Introduction

Heart disease has remained the leading cause of death globally for the last 20 years. Nearly 9 million people died from heart disease in 2019 worldwide1. In China, cardiovascular disease (CVD) is the leading cause of death and premature death, where the absolute number of cardiovascular deaths increased by 46% between 1990 and 20132,3. The prevalent CVDs, such as hypertension, coronary heart disease, and chronic heart failure—need long-term monitoring and treatment.

The peculiarities of the Chinese healthcare system make CVD disease management a monumental and difficult task, given a population of more than 1.4 billion, a land area of 9.6 million square kilometres, and a significant CVD disease burden4. The inverted pyramid of the Chinese healthcare system focuses on treating acute CVDs in tertiary hospitals with resources concentrated at top hospitals5. Moreover, access to primary care for treating of common chronic CVDs must be improved. Thus, many patients with CVD are not able to achieve timely diagnosis and appropriate treatment. For example, while hypertension incidence is at 37.2%, awareness, treatment and control rates are 36.0%, 22.9%, and 5.7%, respectively6.

There are various evident strategies for tackling the CVD burden in China. Previous studies have mentioned “making the most efficient use of available [resources] to maximize the benefits for the people—providing equal, accessible and quality healthcare to everyone”, and focusing on raising awareness of and managing risk factors rather than only treating acute episodes5,7. The difficulty lies in strengthening primary care and striking a balance between primary care and secondary and tertiary care7. One report indicated that more coverage of in-person care does not necessarily lead to more service and better health, especially in rural areas8.

Incorporating technology to provide CVD care, especially mobile health (mHealth) and telemedicine can impact access to high quality care. In 2019, the WHO released guidelines on digital health, to improve the accessibility of medical care services9. In the context of this paper, mHealth is a method of telemedicine delivery (classified under the WHO Digital Health Interventions 2.4.1–2.4.3) that is delivered by mobile applications between a healthcare professional and patient10. mHealth apps with capabilities for telemedicine and telehealth will be referred to as telehealth (tHealth) apps. The Chinese State Council issued the Internet Plus Action Plan in 2015 and the Internet Plus Healthcare in 2018 to use mHealth as a tool to increase the accessibility and affordability of medical services across the country11,12. Previous reports on Chinese mHealth have focused on functions, such as diagnosis support, treatment, rehabilitation, and education regarding heart diseases via smartphone apps13,14. However, information on how mHealth impacts access to care and disease management at the national level is lacking.

Examining the longitudinal changes in the role of mHealth in CVD management can help explain the impact of the technology on CVD burden in China and potentially other developing countries with similar healthcare characteristics. Therefore, the aims of this study are to 1) describe the trajectory of tHealth services offered for cardiovascular care between 2016 and 2020 in China, 2) describe how tHealth services are being utilized in cardiovascular care, and 3) discuss the implications of tHealth for cardiovascular care access in resource-scarce regions and developing countries.

Results

App and mini-program selection

In the initial sample list, there were 1590 tHealth apps in 2016, with this number increasing to 4742 in 2017, and then decreasing to 4631, 4233, and 3255 in 2018, 2019, and 2020, respectively (Table 1). The final number of tHealth apps for CVD care were 207, 599, 417, 403, and 425 in 2016, 2017, 2018, 2019, and 2020, respectively (listed in Supplementary Table 1). The number of hospitals covered by tHealth apps consistently increased during the period of the study (1310, 1564, 1731, 1850, and 1970 in 2016, 2017, 2018, 2019, and 2020, respectively).

Table 1 App selection and screeningd
Full size table

Among the different providers of tHealth apps, public hospitals and private companies consistently provided the highest numbers of tHealth apps across the duration of the study, with 77 and 108 in 2016 and 304 and 92 in 2020, respectively (Table 2). The range of coverage of tHealth apps from government and private company providers, collectively, was 62.2%, which covered more than ten hospitals in 2016, which increased to 84.7% by 2020. In 2016, over 80% of tHealth apps from government and private company providers covered one entire province or more, whereas in 2020, most tHealth apps from the same providers covered only one city or hospital.

Table 2 tHealth app evaluation—Categorization and characteristics of providers
Full size table

Hospitals available

From 2016 to 2020, the numbers of hospitals available via tHealth apps steadily grew from 1310 to 1970, with all provinces having at least one facility providing telemedicine services by 2020. Between 2016 and 2020, Eastern China had the greatest number of hospitals providing telemedicine services, whereas northwestern China and northeastern China consistently had the least (Table 3).

Table 3 Distribution of hospitals providing CVD care via tHealth apps in China
Full size table

Hospital distribution of tHealth apps geographically and socioeconomically

Between 2016 and 2020, the number of hospitals providing telemedicine for CVD in China increased from 1310 to 1970, with an average annual growth rate of 10.7%. The two provinces with the highest GDP per capita had the most hospitals accessible via telemedicine and saw the greatest increase in the number of hospitals available via telemedicine. The provinces with more than a 100% growth rate in hospitals accessible via tHealth apps were the eight provinces with the lowest GDP (per capita) in the country (Fig. 1 and Supplementary Fig. 1). The five provinces with the highest hospital coverage rates of telemedicine are located along the eastern coast of the country, while the provinces in western China experienced the greatest increase in coverage rate (Fig. 2 and Supplementary Fig. 2). The raw data on the tHealth app distribution can be found in Supplementary Table 2.

Fig. 1: Heatmap of the number of hospitals accessible by tHealth apps organized by provincial GDP per capita.
figure 1

The figure shows the change in the absolute number and growth rate of hospitals accessible on tHealth apps from 2016 to 2020. (Note: warmer colours correspond to higher values).

Full size image
Fig. 2: Geographic heatmap of the hospital coverage rate via telemedicine.
figure 2

a Hospital coverage rates (hospitals with tHealth apps divided by total hospitals per province) in 2016 (top left) and 2020 (top right). b Hospital coverage growth rate per province between 2016 and 2020.

Full size image

Functional analysis of the apps and mini-programs

Telemedicine was provided by all the apps and mini-programs in our study. The other functions of hospital and third-party apps or mini-programs had some similarities (Table 4). Appointment scheduling (84.7%), medical records (43.3%), disease awareness (20.5%), and reminders (15.1%) were the most common functions in 2020. In contrast, clinical decision support and discussion forums were rarely provided in the same app. Hospital platforms focused primarily on tracking patients’ medical history and disease awareness, whereas third-party platforms focused on aspects of chronic disease management, such as online pharmacies.

Table 4 Services offered for CVD care by tHealth apps
Full size table

Patient identity verification was required on both hospital and third-party platforms (84.1% in 2016 increasing to over 98.0% in 2017 and plateauing in the following years). Doctor selection availability was 88.0% in 2016 and 94.6% in 2020. Online payments were available in 84.2% of all tHealth apps in 2020.

Patient usage of telemedicine

Open access data from HaoDF were used to examine telemedicine usage for CVD in China. In 2016, there were 222,000 telemedicine cases for CVD and 177,000 cases in 2020. The diagnoses were submitted by the doctor as part of the routine service follow-up by the app provider. Further details of each diagnosis were not collected; hence, the broad category of information available was used. The most common diseases from most to least prevalent identified during telemedicine visits in 2016 were coronary heart disease, arrhythmia, and hypertension; in 2020, they were coronary heart disease, hypertension, and arrythmia (Fig. 3 and Supplementary Fig. 3). Experienced doctors, such as chief physicians and associate chief physicians, were preferred by more than half of the patients (64.0% in 2016, 62.0% in 2020).

Fig. 3: Common CVD diagnosis at telemedicine visits.
figure 3

The radar chart shows the CVD distribution of patients seeking telemedicine care via tHealth apps between 2016 and 2020 on a single large platform44. 44HaoDF Online - Open Access Platform. https://open.haodf.com/.

Full size image

We analysed patients’ consultations with doctors outside their own provinces (Supplementary Table 3). The percentage increased from 34.0% in 2016 to 67.3% in 2020. Nearly half of the interprovincial consultations were to physicians from hospitals in Beijing and Shanghai in 2016, and these regions also had the highest and second highest GDP per capita. In 2020, the inter-provincial consultations to Beijing and Shanghai were lower, at 30%, with regions ranked 3–10 and 21–31 in GDP per capita seeing an increase in consultations. Patients from less developed regions contacted doctors in more developed regions (Fig. 4 and Supplementary Fig. 4), with high percentages of patients domiciled in Xizang (92.2% in 2016 and 94.3% in 2020) and Qinghai (88.6% in 2016 and 93.1% in 2020). Shanghai, Beijing, and Guangdong consistently had the lowest number of patients contacting doctors for interprovincial consultations.

Fig. 4: Interprovincial consultations.
figure 4

Sankey diagram showing the domicile province of the tHealth app patient and the location of the service-providing hospital in (a) 2016 and (b) 2020 for cardiovascular patients on the HaoDF platform44. 44HaoDF Online - Open Access Platform. https://open.haodf.com/.

Full size image

Discussion

This study revealed that tHealth apps can rapidly increase access to cardiovascular care for large populations in ways that physical facilities cannot. tHealth can address China’s CVD burden despite its geographic size and population. On the healthcare provider side, the absolute number and coverage rate of hospitals providing telemedicine services have increased throughout China. In addition to telemedicine, other services, including reminders, medical records, online payments, pharmacies, and disease education, are available for self-management and patient-physician engagement in managing chronic CVDs. On the tHealth utilization side, patients with common CVDs (e.g., coronary heart disease, arrhythmia, and hypertension) most frequently use tHealth apps to receive care. Furthermore, the utilization of these apps realized a large-scale virtual patient flow from poorer to more affluent regions. The findings from China can serve as a reference for developing or resource-poor regions.

The most common conditions addressed via telemedicine were hypertension, coronary artery disease, and arrhythmia, which is consistent with reports on the epidemiological prevalence of CVDs in China15. The data reported in this study include both initial and follow-up visits. Usage numbers indicate that access to care for CVD patients has increased; however, whether the care delivered aligns with guidelines, and whether clinical outcomes are affected remain to be determined.

In 2020, private hospitals and primary hospitals constituted the majority of medical institutions in China (66.5% private hospitals versus 33.5% public hospitals; 34.6% primary hospitals, 29.4% secondary hospitals, and 36.0% tertiary hospitals)16. However, according to the results of this study, this ratio was reversed for telemedicine where public hospitals and tertiary hospitals were the major providers. Although secondary and primary hospitals were a focal point for the government to improve the healthcare system, the main burden remained concentrated on public and tertiary hospitals. Similarly, for in-person clinical visits, approximately about one-third of all public hospitals provide over 80% of medical services, whereas tertiary hospitals provide 57.0% of medical services16. This preference for tertiary care reflects the current imbalances within the Chinese healthcare system. Primary care facilities and private hospitals still have significant room to grow before taking full responsibility for diagnosing and treating common CVDs. The inverted pyramid of the Chinese cardiovascular care system is not sustainable5. More attention should be given to the burden of telemedicine on doctors in these limited hospitals. Therefore, population-level strategies, such as strengthening primary care, hierarchical disease management (where different hospital levels treat based on disease severity and urgency) and doctors practising at multiple sites (which allow for the flow and communication of healthcare professionals between hospitals) should be further popularized in the future17.

In this study, we found that the tHealth services offered are becoming more comprehensive. The recent shift to internet hospitals is driving the push towards comprehensive medical care18. One aspect is the delivery of medical care now including diagnosis, treatment, and continuous care, which started with nondiagnostic consultations and reminders in the 2010s. Additionally, patients can be diagnosed, and have prescriptions refilled and delivered remotely. Supportive services are also becoming more comprehensive and consolidated with non-medical care services such as payments, physician selection, patient education, and health records, which are mainly provided in individual applications13,19. The trend observed in this study is the consolidation of services on applications and uniformity across service platforms. Furthermore, during the period of this study, hospitals (rather than technology providers) made up the majority of the tHealth providers. These trends suggest that patients can more easily consult physicians at top-tier hospitals without needing to be physically present and imply a greater likelihood of reducing overcrowding by patients in top-ranked hospitals20.

The trends observed indicate that adoption of tHealth services occurs at a higher rate in poorer regions, which are more likely to benefit in the coming years. The provinces with the highest growth rates were those with the lowest GDP per capita. Populous provinces also had the top four greatest increases in the absolute number of tHealth hospital services.

tHealth apps can provide interregional support from resource-rich areas to poorer areas. This is indicated by the flow of patients from other provinces to Beijing, Shanghai, and Guangdong, which are the areas with the most top-ranked hospitals (51 of the 100 top-ranked hospitals)21. The implications can carry over to other developing Asian countries with large populations, such as Indonesia, the Philippines, and Bangladesh, where there is a similar trend of top-ranked cardiovascular care facilities residing in three main cities/states in the country22,23,24. In poorer areas having access to care via tHealth can help strengthen primary care to prevent and manage chronic conditions. tHealth can also provide a medium to disseminate cardiovascular health education and promotion to regions that often do not receive such information. This study provides evidence that tHealth as a medium of cardiovascular care delivery can alleviate the pressure on the healthcare system’s inverted pyramid of providers5.

The future development of tHealth should focus on regulation and quality. The regulation of tHealth services from the perspectives of information technology privacy and security, provider licensing, and liability should be considered on all operating platforms25. Across Asian developing countries, tHealth platforms can cover more than 10 million cardiovascular patients per jurisdiction, where not all are subject to clear regulation by local health ministries; such is also the case in developed markets26,27,28. Since the beginning of the COVID-19 pandemic, the number of internet hospitals in China has been increasing18,29. Internet hospitals are regulated tHealth services provided by licenced medical facilities. Practising physicians providing services on online hospitals are regulated similarly to physical hospitals or clinics and are subject to the same legal protections. Additionally, ensuring that states/provinces update licensure requirements and allowing physicians to practice beyond their own province or state jurisdiction is important30.

Furthermore, sufficient triaging infrastructure and patient education should be implemented to ensure the appropriate use of tHealth services. Telemedicine via apps can be helpful for chronic conditions but still cannot fully replace initial visits and diagnoses. Patients should be guided and educated on when in-person or emergency care is needed.

To ensure the continued democratization of healthcare via tHealth, payment options that are seamlessly integrated with commercial and national health insurance are necessary. According to the study data collected, the majority of payments were out-of-pocket, with few integrated payments with an insurance provider. The national government first highlighted the importance of covering fees during online diagnosis and treatment after the COVID-19 pandemic, which was a unique catalyst for tHealth development31. Studies have shown that integrating private or national insurance policies with telemedicine services, especially in rural areas, and across regional jurisdictions, is crucial for maintaining usage and adoption can improve the usage rate and user-adherence or satisfaction30,32.

The authors recognize the limitations of this study. The data reported were from publicly available data rather than officially published reports. However, a range of platforms was selected to ensure the representativeness of the study. Meanwhile, we acknowledge that there are limitations on the representativeness of the android apps sampled. The top app stores in China used for this study accounted for more than three quarters of the market coverage. Additionally, the use of a single app to evaluate patient usage has limitations. The app used has been on the market the longest, has the widest reach in the country, and has the largest market share; thus, the results can be used to describe patient trends across provinces. Second, this study describes the implications pertaining to access to care, not the impact of tHealth on CVD outcomes including incidence, awareness, treatment, and mortality rates. Third, the quality of services delivered across providers was not determined. Future studies can examine the impact of quality and outcomes from tHealth at the population level.

From 2016 to 2020, there was a shift in tHealth services from third-party platforms to hospital platforms targeting CVDs in China. The poorer regions had the highest growth rates and interprovincial and regional care delivery rates, especially when telemedicine was delivered remotely from resource-rich areas. This approach appears to be a sustainable model for developing countries to democratize CVD care across large populous countries. A comprehensive service suite offered across most applications allows for more health equity in managing CVDs.

Developing countries, resource-poor markets, or countries with an imbalanced distribution of medical resources can refer to China’s tHealth adoption as a model. Specific technologies that are available for the whole population that spearheaded the change in China. Even in developing countries or regions with low GDP, internet access and mobile phone usage rates in adults are nearly universal33,34. Additionally, policies encouraging and hospitals willing to adopt tHealth services facilitated comprehensive coverage, thus providing easier access to the population, especially in poorer regions. The technology we believe to be most important for access is hospital and tHealth connectivity, which entails appointment scheduling and telemedicine. Appointment scheduling for an in-person consultation at a local or distant facility can provide assurance to be seen and save patients from making multiple trips. Telemedicine is also an important contributor; however, there must be staff for triaging, and payment and digital literacy from patients and providers.

Methods

Ethics

No personal identifiers or individual data from patients or doctors were collected in this study; thus, informed consent was waived by the ethics committee of Peking Union Medical College Hospital.

Telehealth app selection

mHealth apps offering telemedicine services in China were prospectively selected on publicly available platforms between July and October of 2016 to 2020 using the keyword “appointment making”. The inclusion criteria were patient-facing, operational apps focused on Mainland China and in Simplified Chinese. The exclusion criteria were apps for medical professionals only, not in Simplified Chinese, not serving Mainland China, non-existent/non-operational, providing purely non-care-related services or unable to realize telemedicine functions for CVD. Only the apps that satisfied the inclusion and exclusion criteria were used for subsequent analysis. Two main types of mobile applications in China were selected: stand-alone mobile apps and mini-programs. Mini-programs are apps embedded within another app, a feature allowing users to use a collection of apps without installing multiple applications on their smartphones35. Stand-alone apps were obtained for the Android and iOS operating systems. For Android, apps were selected from the five top Android app stores, which accounted for more than three quarters of the market coverage, namely, MyApp (Tencent Technology (Shenzhen) Co., Ltd., Shenzhen, China), Baidu (Baidu, Beijing, China), 360 (Qihoo 360 Technology, Beijing, China), Huawei (Huawei Technologies Co. Ltd., Shenzhen, China), and Oppo (Guangdong Oppo Mobile Telecommunications Corp., Ltd., Dongguan, China)19,36. For iOS, Apple’s App Store (Apple Inc., Cupertino, CA, USA) was used to obtain apps. WeChat (Tencent Technology (Shenzhen) Co., Ltd., Shenzhen, China) was the source of the mini-programs.

The stand-alone apps or mini-programs (henceforth collectively referred to as tHealth apps unless otherwise specified) described above were evaluated based on above criteria for inclusion into and exclusion from this study. We ensured that the tHealth apps provided services adhering to the definition of telemedicine-the provision of healthcare services at a distance-through the use of information and communication technology, where patients and providers are separated by distance37,38,39,40. Apps providing purely non-care-related services such as health monitoring, exercise, lifestyle management, medical science education, and over-the-counter drugs were excluded from the study. The apps were then evaluated to ensure that the user could obtain access to cardiovascular care, as defined by patients being able to make an appointment at a department of cardiology or cardiac surgery; both online and in-person appointments were included (note that, in the context of China, appointments for online or offline services are paid upon making the appointment; hence, no separate category of paid online consultation was made). The selection process is shown in Fig. 5 and Table 1.

Fig. 5: Flowchart of the tHealth app selection and hospital identification process.
figure 5

In Phase A, the apps were selected from the commonly used app stores and mini-program providers in Mainland China. In Phase B tHealth apps were classified based on the type of entity providing the service. In Phase C, hospitals were identified and aggregated as the final service-providing entity.

Full size image

The initial identification, screening, and evaluation were completed by four authors, and each app’s evaluation was subsequently reviewed by another evaluator to ensure accuracy. Any discrepancies were discussed until a consensus was reached, and the final decision was jointly approved by two other authors.

tHealth app evaluation—outcome variables

The variables related to the tHealth app included operator type, facility location, number of facilities, facility type (primary, secondary, or tertiary), and services offered. Patient usage data included diagnosis, date, the patient’s city of residence, and facility name and location.

tHealth app evaluation—categorization and characteristics of providers

We identified the characteristics of each app, hospital, and hospital location prior to further analysis. Each app was categorized by the operator or owner of the app into one of four categories: public hospital, private hospital, government, or private company. The number of hospitals each tHealth app provided access to was categorized as single, two to three, four to ten, or more than ten. Each hospital that was available for telemedicine was then individually listed with duplicates removed. The hospitals were then tagged to a geographic location and hospital level (primary, secondary, or tertiary). Geographies were defined as one of thirty-one provinces, municipalities, or autonomous regions, henceforth collectively referred to as provinces. Taiwan, Hong Kong and Macau were not included in this study. The following data were identified for each geographic region: CVD burden, GDP per capita, population size, percentage of the population above 65 years old, the percentage of the population with a college degree or higher, and the number of internet users, and were collected for the years 2016, 2017, 2018, 2019, and 2020 (Supplementary Table 4)2,41. When it was not possible to obtain data for the exact year, the most recent available value was used excluding all data prior to 2015 (Supplementary Table 5). Hospitals in China are categorized on a three-tier system based on the following factors including but not limited to, staffing, bed count, specialties offered, and operations performed, with tertiary hospitals being the highest designation42.

tHealth app evaluation—services offered

The services offered by each app were identified according to a previous study’s classification—reminders, medical records, pharmacies, disease awareness, clinical decision support and medical education19. Other telemedicine-related functions, including identity verification, doctor selection, and online payment, were also captured. Definitions of each service can be found in Supplementary Table 6.

tHealth app evaluation—usage for access to CVD care

To investigate real-world patient usage of tHealth apps in China, a large private tHealth app provider called HaoDF (HaoDF Online Corp., Beijing China), which provides medical services to up to 72 million patients was selected43. Deidentified app usage data via an open access data platform granted for scientific research was obtained44. The App usage data for 2016 to 2020 included the following data elements: each episode of care’s diagnosis, date of consultation, the patient’s city of residence, and facility name and location.

Data analysis

This was a longitudinal descriptive study of tHealth apps. Descriptive statistics were used for reporting results where we reported the count and percentage of occurrence of the variables. Various factors affecting access to CVD care were analyzed, including: 1) longitudinal changes in tHealth services provided per region; 2) the hospital coverage rate per province by tHealth apps; 3) the distribution of specific cardiovascular conditions being sought on tHealth apps; and 4) the flow of patients across geographies to receive telemedicine. The outcomes are reported in the tables and were visualized through heatmaps, geographic heatmaps, radar diagrams or Sankey diagrams.