The Impact of the Digital Economy on the Health Industry from the Perspective of Threshold and Intermediary Effects: Evidence from China

Abstract

This study explores the effect of the digital economy driving the development of the health industry and the mechanism behind it. Based on the panel data of 27 Chinese provinces from 2014–2021, this paper uses the entropy weight method to construct a comprehensive indicator evaluation system for the health industry, digital economy, and technological innovation. The two-way fixed effects model and panel threshold model are used to explore the impact of the digital economy on the health industry, and the intermediary effects model is used to analyze the mechanism role of technological innovation in the impact of the digital economy on the health industry. The results show that the digital economy can drive the development of the health industry. The driving effect shows obvious regional heterogeneity, with the strongest in the west, the second strongest in the central part, and the weakest in the east. This driving effect also has non-linear characteristics. Improving technological innovation is an important mechanism for the development of the health industry driven by the digital economy. This study promotes the exploration of the construction path of “Healthy China”, and reflects the importance of implementing dynamic and differentiated digital economy strategies and increasing the R&D of core technologies to drive the development of the health industry.

1. Introduction

Health is a basic need of the people and an important foundation for economic and social development. In developed economies such as the United States, Europe, and Japan, the health industry has become a huge driving force for the growth of the entire national economy, and the added value of the health industry in these economies has accounted for more than 15% of their GDP [1]. Compared with developed countries, the development of China’s health industry is still at a relatively backward level. The health industry’s pulling effect on the modern service industry and even the whole economy is relatively slow in the long run. The overall backwardness of China’s health industry will not be able to meet the needs of the aging population and the rising demand for health products and services with rising income. Total healthcare spending refers to the total amount of money consumed by society for healthcare services. The concept is internationally comparable. It is often used to assess the market size of the health industry in a narrow sense. The China Health Care Statistical Yearbook [2] shows that China’s total healthcare expenditure in 2019 was 6.58 trillion yuan, accounting for 5.4% of GDP, of which government and personal health expenditure accounted for 56% and 44%, respectively. In 2019, the proportion of total healthcare expenditure to GDP was about 16.8% in the United States, 10.2% in the United Kingdom, 10.7% in Japan, and 10.8% in Canada.

The 20th Party Congress report pointed out that people’s health is an important symbol of national prosperity and national strength. The protection of people’s health should be given a strategic position of priority development. In October 2016, the State Council (PRC) issued the “Health China 2030” Planning Outline (hereinafter referred to as “the Outline”), which pointed out that the development of the health industry should be the focus of promoting the construction of “Health China” [3]. In April 2022, the General Office of the State Council issued the “14th Five-Year Plan for National Health”, which clearly pointed out that the health industry should be optimized and strengthened. The integration and development of health-related industries should be promoted, and new health industries and new models should be developed [4].

In addition, the outbreak of the COVID-19 pandemic brought unprecedented challenges and opportunities to the health industry. On the one hand, the outbreak was a test of the level and carrying capacity of healthcare in the country. On the other hand, China’s anti-epidemic measures have been effective. People are confident in the construction and development of China’s healthcare system. The national demand for health products has also increased dramatically [5].

At present, research on the health industry in the academic field mainly focuses on the connotation of the definition of the health industry, statistical caliber, and regional differences. There are two ways of dividing the health industry in a narrow sense and a broad sense [6]. The health industry in the narrow sense includes the integration of the sectors of the economic system that provide services such as prevention, treatment and rehabilitation to patients. In China, health in the narrow sense corresponds to the medical and health service industry. The health industry in the broad sense includes not only the healthcare service industry in the narrow sense, but also the economic activities that provide healthcare products and healthcare services for healthy people. He et al. [7] classified the health industry from the perspective of industry division and the upstream and downstream of the industry, as well as the type of service. In April 2019, the National Bureau of Statistics released the Health Industry Statistical Classification (2019) to the public [8]. It defines the health industry as a collection of production activities based on healthcare, biotechnology, and life sciences to maintain, improve, and promote the health of the people and provide the public with products (goods and services) directly or closely related to health. At the same time, the scope of the health industry is defined by 13 major categories, which are medical and health services; health affairs; health environment management, scientific research, and technical services; health personnel education and health knowledge popularization; health promotion services; health protection and financial services; intelligent health technology services; pharmaceutical and other health products distribution services; other health-related services; pharmaceutical manufacturing; medical instruments, equipment, and apparatus manufacturing; health supplies; equipment and intelligent equipment manufacturing; medical and health institution facilities construction; Chinese herbal medicine planting; and breeding and collection.

Donzé et al. [9] mentioned in the introduction to a Special Issue of The Health Industry in the Twentieth Century that health had become a fast-growing sector of the global economy, the basis of the welfare system, and one of the main reasons for the progress made in the global human development index in the 20th century. However, little was known about the conditions that had shaped this situation. Shu et al. [10] found that the impact of the healthcare industry clustering on health was positive but varied between regions. Pesqueira et al. [11] studied the important aspects of big data in the development of expertise and process-oriented skills, and their impact on organizational business processes. Simonetto et al. [12] analyzed the contribution of PAR technology to workers’ health status in the era of Industry 4.0. Li et al. [13] used principal component analysis to construct an indicator system for measuring the development level of China’s health industry. They studied the spatial distribution characteristics of the development level of China’s health industry based on the method of spatial exploratory data analysis (ESDA). Dong et al. [14] analyzed the advantages and limitations of the development of health industry in Beijing, Tianjin, and Hebei based on the connotation and characteristics of the health industry. They proposed countermeasures to promote the collaborative development of the health industry. Yang [15] measured the level of health construction in China by using the entropy method, and examined the regional differences and degree of polarization through the Dagum Gini coefficient and polarization index. Zheng et al. [16] found that pilot policies for healthy cities contribute significantly to the upgrading of industrial structure, especially for cities in the eastern and central regions. Thus, government departments should promote the transformation of the health industry in different regions according to local conditions.

In the context of the strategy of accelerating the construction of Digital China, a number of support initiatives have been introduced for digital key application areas such as “Internet + Health” and healthcare big data, creating a good endogenous environment for digital health development. According to the China Digital Economy Development Report released by the China Academy of Information and Communications Technology, China’s digital economy reached 45.5 trillion yuan in 2021, accounting for 39.8% of GDP [17]. The development of China’s digital economy has reached a world-leading level, and there is still great potential in the field of digital health. The Statistical Classification of the Digital Economy and its Core Industries (2021) defines the digital economy as a series of economic activities that use data resources as a key production factor, modern information networks as an important carrier, and the effective use of information and communication technologies as an important driving force for efficiency improvement and economic structure optimization [18]. In order to further promote the development of China’s digital economy, the state has issued policies such as the State Council’s Guidance on Actions to Actively Promote the “Internet+” and the Outline of the National Strategy for the Development of Informatization to provide support at the strategic level and emphasize the promotion of the deep integration of the digital economy and the real economy. Jiang et al. [19] found a positive correlation between digitization and health outcomes in BRICS economies. Xu et al. [20] showed that the digital economy was able to maintain the stability of China’s economy during the new coronavirus epidemic. The development of the digital economy has not only contributed to the country’s economic growth, but has also had a significant impact on public health. Lyu et al. [21] noted that the development of China’s digital economy has helped to optimize the efficiency of public health service delivery and enhance the ability to respond to public health crises, particularly in the context of the COVID-19 outbreak. Song et al. [22] pointed out that the digital economy acts on the development of the health industry through the level of education, which plays the role of a full intermediary effect. Qin [23] analyzed the development status of the digital economy and the health industry on a theoretical level. He discussed the importance of the digital economy for the development of the health industry and the risks it faces.

In November 2022, the Ministry of Science and Technology and the National Health and Wellness Commission issued the “14th Five-Year Plan for Health and Health Science and Technology Innovation” which pointed out that health technological innovation is urgently needed to protect people’s health and is a core driving force for the development of the health industry [24]. In the era of digital economy, a new generation of information technology, represented by artificial intelligence, big data, 5G, and industrial internet, has flourished, providing a source of power for technological innovation and industrial development in the health industry. For example, the high-tech sector should increase the development of cutting-edge medical technologies such as gene sequencing, and the use of blockchain technology to guarantee the security of data in the healthcare sector [25]. These initiatives can drive the high-quality development of the health industry. At the same time, technological innovation has a two-way impact on economic growth and human resource welfare [26]. Health enterprises can strengthen their core competitiveness by improving the level of technological innovation, thus supporting the health industry in realizing better economic benefits.

From the previously mentioned literature, it was found that some progress has been made in the construction of indicator systems and scale measurement for the development of the health industry, but some issues need to be further explored. Firstly, the existing studies mainly focus on the theoretical analysis of the impact of the digital economy on the development of the health industry, with less empirical research involved. There is insufficient empirical evidence that the digital economy promotes the development of the health industry. Second, there is insufficient research on the mechanism of action of the digital economy to enable the development of the health industry. There has been no in-depth research conducted on the relationship between the digital economy, technological innovation, and the health industry and the mechanism of action. So, is the digital economy able to drive the development of the health industry? How can the digital economy drive the development of the health industry? In the context of the digital economy, the answers to these questions are of great theoretical and practical significance. Theoretically, the answers to these questions not only explore more development paths for the health industry, but also provide more development ideas for related enterprises in the health field. Practically, these answers can help the government to analyze the current situation of health industry development. Relevant departments can improve the weak links, so as to enhance the quality of the development of China’s health industry. Therefore, this paper attempts to empirically analyze the effects and mechanisms of the digital economy on the health industry in depth based on a complete framework. Specifically, this study uses the entropy weight method to construct a comprehensive index evaluation system to measure the development levels of the digital economy, technological innovation, and health industry in 27 provinces (cities and districts) in China from 2014 to 2021. This paper empirically analyzes the relationship between the digital economy, technological innovation, and health industry using a two-way fixed effect model, threshold effect and intermediary effect. The results show that the digital economy significantly promotes the development of the health industry and reveals a non-linear characteristic. Meanwhile, technological innovation is one of its influencing mechanisms. These findings still hold after robustness testing. By comparing with previous studies and taking into account the realistic context, specific policy recommendations are proposed to make a useful addition to the existing studies.

This paper is structured as follows: Section 2 organizes and collates the relevant literature on the health industry, digital economy, and technological innovation, further specifying the marginal contributions of the paper; Section 3 describes the variables, data, and models; Section 4 analyzes and discusses the empirical results and reports on robustness tests; and Section 5 summarizes the findings, makes relevant policy recommendations, and points to future research directions.

2. Literature Review

2.1. Digital Economy and the Health Industry

The concept of the digital economy was first introduced by Tapscott [27]. Research on the digital economy in China dates back to 1999, and the literature focusing on digital economy research has risen significantly in the last five years [28]. As a rapidly developing new economic model, the digital economy integrates information and communication technologies in industrial development and promotes the digital transformation of all factors in the industry. It plays a powerful role in promoting national economic development [23]. According to the “Electricity and Numbers” database of the Internet Society, the market size of the Internet healthcare market in 2020 was 155 billion yuan, an increase of 48.46% year-on-year; the user size was 662 million people, an increase of 42.06% year-on-year. New ICT-based health services have become a new growth point for the health industry [5].

The core chain of the health industry chain consists of the enterprise chain, value chain, technology chain, product chain, and spatial chain [29]. At present, the development of the health industry is at a preliminary stage. The resources of each link of the industry chain are relatively scattered. This leads to inadequate resource utilization and low production efficiency in the entire industrial chain, preventing it from taking advantage of resources and industrial agglomeration effects. The digital economy can provide a series of value-added services to the industry chain by reconfiguring and integrating digital production factors [30]. Thus, the digital economy can promote the construction and upgrading of the health industry chain.

There is a certain degree of deficiency in the health industry in terms of talent supply, financial support, and information sharing. How can the government play a leading role in encouraging various resources to enter the health industry? At the same time, how can the market play a role in improving the supply of various factors of production? These are all important factors affecting the development of the health industry. Under the digital economy model, the accelerated development of Internet B2B platforms can promote the construction of an Internet platform for the health industry. It effectively removes complex and redundant links in the supply chain, continuously promotes the flattening of the industry, and improves the efficiency of industrial operations [31]. Thus, the digital economy can promote the health industry as a whole to achieve supply chain transformation and upgrading.

With high-quality economic development, rising income levels, and demographic changes, the upgrading of the demand structure of the health industry has given rise to huge market space. People are no longer satisfied with basic medical needs, but have instead upgraded their demand for health management services, high-end medical care, healthy food, and other high-level health needs. In recent years, some domestic Internet health management companies have developed rapidly, and the application software platforms they have built provide health management services to hundreds of millions of users through live fitness streaming, food recommendations, and activity participation [32]. As a result, the digital economy can further give rise to new business models in the health industry.

China has long attached a high level of importance to the development of the health industry. It can boost consumption, promote employment, and has broad market potential. With the continuous improvement of living standards, people’s demand within the health industry further expands, and the health industry ushers in good development opportunities [33]. COVID-19 has had a major impact on the residential industry, negatively affecting many aspects of life [34]. COVID-19 has accelerated the adoption of digital technological innovations in healthcare to support the healthcare system [35]. Without exception, COVID-19 has increased global attention to the healthcare field and the desire for better health services.

The concept of “digital health” was born under the joint strategic deployment of building a healthy China and a digital China. “Digital health” was first mentioned in the 14th Five-Year Plan for the Development of Digital Economy. It was clearly stated that “accelerating the development of digital health services” was one of the “social service digitalization enhancement projects”. The development of digital health has roughly gone through four periods: the budding period, the exploration period, the growth period, and the comprehensive development period. Unlike the vigorous development of digital production and digital life, the development of digital health is relatively slow, and the digitalization level of the health industry is relatively backward [36]. In the context of big data, the rapid growth of the economic scale of the health industry has triggered new problems. Although it has brought economic growth, employment growth, and social and medical service innovation, problems in cybersecurity and personal data leakage have also come to the fore [37]. At the same time, the development of the digital economy may challenge the technological base and resource capacity of health enterprises. The digital economy brings problems of information quality, information security, and information overload to enterprises, which affects their scientific decision making, increases their information processing costs, and increases their operational burden [38]. All of these may weaken its driving effect on the development of the health industry.

It is now generally agreed in the academic community that the innovative integration of digital economy and health industry governance will become an inevitable trend. However, there are very few studies on how the digital economy impacts the health industry, what characteristics this impact has, and how the power of the digital economy can be brought into play to help the development of the health industry. Therefore, this paper investigates what kind of relationship exists between the digital economy and the health industry, and explores the influence mechanism between them. Hypothesis 1.1. and Hypothesis 1.2. are proposed:

Hypothesis 1.1.

The digital economy can drive the development of the health industry.

Hypothesis 1.2.

The driving influence of the digital economy on the development of the health industry is non-linear.

2.2. Digital Economy, Technological Innovation, and the Health Industry

Currently, most of the literature related to the digital economy focuses on the calculation and evolution of digital economy indices [39,40]. The impact of digital economy development on various aspects of society has been studied from different perspectives through the measurement of the digital economy index system. For example, in a study on digital economy and innovation efficiency, Huang et al. [41] established a digital economy index using city-level data in China, showing that China’s digital economy and urban innovation have spatial clustering characteristics and regional differences in spatial distribution. Wang et al. [42] used spatial econometrics to obtain a significant positive direct effect and spatial spillover effect of the digital economy on the growth of innovation efficiency. In a study on digital economy and industry, Chen et al. [28] concluded that the digital economy era has gradually formed a new generation of industrial models based on digital technology, prompting traditional industries to continuously develop transformation paths that create new values with the help of new technologies. Qin [23] explored the importance of the digital economy for the development of the health industry and the challenges it faces.

At present, research on technological innovation focuses on the measurement of the efficiency of technological innovation. Through the measurement of technological innovation efficiency, the impact of technological innovation on the economy and society is studied from different perspectives. For example, Zhu [43] constructed a scientific and technological innovation input–output index system to measure the efficiency of national scientific and technological innovation and made international comparisons. Wu [44] pointed out that the level of technological innovation has a significant threshold effect on the optimization of industrial structure. Qin [45] pointed out that innovation capability plays an important role in the cultivation of core competitiveness of the health industry. Jiang et al. [46] made suggestions for the innovation development of China’s health industry by studying the Swedish health industry’s scientific and technological innovation system. Shi et al. [47] studied the effects of innovation-driven policies on industrial pollution reduction, providing an empirical basis for the promotion of public health.

Currently, the literature on the health industry is generally small and mainly focuses on the definition, division criteria, and scale measurement of the health industry. Empirical studies on the relationship between the digital economy, technological innovation, and the health industry have not yet been conducted. The lack of innovation capacity is still a key issue limiting the development of China’s health industry. Most manufacturers are small in scale and weak in R&D, mainly concentrating on the middle and low-end markets, with few internationally competitive health industries [48]. The core driver of industrial development lies in innovation, so the development of the health industry needs to build an innovation ecosystem and achieve a synergistic symbiosis of industry innovation [49]. In the new journey of Chinese-style modernization, we need to embrace the “new economy” represented by the digital economy, so that we can better use technological innovation to promote the specialization and high level of the health industry. Therefore, in this paper, in addition to exploring the direct impact of the digital economy, we also consider the possible indirect impact of the digital economy on the health industry through technological innovation. This leads to Hypothesis 2:

Hypothesis 2.

The digital economy can contribute to the development of the health industry through technological innovation.

Table 1. Summary of health industry, digital economy, and technological innovation studies.

Objectives of Study	Authors (Year)	Findings
Digital economy and health industry	Tapscott (1996) [27]	The concept of the digital economy was first introduced.
	Liu et al. (2020) [40] Yang et al. (2022) [15]	These scholars measured the level of digital economy development from multiple perspectives.
	He (2019) [31] Liu et al. (2021) [30] Zhang et al. (2022) [5] Wei et al. (2022) [32] Qin (2023) [23]	The digital economy promotes the transformation and upgrading of the overall supply chain of the health industry.
	Zhang et al. (2022) [33] Cutcu et al. (2023) [34] Park (2022) [35]	COVID-19 reinforces global health concerns.
	Zhang (2020) [37] Cheng et al. (2022) [38] Lin (2023) [36]	In the context of the digital economy, new issues have emerged in the health industry alongside the rapid growth in economic scale.
Digital economy and technological innovation	Huang et al. (2022) [41]	China’s digital economy and urban innovation are characterized by spatial agglomeration.
	Wang et al. (2022) [42]	The digital economy can directly contribute to increased efficiency in innovation.
Technological innovation and health industry	Fan et al. (2017) [48] Zhang et al. (2018) [6] Lyu et al. (2022) [21]	The innovation efficiency of the health industry is not high.
	Jiang et al. (2019) [46] Qin et al. (2022) [45] Shi et al. (2023) [47]	Innovation can enhance the competitiveness of the health industry.

presents the current literature related to the digital economy, health industry, and technological innovation. Figure 1 is a graphical presentation of the three hypotheses. Through the review of the health industry literature, it was found that most of the studies involve the scale measurement of the health industry and the theoretical analysis of the influencing factors of health industry development. In this context, the possible marginal contributions of this paper lie in three aspects. First, this paper uses the entropy method to construct a comprehensive index evaluation system for the digital economy, health industry, and technological innovation, and conducts a preliminary investigation into the measurement methods of the three. Second, this paper links the digital economy and the health industry, and quantitatively investigates how they are related to each other to complement the empirical studies in this area. Third, this paper further explores the relationship between the digital economy, technological innovation, and the health industry, which not only enriches the existing theory but also fills the gap of empirical research on the health industry.

3. Variables, Data, and Methods

3.1. Variables

3.1.1. Measurement of Variables

The comprehensive evaluation methods mainly include two types of subjective evaluation methods and objective evaluation methods. The entropy method is one of the objective evaluation methods, which can make full use of the objective message included in the data to assign weights to each index. In order to avoid the impact caused by subjective factors, we measured the development level of the health industry, digital economy, and technological innovation using the entropy method, and the specific steps are as follows.

Step1: Data standardization.

Let $V_{ij}$ be the original data of the jth indicator in the ith evaluation object (i = 1, 2, 3,…, n; j = 1, 2, 3,…,m). In order to make the data of different calibers comparable and eliminate the difference of the dimension between the indicator data, the original data are standardized; if the indicator is a positive indicator, then the processing formula can be written as follows:

$$x_{ij}=\frac{v_{ij}-\min \left(v_j\right)}{\max \left(v_j\right)-\min \left(v_j\right)}$$

(1)

If the indicator is negative, the formula can be written as follows:

$$x_{ij}=\frac{{\max (v}_j)-v_{ij}}{\max \left(v_j\right)-\min \left(v_j\right)}$$

(2)

$V_{ij}$(i = 1, 2, 3,…, n; j = 1, 2, 3,…, m) is the raw data of the jth index of the ith evaluation object, and is the dimensionless value after standardization.

Step2: Calculate the entropy value and weight.

Let $y_{ij}$ be the weight of the jth indicator in the ith evaluation object, $e_j$ be the entropy value of the jth indicator, and $g_j$ be the coefficient of variation in the jth indicator, where the number of evaluation objects is the weight of the first evaluation indicator (j = 1, 2, 3,…, m). Each coefficient is calculated by the following formulas.

$$y_{ij}=\frac{x_{ij}}{\sum _{n=1}^n{x}_{ij}}$$

(3)

$$e_j=-\frac{1}{\mathrm{l}\mathrm{n}\left(n\right)}{\sum }_{i=1}^n{y}_{ij}\mathrm{l}\mathrm{n}{y}_{ij}$$

(4)

$$g_j=1-e_j$$

(5)

$$w_{ij}=\frac{g_j}{\sum _{j=1}^m{g}_j}$$

(6)

Step3: Calculate the score.

According to the weights of each indicator calculated with Equation (6) and the data of each indicator after standardization, the level of health economy, digital economy, and technological innovation development can be calculated with Equation (7), where Score represents the total score, and $w_j$ and $x_{ij}$ represent the weight and standardized value, respectively.

$$Score={\sum }_{j=1}^m{w}_j{x}_j$$

(7)

3.1.2. Explained Variable: Health Industry (HI)

Most of the current academic measurements on the development of the health industry use the classification criteria in the Statistical Classification of Health Industries (2019), which divides the health industry into 13 broad categories [8]. Recent studies on the development of the health industry include Qin [23], Dong [14], Li et al. [13], and others. Based on the Outline and previous research methods, this paper measures the health industry based on the availability of data. The level of development of China’s health industry is examined comprehensively in five dimensions: medical and health services, management of healthy things and healthy environment, health promotion services, health protection and financial services, and pharmaceutical manufacturing. The entropy weight method is used to construct a comprehensive evaluation index system for the development of the health industry (see

Table 2. Health industry development level evaluation index system.

Indicators	Measurement Method	Unit	Weight
Health Services	Beds of Health Institutions	10,000 beds	0.04
	Health Technical Personnel	person	0.039
	Healthcare Institutions	unit	0.053
	Total Health Expenditure	10,000 yuan	0.044
	Per Capita Health Expenditure	yuan	0.237
	Occupancy Rate of Hospital Beds	%	0.007
	Beds of Healthcare Institutions per 1000 Population	bed	0.025
	Tertiary Care Hospitals	unit	0.027
Health Things and Environment Management	Investment Completed in the Treatment of Industrial Pollution	10,000 yuan	0.079
	Rate of Domestic Garbage Harmless Treatment	%	0.005
	Urban Sewage Treatment Rate	%	0.005
Health Promotion Services	Elderly Care Beds per 1000 Elderly Population	bed	0.022
	Number of Aged Care Institutions	unit	0.048
	Tourism Workers	person	0.054
	Travel Agencies	household	0.044
Health Protection and Financial Service	Life Insurance Income	100 million yuan	0.06
	Life Insurance Expenses	100 million yuan	0.052
Pharmaceutical manufacturing	Chemical Medicines Output	10,000 tons	0.16

Note: All indicators are positively attributed to the health industry.

).

In the section on health services, the level of basic health services is measured comprehensively in terms of the number of beds in health institutions, the number of health technicians, the number of health institutions, the bed occupancy rate, the number of beds per 1000 people in health institutions, and the number of tertiary hospitals in the region. The total cost of health is a measure of the overall level of investment in health in a country. The per capita cost of health is the ratio of the total cost of health in a given year to the average population over the same period. The high proportion of health costs per capita is considered in this paper as a possible reason for the large population base in China.

In the section on health affairs and health environment management, this paper measures the level of development of the health industry in terms of the strength of industrial pollution prevention and control and the degree of improvement of the living environment. For industrial pollution prevention and control, the amount of investment completed in industrial pollution control is considered; the degree of improvement of the living environment is measured using the rate of harmless treatment of domestic waste and the rate of urban sewage treatment.

In the section on health promotion services, this paper considers health services for special groups, mainly for the elderly population, measured using the number of beds and institutions per 1000 elderly population. Health promotion also includes leisure and sport, which is measured using the number of people working in the tourism industry and the number of travel agencies.

In the health protection and financial services section of this paper, income and expenses from personal lines insurance businesses are used as measurements. This is because life insurance is an insurance policy that covers the life and body of a person and is closely related to health. In terms of securing the pharmaceutical aspect, this paper adopts an indicator for the production of raw chemical drugs.

3.1.3. Core Explanatory Variable: Digital Economy (DE)

At present, there is no unified standard in the relevant literature dealing with the specific measurement of the digital economy. Zhu et al. [50] selected indicators to measure the digital economy from four aspects: digital economy foundation, digital industrialization, industrial digitization, and digital economy penetration. Yang et al. [39] followed the principles of systematicity, usability, and science to construct an evaluation index system for the development of the digital economy. The system consists of three dimensions: digital infrastructure, digital technology application, and digital industry development. Liu et al. [40] constructed an evaluation index system of China’s digital economy by province from three dimensions: informatization development, Internet development, and digital transaction development, and measured the data of 30 Chinese provinces from 2015 to 2018. Based on Liu et al. [40], Zhao et al. [51] filled the gap of digital economy measurement at city level.

Based on the above research, this paper adopts the entropy weight method to construct a comprehensive index of digital economy development from three aspects: basic installation, digital transactions, and digital industry. Among them, the basic installation aspect mainly selects reference indicators from the digital economy infrastructure construction level, including the length of fiber-optic cable lines, the number of computers used per 100 people, and the number of broadband access users. The number of mobile internet users, the number of enterprises with e-commerce, and e-commerce sales were selected for digital transactions. Indicators were selected for the digital industry in terms of both inputs and outputs, including e-commerce purchases, investment in fixed assets in society as a whole, the proportion of employees in information transmission services, and revenue from information technology services. Specific descriptions of each indicator are shown in

Table 3. Digital economy development level evaluation index system.

Indicators	Measurement Method	Unit	Weight
Digital infrastructure	Cable length per square kilometer	Kilometer	0.054
	Computers Used Per 100 Persons	Unit	0.055
	Internet Broadband Access Users	10,000 households	0.059
Digital transaction	Mobile Internet Users	10,000 households	0.051
	The Number of Enterprises with E-Commerce Transactions	Individual	0.105
	Sales of E-Commerce	10,000 yuan	0.137
Digital industry	Purchases of E-Commerce	10,000 yuan	0.152
	Total Investment in Fixed Assets	10,000 yuan	0.06
	Proportion of Employees in Information Transmission, IT Services, and Software Services	%	0.129
	Income from IT Services	100 million yuan	0.197

Note: All indicators are positively attributed to the digital economy.

.

3.1.4. Intermediary Variable: Technological Innovation (TI)

In the context of innovation-driven strategy, technological innovation is the key to achieve the development of the health industry. This paper draws on the choice of technological innovation indicators by Li et al. [52], Wu [44], and Zhu [43] to measure the level of technological innovation in terms of both inputs and outputs. Technological innovation inputs reflect the scale of factors that firms invest in in terms of innovation activities and are measured through R&D expenditures and R&D full-time personnel equivalents. Technological innovation output measures the ability of innovation inputs to be transformed into actual results. It reflects not only output outcomes but also the economic benefits flowing to the firm through technological innovation activities, measured with the number of patent applications and new product sales revenue. Meanwhile, the supply of health products relies heavily on pharmaceutical manufacturing and medical equipment and instrument manufacturing, so this paper chooses to use data from these two sectors to measure the level of technological innovation.

This paper uses the entropy value method to construct a system of technological innovation indicators, the details of which are shown in

Table 4. Technological innovation level evaluation index system.

Indicators	Measurement Method	Unit	Weight
Innovation Input	Manufacture of Medicines Expenditure on R&D	10,000 yuan	0.105
	Manufacture of Medical Equipment and Meters Expenditure on R&D	10,000 yuan	0.166
	Manufacture of Medicines and R&D Full-time Personnel Equivalent	man-year	0.078
	Manufacture of Medical Equipment and Meters R&D Full-time Personnel Equivalent	man-year	0.142
Innovation Output	Manufacture of Medicines Patent Applications	unit	0.086
	Manufacture of Medical Equipment and Meters Patent Applications	unit	0.154
	Manufacture of Medicines Sales Revenue of New Products	10,000 yuan	0.101
	Manufacture of Medical Equipment and Meters Sales Revenue of New Products	10,000 yuan	0.168

Note: All indicators are positively attributed to technological innovation.

. It shows that the weights of technological innovation inputs and outputs are 49.1% and 50.9%, respectively, indicating that both are equally important in measuring the level of technological innovation.

3.1.5. Control Variables

It is crucial to improve the model fit optimality by introducing control variables into Equation (8). Based on previous studies [22,53,54], the following variables were used as control variables in this study to explain the development of China’s health industry.

• Per capita GDP (LPGDP)

This study uses GDP per capita as a proxy for the level of economic development of each province in China. A higher GDP per capita indicates an increase in the standard of living of the population. An increase in the standard of living of the population can strengthen people’s awareness of health and increase health consumption, which in turn promotes the development of the health industry [22].

• Proportion of total health expenditure to GDP (HEGDP)

The ratio of total health costs to GDP refers to the ratio of total health costs in a given year to the gross domestic product (GDP) in the same period [53]. It is used to reflect the strength of the state’s financial investment in health over a certain period, as well as the importance that the government and society as a whole attach to health and the health of the population.

• Educational level (EL)

Education promotes social stratification, with the public having different health philosophies and aspirations depending on their level of education. In particular, groups with higher education place a higher value on health. Therefore, this paper argues that the level of education has a driving effect on the development of the health industry. Drawing on the methodology of Song et al. [22], the number of students with higher education as a proportion of the total population is used to measure the level of education.

• Population ageing (PA)

Populations are now ageing at a much faster rate than ever before. The United Nations General Assembly has declared 2021–2030 as the United Nations Decade of Action on Healthy Ageing, a decade of coordinated, catalytic, and collaborative action to promote longer and healthier lives. Population ageing presents opportunities and challenges for the development of the health industry. In the era of longevity, human life expectancy is gaining ground while facing health challenges that are quite different from those that preceded it. Health has become a more urgent need. This need will be more diversified and long-term, which becomes a strong driving force for the growth of the health industry, and the era of health is coming [54]. Therefore, we measure the level of ageing as the proportion of the population aged 65 years and over to the total population.

3.2. Data Description

This study collected panel data for 27 provinces in China (excluding Hong Kong, Macau, and Taiwan) from 2014–2021, with a total of 216 observations. The data selected for this study were all obtained from the China Statistical Yearbook [55], China Health and Health Statistical Yearbook [56], China Statistical Yearbook on High Technology Industry [57], and the statistical yearbooks of each province. In the process of collecting data, it was found that the data on the production of raw chemical drugs in Tibet, Qinghai, Xinjiang, and Hainan provinces were seriously missing. Thus, they were excluded from the sample in this paper. Meanwhile, there are more vacant values in China’s data on the digital economy before 2014, such as: the number of computers used per 100 people, e-commerce sales, and IT service revenue. This paper does not take 2022 data into account, because the data on the health industry in the China Health and Health Statistical Yearbook currently only covers up to 2021. In this paper, to keep the empirical results robust and realistic, we select data from the date range of 2014–2021 for analysis. Missing values for some provinces are supplemented with interpolation. Also, to eliminate the phenomenon of “heteroskedasticity” as much as possible, variables such as GDP per capita are introduced into the model after taking the natural logarithm. Descriptive statistics for the corresponding variables were calculated using Stata 17.0 software, as shown in

Table 5. Descriptive statistics of raw data.

Variables	Definition	N	Mean	Std.	Min	Max	Kurtosis	Skewness	Jarque-Bera
HI	Health industry	216	0.206	0.108	0.0407	0.538	3.011	0.858	26.483
DE	Digital economy	216	0.172	0.142	0.0103	0.785	6.301	1.757	209.213
TI	Technological innovation	216	0.106	0.149	0.00222	0.944	10.875	2.658	812.412
LPGDP	Ln per capita GDP	216	10.97	0.427	10.13	12.14	2.767	0.518	10.165
HEGDP	Proportion of total health expenditure to GDP	216	0.0690	0.0176	0.0401	0.148	4.803	0.913	59.231
EL	Educational level	216	0.0220	0.00502	0.0122	0.0425	5.131	1.164	89.614
PA	Population ageing	216	0.119	0.0248	0.0678	0.188	2.499	0.376	7.340

.

3.3. Empirical Model

3.3.1. Baseline Model

The baseline regression to examine the impact of the digital economy on the health industry in China is based on Equation (8). Considering the inclusion of a range of control variables including: GDP per capita, total health expenditure as a proportion of GDP, education level, and population ageing, Equation is (9):

$${HI}_{it}=c+\alpha {DE}_{it}+u_i+z_t+{\epsilon }_{it}$$

(8)

$${HI}_{it}=c+\alpha {DE}_{it}+{\beta {Control}_{it}+u}_i+z_t+{\epsilon }_{it}$$

(9)

where ${HI}_{it}$ is a provincial index measuring the year of the health industry and ${DE}_{it}$ indicates the digital economy index. ${Control}_{it}$ shows a set of control variables reflecting region-specific characteristics, namely LPGDP, HEGDP, EL, and PA. $u_i$ represents individual fixed effects, controlling for heterogeneity across provinces and varying over time, whereas $z_t$ signifies the time-fixed effects which capture the heterogeneity that vary over time but are constant across provinces. ${\epsilon }_{it}$ is the random error term, $c$ is the constant term, and $\alpha$ and $\beta$ are the parameters to be estimated. In particular, $\alpha$ captures the impact of the digital economy on the health industry.

3.3.2. Panel Threshold Effect Model

To test Hypothesis 1.2, this study further speculates on whether there is a non-linear association between the digital economy and the level of development of the health industry. Threshold regression extends linear regression by allowing the coefficients to vary between regions with heterogeneous characteristics. These regions are distinguished by the threshold value of one or more specific threshold variables. The presence of non-linear associations was tested using fixed effects panel threshold effects as proposed by Hansen [58]. This study further incorporates cross-sectional factors for each control variable as threshold variables in the tests. As a specific setting, the model takes the following form. Equations (10) and (11) represent a fixed-effects panel threshold model with a single threshold:

$${HI}_{it}=c_1+{\alpha }_1{DE}_{it}+{\beta {Control}_{it}+u}_i+z_t+{\epsilon }_{it},q_{it}<\gamma$$

(10)

$${HI}_{it}=c_2+{\alpha }_2{DE}_{it}+{\beta {Control}_{it}+u}_i+z_t+{\epsilon }_{it},q_{it}\ge \gamma$$

(11)

where $\gamma$ is the threshold parameter for the threshold variable $q$, which is estimated to investigate whether the coefficients on the digital economy (${\alpha }_1$ and ${\alpha }_2$) change significantly under different regimes below and above the threshold. Similarly, Equations (12)–(14) represent a two-threshold model with fixed effects:

$${HI}_{it}=c_1+{\alpha }_1{DE}_{it}+{\beta {Control}_{it}+u}_i+z_t+{\epsilon }_{it},q_{it}<{\gamma }_1$$

(12)

$${HI}_{it}=c_2+{\alpha }_2{DE}_{it}+{\beta {Control}_{it}+u}_i+z_t+{\epsilon }_{it},{\gamma }_1{\le q}_{it}<{\gamma }_2$$

(13)

$${HI}_{it}=c_3+{\alpha }_3{DE}_{it}+{\beta {Control}_{it}+u}_i+z_t+{\epsilon }_{it},q_{it}{\ge \gamma }_2$$

(14)

where ${\gamma }_1$ and ${\gamma }_2$ are the two threshold parameters of the threshold variable $q$, dividing the relationship between the digital economy and the health industry into three regimes.

3.3.3. Intermediary Effect Model

To test Hypothesis 2, an intermediary effects model was built on the basis of Equation (9):

$${TI}_{it}={\gamma }_0+{\gamma }_1{DE}_{it}+{\gamma }_2{Control}_{it}+u_i+z_t+{\epsilon }_{it}$$

(15)

$${HI}_{it}={\lambda }_0+{\lambda }_1{DE}_{it}+{\lambda }_2{TI}_{it}+{\lambda }_3{Control}_{it}+u_i+z_t+{\epsilon }_{it}$$

(16)

where ${TI}_{it}$ is the intermediary variable and ${Control}_{it}$ is a set of control variables in addition to the intermediary variable. $u_i$ and $z_t$ are province-fixed effects and year-fixed effects, respectively. ${\epsilon }_{it}$ is the random error. First, $\alpha$ is significant, otherwise the intermediary effect is insignificant. Second, when ${\gamma }_1$ and ${\lambda }_2$ are both significant, the intermediary effect is partially intermediated if ${\lambda }_1$ is significant and fully intermediated and if ${\lambda }_1$ is insignificant. Third, when either ${\gamma }_1$ or ${\lambda }_2$ is significant, Wen and Ye [59] suggest using the Bootstrap method for the intermediary effect test. If ${\lambda }_1$ and the test results of the Bootstrap method are significant, then a partial intermediary effect is created.

4. Empirical Results and Analysis

In order to explore the relationship between the digital economy and the health industry, this paper adopts a two-way fixed effects model [54,57,58]. Based on this, a panel threshold effect model and an intermediary effect model are used to further analyze how the digital economy affects the development of the health industry and to analyze the characteristics that this effect has [21,50].

4.1. The Spatial and Temporal Evolution of the Health Industry

In order to visually reflect the spatial and temporal evolution characteristics of the development level of the health industry in each province and city, this paper uses Arcgis10.6. to depict the current distribution of the health industry development in 2014 and 2021 in 27 provinces of China, as shown in Figure 2.

The overall level of development of the health industry in China has increased significantly in terms of time trends. At the same time, the number of cities with a high level of health industry development has increased from 10 to 12. In terms of spatial distribution, the imbalance of “high in the east and low in the west” in the level of health industry development persists and increases in 2021. This feature is consistent with the findings of Yang [15]. This paper argues that the level of regional economic development is one of the main reasons for this distribution. The western region has a weak economic base, relatively poor environmental carrying capacity, and insufficient investment in scientific and technological innovation, resulting in a low level of development of the health industry. The eastern region, as an economically developed region, pays more attention to investment in scientific and technological innovation and the optimization and upgrading of the health industry structure, which is conducive to the development of the health industry. The level of development of the health industry in midstream cities is relatively stable and has a tendency to develop for the better.

4.2. Baseline Regression

Currently, most empirical studies investigating the digital economy of China’s economic development use two-way fixed effects models, such as the effect of the digital economy on the coordinated development of the regional economy [60], the digital economy affecting rural environmental governance [61], and the digital economy driving the development of the rural consumer market [62]. Therefore, in order to systematically reveal the influence relationship between the digital economy and the health industry, based on the Hausman test, this paper also constructs a two-way fixed effects model that includes both time and individual effects.

Table 6. Results of baseline regression.

Explained Variable	Pooled OLS		Random Effects		Double Fixed Effects
	(1)	(2)	(3)	(4)	(5)	(6)
DE	0.500 ***	0.707 ***	0.327 ***	0.291 ***	0.220 ***	0.338 ***
	(0.119)	(0.102)	(0.0265)	(0.0456)	(0.0434)	(0.0512)
LPGDP		−0.157 ***		0.0150		0.209 ***
		(0.0398)		(0.0220)		(0.0513)
HEGDP		−1.912 ***		−0.267		0.330
		(0.538)		(0.234)		(0.270)
EL		−1.949		2.001		5.503 ***
		(1.734)		(1.335)		(1.695)
PA		1.720 ***		−0.0335		0.102
		(0.373)		(0.213)		(0.248)
Constant	0.120 ***	1.772 ***	0.150 ***	−0.0304	0.161 ***	−2.230 ***
	(0.0178)	(0.430)	(0.0161)	(0.216)	(0.00636)	(0.562)
Year fixed effects	No	No	No	No	Yes	Yes
Province fixed effects	No	No	No	No	Yes	Yes
Observations	216	216	216	216	216	216
R-squared	0.435	0.676	-	-	0.455	0.538
Hausman test	-	-	-	-	9.887 ***	1186.38 ***
p-value	-	-	-	-	0.002	0.000

Note: Significance at 0.01 levels indicated by ***. Robust standard error in parentheses.

reports the results of the baseline regressions. Columns (1) and (2) are fitted using pooled ordinary least squares (OLS) techniques, Columns (3) and (4) are random effects models, and Columns (5) and (6) are two-way fixed effects models. The Hausman test results confirm that fixed effects are better than random effects. Therefore, the following analysis is based on the estimation results in Columns (5) and (6).

The regression results in Column (5) show that there is a significant positive correlation between DE and HI. After further inclusion of control variables, the coefficient of DE changed from 0.220 to 0.338 and was significant at the 1% level of significance, which to some extent reflects the robustness of the estimation results. Thus, Hypothesis 1.1 is proved. Our empirical results further enrich the findings of Qin [23]. It also suggests that the Chinese government should give full play to the positive role of the digital economy on the development of the health industry. By fully releasing the dividends of the digital economy, the development of the digital economy can become an effective means to pull the development of the health industry.

In terms of the control variables, the coefficient on LPGDP is 0.209, which is significant at the 1% level of significance, according to the results of the two-way fixed effects model shown in Column (6) of Table 6. This means that, for every 1% increase in GDP per capita, HI will increase by 0.209 units. The reason for this is that a higher GDP per capita indicates a higher standard of living for the population. It enhances people’s health awareness and increases health consumption, which in turn promotes the development of the health industry.

The coefficient of EL was 5.503, which was significant at the 1% level of significance. This means that a one unit increase in EL leads to a 5.503 unit increase in DE, a result similar to previous findings by scholars [22]. The reason for this is that groups with different levels of education have different health philosophies and pursuits. Those with higher education are more motivated to consume, thus promoting the development of the health industry.

The coefficient on HEGDP in Column (6) is not significant. The possible reason for this is that total health costs in China are currently very low as a proportion of GDP. There is considerable room for growth. The coefficient of PA in Column (6) is not significant. This paper suggests that there may be two reasons for this. On the one hand, it is due to the family structure, as the Chinese attach particular importance to the concept of family and children generally have to support the elderly. On the other hand, China’s elderly health industry started late and has not yet formed a mature industrial chain. What is foreseeable is that its impact on the development of the health industry has great potential.

4.3. Robustness Test

4.3.1. Multicollinearity Test

Drawing on Xu et al. [63] and Huang et al. [64], before the model is estimated, the variance inflation factor (VIF) is calculated for each variable to determine whether there is multicollinearity. The results of the test are shown in

Table 7. Multicollinearity test results.

Variables	VIF	1/VIF
LPGDP	3.49	0.286156
DE	2.55	0.392606
EL	1.75	0.571616
PA	1.46	0.686919
HEGDP	1.29	0.775940
Mean VIF	2.11

. The maximum value of the VIF was 3.49, the minimum value was 1.29, and the mean value was 2.11. It is generally accepted that the value of the variance inflation factor does not exceed 10. There is no serious problem of multicollinearity, so the model can be estimated. Compared to the baseline regression results, the significance and sign of the explanatory variables did not change significantly. This indicates that the model passes the robustness type test.

4.3.2. Shortened Time Window

The sample period selected for this paper is 2014–2021. Since the 2023 China Health and Health Statistical Yearbook has not yet been published, the data related to 2021 in the health industry system are missing. This paper complements certain missing data via interpolation. To avoid the processed data from interfering with the empirical results and drawing on scholars such as Li [65] and Sun [66], this paper adjusts the sample to 2014–2020 and re-estimates it. The regression results are shown in Column (7) of

Table 8. The results of shortened time window and sample data tailing.

Variables	HI	HI
	(7)	(8)
DE	0.311 ***	0.439 ***
	(0.0584)	(0.0549)
LPGDP	0.255 ***	0.161 ***
	(0.0727)	(0.0552)
HEGDP	0.357	0.200
	(0.417)	(0.291)
EL	4.454 **	5.355 ***
	(2.192)	(1.818)
PA	0.0665	0.335
	(0.297)	(0.266)
Constant	−2.696 ***	−1.725 ***
	(0.790)	(0.604)
Year fixed effects	Yes	Yes
Province fixed effects	Yes	Yes
Observations	189	216
Number of provinces	27	27
R-squared	0.483	0.631

Note: Significance at 0.01 and 0.05 levels indicated by *** and **. Robust standard error in parentheses.

. The significance and sign of each explanatory variable did not change significantly compared to the baseline regression results. This indicates that the model passes the robustness test.

4.3.3. Sample Data Tailing

Extreme values exist due to the measurement of control variables such as the digital economy and technological innovation indicator system. Drawing on the approach of scholars such as Chen et al. [67], this paper next applies a 1% and 99% tailing process to each variable to eliminate the effect of outliers and extreme values on the estimation results. The results are shown in Column (8) of Table 8. The regression results are not significantly different from the previous paper: an increase in the digital economy significantly contributes to an increase in the level of development of the health industry; the coefficients on LPGDP and EL remain significant; and the coefficients on HEGDP and PA are not significant.

4.4. Regional Heterogeneity Analysis

Considering the unevenness of China’s regional development and drawing on the approach of scholars such as Zheng et al. [68], this paper divides the 27 provinces into three regions—east, central, and west—for regional heterogeneity analysis (see Appendix A). The specific regression results are shown in

Table 9. Subregional baseline regression results.

Variables	Eastern	Central	Western
	(9)	(10)	(11)
DE	0.306 ***	0.317 ***	0.716 ***
	(0.0686)	(0.111)	(0.184)
Control variables	Yes	Yes	Yes
Observations	80	64	72

Note: Significance at 0.01 level indicated by ***. Robust standard error in parentheses.

. It can be seen that the impact of the digital economy on the development of the health industry in all three regions is significant at the 1% confidence level.

In terms of effect size, the western region has the largest coefficient of 0.716, which may be due to the greater need to develop the digital economy in the western region in order to reduce the transaction costs of health industry development, increase productivity and innovation, and promote economic transformation. The coefficients for the eastern and central regions were 0.306 and 0.317, respectively, with little difference between them. The level of digital economy development in the eastern and central regions is higher, and the development of the digital economy and the health industry is in a virtuous cycle [68].

4.5. Panel Threshold Analysis

To test Hypothesis 1.2, a panel threshold model was used for empirical testing. The panel threshold was tested using Hansen’s [58] approach before estimating the threshold model. In this paper, a panel threshold regression was also considered with each control variable together as a threshold variable. Single, two-fold, and three-fold threshold tests were conducted for the different threshold variables using Bootstrap’s method. The results are presented in Appendix B.

The results In Appendix B show that the digital economy as a threshold variable only passed the single threshold test, while all the control variables as threshold variables failed the threshold test. On this basis, regression models for the corresponding threshold numbers were constructed, and the regression results are shown in

Table 10. Results of threshold estimates and confidence intervals.

Threshold Variable (q)	Threshold	Estimated Threshold Value	95% Confidence Interval
DE	Single	0.0250	[0.0230,0.0280]

and

Table 11. Threshold regression results.

Threshold Variable	HI
de_1(Th ≤ q)	0.768 ***
	(0.184)
de_2(Th > q)	0.232 *** (0.0618)
Control variables	YES
R-squared	0.490
N	216

Note: Significance at 0.01 level indicated by ***. Robust standard error in parentheses.

. From Table 10 and Table 11, it can be seen when the threshold variable is the digital economy. The digital economy is below the threshold value of 0.025; for every 1 unit increase in the digital economy, the development level of the health industry increases by 0.768 on average. After crossing the threshold value of 0.025, the development level of the health industry increases by an average of 0.232 for every 1 unit increase in the digital economy.

This suggests that, as the digital economy develops to a certain level, its impact on the development of the health industry shows a non-linear trend. In other words, the degree of positive impact on the health industry varies at different stages of the development of the digital economy. This is mainly because when the health industry is at a lower level of development, the digital economy can expand the supply and demand market of the health industry and promote the construction of the health industry chain. With the continuous improvement of the health industry chain, the role of the digital economy in the development of the health industry reaches saturation. At this point, new factors will provide new impetus for the development of the health industry.

4.6. Intermediary Effect Analysis

To test Hypothesis 2, this paper uses an intermediary effect model for empirical testing (see

Table 12. The test results of the function mechanism of the digital economy affecting the health industry development level.

Variables	HI	TI	HI	HI1
	(12)	(13)	(14)	(15)
DE	0.338 ***	0.619 ***	0.328 ***	0.338 ***
	(0.0512)	(0.0706)	(0.0614)	(0.0512)
TI			0.0167
			(0.0546)
LPGDP	0.210 ***	−0.0827	0.212 ***
	(0.0514)	(0.0709)	(0.0517)
HEGDP	0.00341	−0.00108	0.00343
	(0.00271)	(000374)	(0.00272)
EL	5.494 ***	−3.271	5.548 ***
	(1.693)	(2.337)	(1.707)
PA	0.102	−0.747 **	0.114
	(0.248)	(0.342)	(0.252)
_bs_1				0.125 ***
				(0.0438)
_bs_2				0.582 ***
				(0.0725)
Constant	−2.506 ***	1.055	−2.523 ***
	(0.608)	(0.839)	(0.613)
Year fixed effects	YES	YES	YES
Province fixed effects	YES	YES	YES
Observations	216	216	216	216
R-squared	0.966	0.966	0.966

Note: Significance at 0.01 and 0.05 levels indicated by *** and **. Robust standard error in parentheses.

).

Columns (12)–(15) of Table 12 show that the digital economy affects the level of health industry development through the level of technological innovation, confirming Hypothesis 2. Column (12) verifies the positive effect of the digital economy on the level of health industry development. Column (15) verifies that the digital economy promotes the level of technological innovation in the provinces. The regression coefficients of the digital economy in both columns are positive and significant at the 1% level. Relative to Column (12), the coefficient of the digital economy in Column (14) decreases, indicating that the increase in the level of technological innovation is the mechanism of action of the digital economy to promote the increase in the level of development of the health industry. The results in Columns (14) and (15) suggest that the level of technological innovation plays an intermediary role [69]. These empirical results support Hypothesis 2.

4.7. Further Discussion

From the previous empirical results, the impact of digital economy on the health industry is non-linear, and the turning point occurs when the digital economy development is 0.025. Of the 216 sample observations, 210 are greater than the turning point of 0.025, which means that the driving effect of the digital economy on the health industry has slowed down in most provinces. This paper considers two possible reasons for this. First, the driving effect of the digital economy on the health industry is significant in the initial integration stage of the digital economy and the health industry. The rapid improvement of digital health infrastructure and the rapid construction of links between the production, retail, and consumption ends of the health industry can effectively drive the development of the health industry in the short term. However, China’s health industry started late. The current health industry has the problems of imperfect health information management mechanisms, information sharing mechanisms, and health industry business management mechanisms, which have become important factors restricting the development of the health industry [23]. This may, to a certain extent, weaken its driving effect on the development of the health industry. Secondly, the empirical results show that the driving effect of digital economy on the health industry in most regions has changed from high-speed growth to medium-low growth, indicating that the health industry is in a critical period of transition to “high-quality” development, which is consistent with the results of Han [70]. Under the strategy of “Health China”, it is important to explore and construct a scientific, comprehensive, and China-friendly evaluation index system for the high-quality development of the health industry in China [71]. Song et al. [22] studied the intermediary role of education level in the development of the health industry driven by the digital economy. To enrich the existing theoretical results, this paper further explores the influence mechanism of the digital economy on the health industry. The intermediary role of technological innovation was analyzed by including education level as a control variable in the model. It was found that technological innovation significantly plays an intermediary effect.

5. Conclusions and Policy Recommendations

5.1. Conclusions

Based on the panel data of 27 provinces from 2014–2021, the entropy weight method was used to construct a comprehensive indicator evaluation system for the digital economy, technological innovation, and the health industry. Then, using a two-way fixed effects model, intermediary effect model, and panel threshold model, the impact of the digital economy on the health industry was empirically studied. The main findings are as follows: First, the digital economy is generally able to drive the development of the health industry. This conclusion is consistent with the findings of Song et al. [22] and passes the robustness test. In addition, this conclusion empirically verifies the theoretical analysis of Qin [23] on the digital economy driving the development of health industry. Meanwhile, there is regional heterogeneity in the impact of the digital economy on the health industry, with the strongest in the west, the second strongest in the center, and the weakest in the east. Second, the digital economy has a single threshold effect on the health industry. The driving effect of the digital economy on the health industry is significantly greater on the left side of the threshold than on the right side. The vast majority of provinces have currently crossed the threshold. This conclusion is not only an enrichment and deepening of the research field of health industry, but also a reflection of the remarkable effectiveness of the current construction of digital China and health China. Third, the digital economy not only has a direct impact on the development of the health industry, but can also indirectly promote the development of the health industry through technological innovation. Improving technological innovation is the mechanism of action for the digital economy to empower the development of the health industry. The findings fill the gap in the literature in the field of health industry and explore more potential paths for promoting the development of the health industry.

5.2. Policy Recommendations

In addition, to providing a series of empirical evidence for the digital economy to promote the development of the health industry, the findings of this paper also have the following policy implications: First, China should continue to promote the integration of the digital economy and the health industry. China can promote the construction of new infrastructure for “Internet + medicine”, encourage the development and application of digital technologies in the health industry, and reduce the cost of data development. The integration of digital technology industry, academia, and research in the health industry promote the integration of the digital economy and the health industry in terms of depth and breadth. Facing the current problems of information management mechanisms, information sharing mechanisms, and lack of business management qualifications in the health industry, the construction of information sharing platforms should be coordinated and promoted. The information barriers between relevant departments should be broken. China should improve the management system of the health industry, make use of the opportunity of digital economy development to carry out multi-industry innovation, and provide high-quality health services. China should actively build an international platform for digital healthcare, encourage the creation of Internet hospitals, and develop services such as teleconsultation and online education and training, further accelerate the construction of pilot bases for the export of digital medical technology, and rely on the digital economy to enhance the level of openness and international competitiveness of China’s health industry.

Second, the Chinese government should implement a dynamic and differentiated digital economy strategy. Considering the different driving effects of the digital economy on the development of health industries in different regions, it foreshadows the need to tailor to local conditions and tap the development potential of regional health industries. It is better to make the digital economy into a technical support system to effectively alleviate the imbalance in the development of regional health industries.

Finally, the Chinese government must insist on leading the way in terms of technological innovation. At present, the health industry as a whole shows cross-border integration and cluster development, with the rapid growth of the pharmaceutical industry, mainly innovative drugs, and high-end medical devices. Telemedicine, mobile medicine, precision medicine, intelligent medicine, and other medical models are developing rapidly, and new healthcare services such as Internet medicine, health consultation, and recreation services are taking shape. However, China’s health industry still has some obvious shortcomings. Targeting new drugs and high-end medical equipment are still mainly dependent on imports. The lack of independent innovation of key core technologies needs to be strengthened to improve the international competitiveness of the health industry. Health enterprises can accomplish digital transformation with the popularization and application of information technology. High-level technological innovation in the field of health needs to enhance the technological innovation capacity of enterprises and the transformation rate of innovation results. This can enrich the types and levels of health products and inject endogenous power into the development of the health industry.

All in all, the research on the digital economy and the health industry in this paper has important practical significance. For the Chinese government, “Internet + Healthcare” puts forward higher requirements for the government’s macro-governance. As a new economic form, the development of the digital economy cannot be guided by the government, and it requires the state to formulate a forward-looking strategic development plan. Moreover, the successful experience of China’s anti-epidemic program shows that the government actively plays the function of macro-governance to oversee the orderly development of the health industry. For enterprises, the integration and development of the digital economy and the health industry fits the requirements of the supply side of structural reform. The digital economy reduces the threshold of investment and entrepreneurship, which can mobilize the entrepreneurial enthusiasm of micro-subjects and enhance the innovation power of health enterprises. With the background of big data, health enterprises can also use the SP-LACE method to improve the economic efficiency of the health industry. The SP-LACE method can align inventory deployment and supply policies with changes in customer demand, improve customer satisfaction, and reduce the total cost of the supply chain [72].

5.3. Shortcomings and Suggestions for Further Research

We acknowledge that there are certain limitations to this study. Firstly, the relationship between the digital economy, technological innovation, and the health industry has not been studied in-depth, and only provincial data are used in this paper. Secondly, this study only analyzes the mechanism effect played by a single intermediary variable, and more intermediary variables can be selected in future studies. Finally, in future research, more new generation panel data analysis techniques could be applied for more in-depth analysis. For example, the systematic GMM method could be used to analyze the different situations between individuals and the dynamic change characteristics of individuals using bootstrapping panel data to improve the precision and accuracy of the estimation results.