Skip to main content
Download PDF
Download ePub

Topic identification and content analysis of internet medical policies under the background of Healthy China 2030

Health Research Policy and Systems volume 22, Article number: 132 (2024) Cite this article

Abstract

Objective

This study aims to analyse the content of internet medical policies, grasp the inherent laws of the development of internet medical policies and provide references for improving the policy system.

Methods

On the basis of web crawler technology, 436 internet medical policies issued by the central and local governments after the Healthy China Strategy was proposed were collected. The BERTopic model was used to extract topics, and a comprehensive analysis of China’s internet medical policy texts was conducted through the analysis of topic content, topic hierarchy and topic clusters.

Results

A total of 27 topics were identified in China’s internet medical policies, mainly focussing on five topic clusters: medical service regulation, elderly nursing and children’s healthcare using traditional Chinese medicine, user safety guarantees, health education and communication and infectious disease treatment and recovery guidance.

Conclusions

Different topic contents play a role in promoting the comprehensive and standardized development of internet medical services. However, there is still room for further improvement in policy integrity, consideration of multiple types of users, and implementation effectiveness. Continuous efforts are needed to optimize the effectiveness of policies.

Peer Review reports

Introduction

The “Internet + ” action plan was first introduced in China’s 2015 Government Work Report with the aim of integrating the internet into fields such as healthcare, education and finance, fostering the growth of emerging industries [1]. In recent years, “Internet + Healthcare” has rapidly developed as an innovative medical service using the Internet as a platform. By leveraging cloud computing and big data technologies, it has deeply integrated with traditional medical services, continuously expanding service offerings and gradually widening its reach, demonstrating significant success. This has played a critical role in supporting the comprehensive promotion of the Healthy China initiative. According to the 2023 China Digital Health Market Data Report, China’s internet healthcare market is projected to reach 385.4 billion RMB, marking a 24.25% year-on-year growth. The robust growth of “Internet Plus Healthcare” has garnered government attention, with a series of supportive policies, such as the 2018 State Council’s Guidelines for Promoting the Development of “Internet Plus Healthcare”, providing a favourable policy environment for its advancement. This has further propelled the regulated and healthy growth of “internet healthcare”. Under proactive policy guidance, internet healthcare services have achieved remarkable results, progressively meeting the increasing demands of society and the market and offering convenient medical services to the public.

Understanding the key focus of internet medical policies (including legal documents or broader policy pronouncements) is crucial for guiding its direction, fostering its robust development and advancing the Healthy China initiative. From a policy perspective, in-depth comprehension of the central topics of internet healthcare policies provides insight into the critical elements of its development, aiding in promoting internet healthcare work in China and warranting widespread attention and discussion. This study examines the policies issued by central and local Chinese authorities to explore the orientation and trends of the core topics of policies, understanding the inherent patterns of internet healthcare policy development. This serves as a reference to improve the policy framework and ultimately support the future growth of China’s “internet healthcare”.

Related work

Internet healthcare policy

Many scholars have conducted research on internet healthcare policies using methods such as content analysis and bibliometrics. For example, Yang et al. used content analysis to examine China’s “Internet + Healthcare” policy and concluded that it has a promising future, emphasizing the need for infrastructure improvement and high-quality medical services [2]. Thorlu-Bangura et al. analysed UK government policies on health inequalities and digital health, finding that future policies must consider the heterogeneity of different ethnic groups [3]. Wei et al. combined policy bibliometrics, content analysis and the The Policy Modeling Research Consistency Index (PMC) index model to analyse the characteristics and distribution of policy tools in internet healthcare policies. Their study revealed that current Chinese internet healthcare policies align with the stage of internet healthcare development, although there is an imbalance in implementation [4]. From a methodological standpoint, methods based on word frequency statistics provide a superficial level of analysis and do not account for semantic relationships between words. Meanwhile, content analysis requires significant manual effort, is less efficient and can be subjective. By automatically extracting relevant topics from policy texts and summarizing key content, this study can effectively consider semantic relationships between words, understand the deep meanings of topics in texts and explore the development direction and regulatory approaches of internet healthcare more deeply, systematically analysing the formulation and implementation of policies.

Policy text topic mining

Conducting thematic mining on policy texts requires integrating text representation models from natural language processing with various techniques from text mining. This approach offers higher analytical efficiency, lower labour costs and stronger result replicability compared with traditional policy content analysis and bibliometric analysis [5]. The most widely used topic model currently is the latent Dirichlet allocation (LDA) model [6]. For instance, Zhou et al. applied the LDA model to discover topics in China’s wind power policies and discussed potential future changes needed in the policy domain [7]. Ma et al. used the LDA model to extract six major topics from family education policy documents and proposed three optimization paths for policy design [8]. Song et al. combined the LDA model with k-means clustering to perform a quantitative analysis of Chinese central and local food safety policies, revealing continuity in the number and content of central and local policies [9].

Regarding research content, no scholar has yet applied topic modelling to the study of internet healthcare policies. In terms of topic modelling methods, although the LDA model has produced abundant results in policy text thematic mining, it cannot avoid the incompatibility between density-based clustering and centroid-based sampling methods. As artificial intelligence technology advances, BERTopic – a topic model based on pre-trained language models – has gained widespread application. This model can better interpret topics and retain important keywords within topics [10].

BERTopic model

Currently, the application of the BERTopic model is primarily focussed on extracting topics from scientific literature and social media content. For example, Lee et al. used the BERTopic model to conduct thematic modelling and analysis of news and academic papers related to environmental, social and governance (ESG) factors. Their work emphasized the significance of ESG factors in business management and highlighted key topics and trends in discussions surrounding ESG [11]. Uncovska et al. used the BERTopic model to analyse the ratings and user reviews of German mobile medical applications. The results showed that regulated applications tend to receive more positive feedback [12]. Lalk et al. employed the BERTopic model to conduct thematic modelling on 552 psychotherapy transcripts from 124 patients. They used machine learning methods to predict symptom severity and treatment approaches, demonstrating the potential of thematic modelling in psychotherapy [13].

These studies demonstrate that the BERTopic model, as one of the most advanced topic modelling methods, can effectively perform thematic modelling, identification and content analysis of texts. Therefore, this study aims to apply the BERTopic model to policy research and visualize the distinctions and relationships between topics.

Review of research

Both national and local governments have actively introduced policies to promote the development of internet healthcare, emphasizing the importance of understanding policy content and trends to further drive its growth. However, current research on internet healthcare policy is limited, with methods predominantly focussing on content analysis and bibliometrics, resulting in conclusions that lack depth and detailed insights. Topic modelling, a method capable of extracting abstract topics from documents, is commonly used to identify key concepts in texts [14]. Besides the widely used LDA method, the application of BERTopic, a deep learning-based model, is becoming increasingly popular [10]. However, this approach is often applied to scientific literature and social media content, with limited research on policy texts. Therefore, we attempt to introduce the BERTopic model for discovering topics in internet healthcare-related policies, leveraging its strong performance in handling large-scale documents. This approach aims to improve the quality of policy text topic discovery, identify key areas of focus and existing issues in internet healthcare policies and provide valuable insights for future policy-making and optimizing internet healthcare services.

Research design

Research framework

This study investigates the orientation and trends of relevant policy topics. The research process consists of four steps: (1) collecting all relevant central and local policy documents from various policy document repositories; (2) preprocessing the study samples; (3) training the selected thematic analysis model and manually merging similar topics to finalize the core topics of policy texts and conduct in-depth analysis of the specific orientations within each topic; and (4) revealing the evolution trends of core topics in policy texts on the basis of the proportion and quantity of each topic over the years. The research framework is illustrated in Fig. 1.

Fig. 1
figure 1

Research framework

Full size image

Data sources and processing

To comprehensively collect relevant policy texts, this study integrates state policies and recent literature research. It uses keywords such as “Internet healthcare”, “Internet health”, “online healthcare” and “telemedicine” for searches in titles and full texts. A web crawler designed using Octoparse v8.68 was used to search the Peking University Law Database (an example of the collected policies is shown in Fig. 2). In addition, the State Council’s policy document repository, the national laws and regulations database and provincial government websites were supplementary sources of data. The search period covered October 2016 (following the release of the Healthy China 2030”initiative) to March 2024, and the retrieval date was 1 April 2024. A total of 1676 documents were retrieved. These documents were manually screened according to specific criteria, resulting in a final set of 436 policy documents, which includes all the policies cited in this study, with an average length of 3293.35 characters. The screening criteria were: (1) all currently effective or soon-to-be-effective central and local (provincial, city-level) policy texts; (2) policy types such as laws, regulations, ordinances, rules, guidelines, notices and measures, excluding informal decision-making documents such as approvals, replies and reports; and (3) policy text content must relate to Internet healthcare.

Fig. 2
figure 2

Examples of policy text capture

Full size image

Upon manually reviewing the 436 policy documents, it was found that all were related to the healthcare field. However, some had a weaker connection to internet healthcare. To ensure comprehensiveness, Python 3.7 was used to extract and consolidate paragraphs related to “internet” from these documents into a corpus for topic identification. Subsequently, Python’s regex function was employed for regular expression cleaning, using patterns such as “a-zA-Z0-9\s + & = $:” to remove symbols, labels and numbers from policy texts. Harbin Institute of Technology’s stop word list was used to eliminate words with unclear meanings, such as “some” and “among others”. The jieba Chinese word segmentation library was utilized to segment policy texts. The resulting corpus of internet healthcare policy texts contained a total of 1,353,584 characters.

Methodology

We employ a deep learning-based topic modelling approach using the BERTopic model to extract, mine and cluster topics from internet healthcare policy texts. The process consists of several steps: First, the pre-processed word sequences are input into the paraphrase-multilingual-MiniLM L12-v2 pre-trained model on the basis of the SBERT algorithm to generate high-dimensional vectors. Then, the UMAP dimensionality reduction algorithm is used to reduce the dimensionality of the generated high-dimensional vectors, followed by the application of the HDBSCAN algorithm to cluster the reduced vectors. Finally, c-TF-IDF scores for topic words are calculated, and the top five words by score are selected as topic representatives.

Results

Distribution of policy release dates

The distribution of policy release dates obtained in this study is illustrated in Fig. 3. As shown in the figure, the number of internet healthcare policies released from 2016 to 2024 exhibits significant fluctuations. On 25 October 2016, the Central Committee of the Communist Party of China and the State Council officially issued the Healthy China 2030 Planning Outline. In the following 2 months, 28 related policy documents were released. Over the next 2 years, the number of policies continued to increase, reaching a first peak in 2018 with 65 documents, indicating a growing governmental focus on internet healthcare. The number of policy releases slightly declined in 2019 and 2020, with 52 and 55 documents issued, respectively. However, in 2021, the number of policies surged to 74, reaching its peak, likely due to the increased governmental emphasis on internet healthcare in response to the coronavirus disease 2019 (COVID-19) pandemic. In the following 2 years, the number of policies returned to levels similar to those seen in 2019 and 2020.

Fig. 3
figure 3

Number of internet healthcare policy releases, 2011–2024

Full size image

Topic content

By using the BERTopic model for topic modelling on internet healthcare policy texts, 27 topics were automatically generated. The optimal clustering effect and the clearest semantics were selected, resulting in eight highly significant topics. The top five keywords for each topic were extracted and visualized on the basis of their importance to the topic. Figure 4 illustrates the proportion of keywords for each topic. Topics 0, 1, 4 and 7 displayed relatively balanced keyword distributions, while topics 2, 3, 5 and 6 exhibited some variation in keyword proportions. In topic 0, “internet” had the highest proportion at 0.04, while “hospital” had the lowest at 0.02. In topic 1, “health” was the most significant topic, accounting for 0.03, while “development” had the lowest share at 0.02. In topic 2, “elderly” was the most prominent topic, with the highest proportion at 0.07, while “eldercare” had the lowest at 0.03. In topic 3, “regulation” held the highest proportion at 0.04, whereas “industry” had the lowest at 0.02. These topics reflect that internet healthcare policy texts emphasize technical applications, target audiences and regulatory issues.

Fig. 4
figure 4

The top eight topics and their subject terms

Full size image

Topic hierarchy

To understand the underlying hierarchical structure of topics, the HDBSCAN hierarchical clustering algorithm was used to cluster the relationships between topics and visualize the results, as shown in Fig. 5. The clustering results provide a clear representation of the connections between different topics at various levels. For example, topic 0 (internet healthcare services) and topic 3 (regulation and oversight) are closely connected; the 2022 guidelines for the supervision of online diagnosis and treatment services from the National Health Commission set specific regulations for Internet healthcare services, covering aspects such as medical institution supervision, personnel oversight and quality safety management. Topic 1 (traditional Chinese medicine and health) and topic 4 (rural and township health centres) are directly related. In 2022, the National Administration of Traditional Chinese Medicine issued the notification, Development Plan for Traditional Chinese Medicine Informatization during the 14th Five-Year Plan, which promotes the deep integration of traditional Chinese medicine health services with the internet. In the same year, the State Council’s Office issued the Development Plan for Traditional Chinese Medicine during the 14th Five-Year Plan, calling for strengthening traditional Chinese medicine departments in primary healthcare institutions and striving to equip all community health service centres and township health centres with traditional Chinese medicine clinics and practitioners. Topic 2 (eldercare) and topic 9 (traditional Chinese medicine rehabilitation) are directly related. Research indicates that traditional Chinese medicine nursing has unique advantages and has been widely used in clinical practice, playing an indispensable role in rehabilitation care [15]. In 2019, the National Health Commission’s “Internet Plus Nursing Service Pilot Program” focussed on providing nursing services such as chronic disease management, rehabilitation care and specialized care to the elderly, disabled or terminally ill patients. Topic 5 (child healthcare) and topic 17 (traditional Chinese medicine health) have an indirect relationship. The National Health Commission’s “Healthy Children Initiative (2021–2025)” emphasized strengthening children’s traditional Chinese medicine services while leveraging various internet communication platforms to conduct online child health assessments and guidance.

Fig. 5
figure 5

Topic hierarchy

Full size image

Topic cluster

Topic spatial distribution

To visualize the concentration of internet healthcare policies among topics, core topic groups and the number of outlier topics, the model.visualize-topics() function was used to convert topic vectors into two-dimensional vectors and generate a scatter plot of the thematic space distribution, as shown in Fig. 6. Each circle represents a topic, with the size of the circle indicating the frequency of the topic across all documents. The distance between circles is determined by the semantic relatedness between different topics; the closer the distance, the higher the semantic association between the topics. The results demonstrate that internet healthcare policies can be grouped into six thematic clusters, with two clusters near the centre representing core topic groups and the other four clusters as outlier groups. The larger area of the points in the two core topic groups suggests that most internet healthcare policy topics are concentrated around the core groups.

Fig. 6
figure 6

Topic spatial distribution

Full size image

Topic similarity

To explore the synergistic relationships between topics, cosine similarity between each topic and the others was calculated and used to construct a similarity matrix. The thematic similarity heatmap was then generated, as shown in Fig. 7, where the intensity of colour indicates the level of similarity between topics – the darker the colour, the higher the similarity. The results reveal that the similarity between topics 0 and 20 is higher than that between topics 21 and 26, suggesting that the first 21 topics share more concentrated semantics and likely belong to the core topic groups, while the latter six topics have more dispersed semantics and may be more related to the outlier groups.

Fig. 7
figure 7

Topic similarity heatmap

Full size image

Document spatial distribution

To analyse the thematic content by displaying the distribution of documents within each topic, the model.visualize_documents() function was used to generate a visualization of the Internet healthcare policy document-to-topic distribution, as shown in Fig. 8. The distance between points reflects the semantic similarity of documents, while the colour of the points represents the topic they belong to. This visualization provides insight into the specific policies covered under each topic, which assists in further refining the topic clusters for this study.

Fig. 8
figure 8

Documents–topic distribution atlas

Full size image

Dynamic topic model

Dynamic topic modelling (DTM) is a technique used to analyse changes in topics over time. In BERTopic, DTM is executed by calculating the c-TF-IDF for each topic and time period. Figure 9 illustrates the trend of different topics from 2016 to 2024. Among these, topic 0 (internet healthcare services) consistently maintained a high frequency each year, peaking in 2018 before gradually declining. In contrast, the distribution of other topics, such as topic 2 (eldercare), remained relatively stable, indicating that policy development in these areas progressed more slowly and did not experience significant shifts in priority.

Fig. 9
figure 9

Visualization of topics over time

Full size image

Topic cluster

On the basis of the thematic space distribution, topic similarity and document space distribution results, as well as manual selection of the most optimal clustering results and clearest semantics, the internet healthcare policy texts were ultimately categorized into five thematic clusters. The results are presented in Table 1.

  1. (1)

       Regulation of internet medical services

Table 1 Internet healthcare topic group
Full size table

The thematic cluster includes seven topics, such as Topic 0: Internet healthcare services and Topic 3: Regulation and supervision. These topics focus on the safety, reliability and rationality of internet healthcare practices, as well as the development and improvement of relevant standards and norms.

Cases of unregulated behaviour in internet healthcare have repeatedly sparked public outcry, with numerous media outlets exposing problems within the industry [16]. In response, national and local health administration departments have developed and issued policies to oversee and manage internet healthcare services, including the eligibility for entry, treatment process and service orientation. They also intervene and restrict medical practices that endanger patient health and safety [17, 18]. In 2018, the National Health Commission released the Interim Measures for the Management of Internet Hospitals, outlining specific directions for the admission and standardized management of Internet healthcare services. In 2022, the National Health Commission and the National Administration of Traditional Chinese Medicine jointly issued the Interim Measures for the Supervision of Internet Diagnosis and Treatment, further clarifying the rights and responsibilities of internet hospitals and regulatory plans. However, gaps remain in existing regulatory policies as internet healthcare rapidly evolves. Different online platforms and hospitals providing Internet healthcare services are subject to different administrative authorities for entry review and regulation, which may lead to inconsistencies in policy implementation due to varying interpretations and execution [19]. Additionally, there is a lack of guiding documents and industry standards for the use and management of electronic medical records and medical insurance in internet healthcare services, which complicates regulation [20].

In summary, the regulation of internet healthcare services remains a key focus for future policy design. Moving forward, national and local health administration departments should establish clear guidelines for the overall direction of internet healthcare development and provide detailed regulations that specify responsibilities and strengthen the operability of policy implementation to ensure the protection of patients’ health and legal rights.

  1. (2)

       Internet-based traditional Chinese medicine eldercare and child healthcare

The thematic cluster includes eight topics, such as Topic 1: Traditional Chinese medicine and health and Topic 2: Eldercare, which focus on advancing the efficient use of internet healthcare platforms to deliver eldercare, child health services and more, thereby enhancing service capabilities and coverage.

With the increasing challenges of an ageing population and rising rates of chronic diseases, there has been growing demand for traditional Chinese medicine services [21]. In October 2019, the National Health Commission issued the Several Opinions on Further Promoting the Development of Integrated Medical and Elderly Care, highlighting the unique role of traditional Chinese medicine in preventive care and chronic disease management. It also emphasized the promotion of suitable traditional Chinese medicine techniques and services, the strengthening of community-based integrated medical and eldercare services and the healthy development of the traditional Chinese medicine sector. Additionally, the government has actively promoted the development of internet-based traditional Chinese medicine services. Utilizing internet platforms enables broader coverage and provides safe, effective and convenient traditional Chinese medicine services to more people [22]. In 2018, the National Administration of Traditional Chinese Medicine issued the Guiding Opinions on Promoting the Integration of Traditional Chinese Medicine Health Services and Internet Development, emphasizing the need to strengthen information services for traditional Chinese medicine rehabilitation. It advocates for using cloud computing, big data and mobile internet technologies to promote the establishment of digital rehabilitation systems in traditional Chinese medicine medical institutions, offering specialized traditional Chinese medicine rehabilitation medical and nursing services.

However, current policies predominantly focus on promoting technological advancement and may overlook the challenges faced by the service recipients themselves. The elderly population, who have a higher likelihood of suffering from various diseases and greater needs for traditional Chinese medicine care, often face mobility limitations and other obstacles to in-person visits. Utilizing internet-based medical services can provide an effective solution. Nonetheless, the proportion of elderly individuals who can use internet and information technology remains relatively low [23], and national and local policies lack robust guidelines and standards for the design of internet healthcare products to accommodate the elderly, hindering the promotion and application of internet-based traditional Chinese medicine services. For children, the National Health Commission and the National Administration of Traditional Chinese Medicine issued the Interim Measures for the Management of Internet Diagnosis and Treatment, which require confirmation of the presence of a guardian and relevant professional medical personnel when prescribing internet-based children’s medications for children under 6 years old. However, verifying the presence of the child’s guardian and accurately supervising the situation in real time during the implementation process can be challenging [24], and children may face difficulties in using internet healthcare services due to a lack of familiarity with the process or limited comprehension.

In summary, the ultimate goal of using internet healthcare platforms for traditional Chinese medicine to provide eldercare and child health services is to expand service coverage and accessibility. Thus, future policy design should not only advance technological development, but also address the needs and challenges of service recipients by offering solutions and clear guidelines to enhance the usability and applicability of internet healthcare. This includes promoting internet healthcare products tailored to the habits and needs of vulnerable populations, thereby bringing health and wellbeing to a wider audience.

  1. (3)

       Internet medical security assurance

The thematic group includes two topics, Topic 11: Data security and Topic 13: Network security, focussing on ensuring the security, integrity and availability of patient information while clarifying standards for data collection, storage, transmission and processing.

The 2020 National Health Commission’s “Basic Healthcare and Health Promotion Law of the People’s Republic of China” mandates that medical providers have an obligation to protect citizens’ personal health information and should establish sound information security systems and safeguards. Research indicates that due to the “transparency” of online information, patient data in internet healthcare is exposed to multiple parties, including medical institutions, healthcare professionals and internet companies, significantly increasing the risk of data breaches [25]. In response, in 2022, the National Health Commission and several other departments jointly issued the Measures for the Network Security Management of Medical and Health Institutions, which emphasized the need to protect data confidentiality, integrity and availability through data encryption and backup technologies. These measures strengthen the security protection across the entire data lifecycle, including collection, transmission, storage, usage, exchange and destruction.

Despite existing policies that strictly regulate how internet healthcare institutions handle patient information and the adoption of privacy policies by various internet healthcare platforms for the collection, processing and use of patient data, research shows that privacy policies on different platforms vary greatly in quality. Issues related to reducing user cognitive load and ensuring information security persist, making it difficult for users to understand policies and leading to potential misuse of information without users’ knowledge [26]. Therefore, future policy direction should focus on enhancing the standardization of internet healthcare platforms’ privacy policies. Additionally, efforts should be made to educate patients about personal information safety, encouraging individuals to actively participate in strengthening self-regulation to ensure proper use and protection of patient privacy [27].

  1. (4)

      Internet medical health education and popularization of science

The thematic group comprises three topics: Topic 18: Health education, Topic 25: Oncology diagnosis and treatment and Topic 26: Ophthalmology and eye diseases. These topics focus on effectively using internet technology and platforms to promote the dissemination of medical knowledge and health education, thereby enhancing the public’s health literacy and understanding of medical information.

Internet healthcare plays a crucial role in health education and promotion. The 2018 State Council’s Guiding Opinions on Promoting the Development of “Internet + Medical Health” emphasized the need to strengthen “Internet + medical education and public service”. For patient groups, health education can improve understanding of diseases and treatment importance, leading to higher health knowledge levels and better patient habits, ultimately enhancing treatment outcomes [28]. For the general public, the internet can effectively leverage expert resources, facilitating the dissemination of popular science information and increasing health knowledge awareness to promote healthy behaviour [29]. However, the thematic group consists of topics numbered 18, 25 and 26, indicating a lower priority and a lack of emphasis in relevant policy documents. Additionally, the limited time doctors spend on online consultations in internet healthcare affects the quality of health education content [30].

Thus, there is a need to strengthen the importance of health education and promotion in internet healthcare and to clarify support for these areas. Encouraging and guiding relevant institutions and individuals to participate actively is essential. Furthermore, it is crucial to control the quality of information shared on platforms to prevent the spread of false or misleading information. Cross-sector collaboration should be promoted, inviting healthcare, education and technology sectors to work together on internet healthcare education and promotion, driving its progress to a higher level.

  1. (5)

      Internet-based guidance for treatment and rehabilitation of infectious diseases

The thematic group encompasses two topics: Topic 8: Epidemic prevention and control and Topic 10: Patient treatment. These topics focus on leveraging internet technology to enhance the efficiency and effectiveness of infectious disease prevention efforts while also providing patients with comprehensive and personalized rehabilitation guidance.

During the ongoing COVID-19 pandemic, internet healthcare has rapidly expanded to alleviate the strain on in-person hospital treatment, reduce crowding within hospitals and lower the risk of cross-infection [31]. In 2020, the National Health Commission issued several documents, such as the Notice on Strengthening Information Technology Support for COVID-19 Epidemic Prevention and Control, which emphasized the importance of promoting online medical services. Additionally, the National Health Commission’s Guiding Opinions on Implementing “Internet + Healthcare” Medical Insurance Services during the COVID-19 Pandemic incorporated eligible “Internet + ” healthcare services into medical insurance payment scopes, helping to reduce the financial burden for patients using online healthcare services.

With the shift in COVID-19 prevention strategies and the increase in home-treated patients, the advantages of internet healthcare have become even more pronounced. In 2022, the State Council issued the Notice on Providing COVID-19 Internet Healthcare Services, which noted that medical institutions could use online consultation platforms to prescribe treatment for patients exhibiting COVID-19-related symptoms who are staying home according to the Guidelines for Home Treatment of COVID-19. In the post-pandemic era, contagious diseases such as influenza and bronchial pneumonia may continue to intermittently surge. Utilizing the internet to provide medical services remains crucial for controlling and treating infectious diseases and alleviating pressure on hospitals [32], and it is expected to remain a key focus of future policies.

Conclusions

In this study, we examine “Internet healthcare policy” by analysing 436 policy documents issued by central and local governments under the Healthy China initiative. Using the BERTopic model, the study systematically analyses the thematic content and structure of internet healthcare policies under the Healthy China initiative, resulting in 27 topics that were grouped into 5 clusters. Each cluster is thoroughly explained and analysed, offering suggestions for future policy directions. However, the study has certain limitations. First, the analysis focusses primarily on the content of policy texts and does not incorporate empirical research with specific case studies or data. Second, the data are limited to domestic policy texts and do not include a comparative analysis with foreign policies. Future research will incorporate practical case studies or data to evaluate and analyse policy implementation effects and compare policy issuance and enforcement in different countries and regions to provide more targeted and actionable policy recommendations.

Availability of data and materials

Datasets used or analysed during the current study are available from the corresponding author on reasonable request.

References

  1. Wang XY, Xie Y, Yang XJ, Gu DX. Internet-based healthcare knowledge service for improvement of Chinese medicine healthcare service quality. Healthcare. 2023;11(15): 2170. https://doi.org/10.3390/healthcare11152170.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Yang F, Shu H, Zhang X. Understanding “internet plus healthcare” in china: policy text analysis. J Med Internet Res. 2021. https://doi.org/10.2196/23779.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Thorlu-Bangura Z, Poole L, Sood H, Khan N, Stevenson F, Khunti K, Gill P, Sajid M, Hanif W, Bhala N, Modha S, Patel K, Blandford A, Banerjee A, Ramasawmy M. Digital health, cardiometabolic disease and ethnicity: an analysis of United Kingdom government policies from 2010 to 2022. J Public Health Policy. 2023;44(2):179–95. https://doi.org/10.1057/s41271-023-00410-z.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Wei Q, Wang XY, Zhang GR, Li XG, Yang XJ, Gu DX. Internet healthcare policy analysis, evaluation, and improvement path: multidimensional perspectives. Healthcare. 2023;11(13): 1905. https://doi.org/10.3390/healthcare11131905.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Allen P, Pilar M, Walsh-Bailey C, Hooley C, Mazzucca S, Lewis CC, Mettert KD, Dorsey CN, Purtle J, Kepper MM, Baumann AA, Brownson RC. Quantitative measures of health policy implementation determinants and outcomes: a systematic review. Implement Sci. 2020;15(1): 47. https://doi.org/10.1186/s13012-020-01007-w.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Sheng B, Wang ZY, Qiao YJ, Xie SQ, Tao J, Duan CQ. Detecting latent topics and trends of digital twins in healthcare: a structural topic model-based systematic review. Digit Health. 2023;9: 20552076231203672. https://doi.org/10.1177/20552076231203672.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Zhou H, Zhang J, Zhao Y. Wind power policies in China: development themes, evaluation, and recommendations. Energy Sci Eng. 2022;11(3):1044–59. https://doi.org/10.1002/ese3.1368.

    Article  Google Scholar 

  8. Ma CW, Qirui C. Spatial-temporal evolution pattern and optimization path of family education policy: an LDA thematic model approach. Heliyon. 2023;9(7): e17460. https://doi.org/10.1016/j.heliyon.2023.e17460.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Song C, Guo J, Gholizadeh F, Zhuang J. Quantitative analysis of food safety policy—based on text mining methods. 2022. Foods. https://doi.org/10.3390/foods11213421.

  10. Grootendorst M. BERTopic: neural topic modeling with a class-based TF-IDF procedure. 2022.

  11. Lee H, Hong Lee S, Re Lee K, Hyun Kim J. ESG discourse analysis through BERTopic: comparing news articles and academic papers. Computers Mater Contin. 2023;75(3):6023–37. https://doi.org/10.32604/cmc.2023.039104.

    Article  Google Scholar 

  12. Uncovska M, Freitag B, Meister S, Fehring L. Rating analysis and BERTopic modeling of consumer versus regulated mHealth app reviews in Germany. NPJ Digit Med. 2023. https://doi.org/10.1038/s41746-023-00862-3.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Lalk C, Steinbrenner T, Kania W, Popko A, Wester R, Schaffrath J, Eberhardt S, Schwartz B, Lutz W, Rubel J. Measuring alliance and symptom severity in psychotherapy transcripts using bert topic modeling. Adm Policy Ment Health Ment Health Serv Res. 2024. https://doi.org/10.1007/s10488-024-01356-4.

    Article  Google Scholar 

  14. Wang ZY, Chen J, Chen JP, Chen HH. Identifying interdisciplinary topics and their evolution based on BERTopic. Scientometrics. 2023. https://doi.org/10.1007/s11192-023-04776-5.

    Article  Google Scholar 

  15. Chen L, Zhong WH, Ma YD, Zhang YJ, Cai GC, Huang HQ, Lu FF, Su XD. Changes of wound area and inflammatory factors in diabetic foot patients after comprehensive nursing of traditional Chinese medicine. Eur Rev Med Pharmacol Sci. 2023;27(4):1609–13.

    CAS  PubMed  Google Scholar 

  16. Rehman MU, Shafique A, Usman AB. Securing medical information transmission between IoT devices: an innovative hybrid encryption scheme based on quantum walk, DNA encoding, and chaos. Internet Things. 2023;24: 100891. https://doi.org/10.1016/j.iot.2023.100891.

    Article  Google Scholar 

  17. Xie XX, Lin LY, Fan S, Zhou WM, Lin F, Wang L, Guo TJ, Ma X, He Y, Chen YX. Internet hospital in China: a cross-sectional survey. Lancet. 2017;390:S40–S40. https://doi.org/10.1016/S0140-6736(17)33178-1.

    Article  Google Scholar 

  18. Xie XX, Zhou WM, Lin LY, Fan S, Lin F, Wang L, Guo TJ, Ma CY, Zhang JK, He Y, Chen YX. Internet hospitals in China: cross-sectional survey. J Med Internet Res. 2017;19(7): e239. https://doi.org/10.2196/jmir.7854.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Sang L, Song L. The current status of the use of internet hospitals for outpatients with pain: retrospective study. J Med Internet Res. 2023;25: e44759. https://doi.org/10.2196/44759.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Chen XJ, Wu XX, Zhang QH, Jing R, Cheng WB, Tian JZ, Jin CX. The construction and operational models of internet hospitals in China: a hospital-based survey study. BMC Health Serv Res. 2023;23(1): 669. https://doi.org/10.1186/s12913-023-09675-2.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Singh S, Kumar D. A public key authentication and privacy preserving model for securing healthcare system. IETE J Res. 2023;69(8):5031–43. https://doi.org/10.1080/03772063.2021.1977189.

    Article  Google Scholar 

  22. Yang B, Wen Q, Zhang Y, Wang XL, Yin XD, Li QB, Li QL, Song L. Economic value and characteristics of cloud pharmacy for children based on internet hospital in western China during the COVID-19 pandemic: cross-sectional survey study. Front Public Health. 2022;10: 1034450. https://doi.org/10.3389/fpubh.2022.1034450.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Landgren S, Cajander Å. Non-use of digital health consultations among Swedish elderly living in the countryside. Front Public Health. 2021;9: 588583. https://doi.org/10.3389/fpubh.2021.588583.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Cui WB, Zhu WJ, Li XJ, Wu DM, He P, Yu GJ. Attitudes and perspectives of 534 Chinese pediatricians toward internet hospitals. Front Pediatr. 2022;10: 948788. https://doi.org/10.3389/fped.2022.948788.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Yu F, Shen H, Yu QL, Kong XX, Sharma PK, Cai S. Privacy protection of medical data based on multi-scroll memristive hopfield neural network. IEEE Trans Netw Sci Eng. 2023;10(2):845–58. https://doi.org/10.1109/TNSE.2022.3223930.

    Article  Google Scholar 

  26. Robillard JM, Feng TL, Sporn AB, Lai JA, Lo C, Ta M, Nadler R. Availability, readability, and content of privacy policies and terms of agreements of mental health apps. Internet Interv. 2019;17: 100243. https://doi.org/10.1016/j.invent.2019.100243.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Yi GH, Cao ZC. An algorithm of image encryption based on AES & Rossler hyperchaotic modeling. Mob Netw Appl. 2023. https://doi.org/10.1007/s11036-023-02216-5.

    Article  Google Scholar 

  28. Zou T, Guo P, Li F, Wu Q. Research topic identification and trend prediction of China’s energy policy: a combined LDA-ARIMA approach. Renew Energy. 2024. https://doi.org/10.1016/j.renene.2023.119619.

    Article  Google Scholar 

  29. Sedaghat S. Early applications of ChatGPT in medical practice, education and research. Clin Med. 2023;23(3):278–9. https://doi.org/10.7861/clinmed.2023-0078.

    Article  Google Scholar 

  30. Liu LY, Shi LZ. Chinese patients’ intention to use different types of internet hospitals: cross-sectional study on virtual visits. J Med Internet Res. 2021;23(8): e25978. https://doi.org/10.2196/25978.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Sadique MA, Yadav S, Khan R, Srivastava AK. Engineered two-dimensional nanomaterials based diagnostics integrated with internet of medical things (IoMT) for COVID-19. Chem Soc Rev. 2024;53(8):3774–828. https://doi.org/10.1039/d3cs00719g.

    Article  CAS  PubMed  Google Scholar 

  32. Kumar D, Sood SK, Rawat KS. IoT-enabled technologies for controlling COVID-19 spread: a scientometric analysis using CiteSpace. Internet Things. 2023;23: 100863. https://doi.org/10.1016/j.iot.2023.100863.

    Article  Google Scholar 

Download references

Funding

This study was supported by the State Administration of Traditional Chinese Medicine of China under cooperative agreement GZY-FJS-2022-045. The information, conclusions and opinions expressed in this brief are those of the authors.

Author information

Authors and Affiliations

  1. Management School of Beijing University of Chinese Medicine, Beijing, China

    Guoping Wu, Junbai Chen & Zhiwei Wang

  2. School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, China

    Youshu Yuan

  3. Institute of Information of Traditional Chinese Medicine, China Academy of Chinese Medical Sciences, Beijing, China

    Guoxiang Li & Shuang Liu

  4. National Institute of Chinese Medicine Development and Strategy, Beijing University of Chinese Medicine, Beijing, China

    Hao Ning & Zhiwei Wang

Authors
  1. Guoping Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Hao Ning
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Youshu Yuan
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Junbai Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Guoxiang Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Shuang Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Zhiwei Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

G.W., H.N. and J.C. participated in data collection and collation. G.W. and J.C. participated in the method design, analysed data and drafted the initial manuscript. G.W., J.C., Y.Y., G.L., S.L. and Z.W. participated in text checking correction and helped to draft the manuscript. Z.W. and Y.Y. oversaw and provided input on all aspects of manuscript writing and the final analytical plan. All the authors read and approved the final manuscript.

Corresponding author

Correspondence to Zhiwei Wang.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wu, G., Ning, H., Yuan, Y. et al. Topic identification and content analysis of internet medical policies under the background of Healthy China 2030. Health Res Policy Sys 22, 132 (2024). https://doi.org/10.1186/s12961-024-01226-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12961-024-01226-3

Keywords

Health Research Policy and Systems

ISSN: 1478-4505

Contact us