded in solid evidence. Given the short observation window, innovative drugs may warrant alternative regulatory approaches like special management measures to reflect limited data. On one hand, this encourages uptake of newly approved, often higher-priced therapies; on the other hand, it allows the market to validate performance. If outcomes are favorable, early investments can be recovered more effectively while patients gain access to newer, more valuable treatments.

In China’s healthcare system, pharmaceuticals account for a large share of expenses. Scientific evaluation, therefore, should be applied to drive further savings in this area, so as to direct freed-up resources to patients with essential needs.

Finally, one more point that can improve the quality and effectiveness of medical services: artificial intelligence, which everyone is talking about. AI is a digital technology; unlike physical technologies, it does not need to exist in visible, tangible form. It produces, delivers, and enables the use of products and services through data. Before digital technology and AI, sharing top-tier medical resources from Beijing or Shanghai with inpatients in Tibet, Yunnan, or Guizhou—so they could receive the same high-level surgical care—meant flying surgeons thousands of kilometers and spending days in transit to deliver services comparable to those in the major cities.

Imagine the high transaction costs, and patients often suffer more. Now, with AI, digital health, and telemedicine, a receiving site with a terminal and a stable 5G connection allows a surgeon in Beijing (e.g., at Peking Union Medical College Hospital) to operate a surgical robot remotely and provide procedures to patients thousands of kilometers away.

The associated costs should be lower than flying surgeons back and forth for several days, and the gap in medical services between developed cities and less-developed regions can gradually narrow. In short, AI has enormous scope for application in healthcare.

Another point: with AI in medicine, the path of a young surgeon looks very different. By traditional routes, attaining the refined, top-tier skills of a hospital’s master surgeons can take decades; matching that level is exceptionally difficult. This is why many of the most accomplished surgeons are older: decades are needed to build manual technique, achieve high-level whole-body coordination, and apply exceptional intellect. With AI and surgical robotics, however, the learning cost and the time required for a young surgeon to reach top-level proficiency are drastically reduced.

Lengthy, traditional experience-building becomes less necessary because AI and surgical robots integrate global case data. Once familiar with these tools, the practical time and effort required can shrink from twenty or even thirty years to perhaps two or three to reach a top surgeon’s level.

What does this mean for patients, and for the goal of making care accessible to everyone? In a top hospital, a department may have only one or two master surgeons, which is why getting an appointment with them is so difficult. People travel from other regions, call in favors, and insist on seeing that one expert because only one or two surgeons in the department operate at that level.

With AI and surgical robotics, however, more surgeons in the same department can rapidly deliver high-quality care. As a result, complex cases requiring surgery can be treated more effectively.

That is why I am convinced that AI’s applications in healthcare are so far-reaching. Its greatest benefit is that today’s scarce, high-level medical resources can extend to far more people, including those in remote and underserved areas.

To conclude, this talk has outlined the following entry points for making care accessible to everyone:

We first called for transformation of the service model: shift from a hospital-centered system to one rooted in primary care at people’ doorsteps, with large numbers of physicians providing general practice as first-line guardians. This would allow medical needs to be met more effectively.

We then talked about the potential to reduce unnecessary demand for services. Afterwards, we introduced the ways to optimize the use of medicines through pharmacoeconomic evaluation to curb waste, overuse, and misuse.

Finally, we emphasized digital health to apply artificial intelligence to raise efficiency and lower costs in service delivery. Most importantly, with digital technologies such as AI, scarce physician resources concentrated in major cities can be shared with residents in remote areas. With progress on these fronts, the goal of ensuring that everyone has access to healthcare will come steadily closer.