The spatial spillover effects of digital transformation on healthcare services quality

Abstract

Objective

Economic growth and improved material living standards have raised people's expectations for healthcare service quality. The digitalization level of healthcare organizations can significantly impact meeting these expectations.

Methods

This study uses Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to calculate the digital transformation and healthcare service quality composite index. Digital transformation and healthcare service quality spatiotemporal evolution are studied using kernel density estimation, spatial Moran's I index and trend surface analysis. Second, the spatial Durbin model explores how digitalization directly impacts healthcare quality. Finally, digital transformation's spatial spillover impacts on healthcare service quality are examined using partial differential decomposition.

Results

The digital transformation gap is expanding as areas develop differently. Notwithstanding west-east expansion of digital transformation across China, the centre region demonstrates greatest expansion compared with northern or southern regions. Beijing-Tianjin-Hebei, the eastern coastline region, Sichuan-Chongqing and Guangdong are high-level, whereas the northeast, northwest and Yunnan-Guizhou are low. Healthcare quality has improved annually, although regional gaps have grown. The centre was found to have a greater healthcare gap than the east and west. North exceeds south, with the north-south gap growing in 2021 over 2012. Digital transformation improves local healthcare but degrades neighbouring care.

Conclusion

Situated within a digital framework, this research examines how digital transformation might improve the quality of healthcare services and the spatial spillover effects. The results indicate that digital transformation may markedly improve the quality of medical services and have spatial spillover effects. Limitations identified in this study include constraints in research methodologies and modest sample size. Consequently, future studies may refine the provincial sample to the level of prefecture-level cities, employing moderation and mediation effect models to more precisely evaluate the impact mechanism of digital transformation on the quality of medical services.

Keywords

Digital transformation healthcare quality Moran index spatial spillovers

Introduction

The criteria for the division of China's industrial structure indicated that the development of the service industry could effectively promote the upgrading of the industrial structure, which was due to the sustained growth of the economy and the improvement of people's living standards, prompting people's consumption demand to gradually change from the material to the service-oriented, resulting in a steady upward trend in the share of the service sector in gross domestic product (GDP), and to promote the industrial structure to the higher level of the evolution of the form.^1,2 Healthcare services are a unique feature of the service sector. They are at the centre of the tertiary sector (The tertiary sector, or service industry, refers to industries other than the primary and secondary sectors. In national economic accounting, the tertiary sector is typically divided into multiple industries, including but not limited to wholesale and retail trade, transportation, warehousing, postal services, accommodation and food services, finance, real estate and other services – including for-profit and non-profit services.) because they are directly related to people's health and quality of life.³ With the improvement of people's health awareness, the demand for healthcare services was also growing, and its proportion in the service industry was gradually increasing. In China, for example, the market size of the healthcare service industry continued to grow and was expected to remain on a growth trend in the coming years.^4-6 This trend reflected the importance of healthcare services in the tertiary sector and its significant contribution to socio-economic development. The healthcare service industry chain involved several links and participants, including hospitals, doctors, patients, drugs, equipment, insurance, etc. The healthy development of this industry chain was of great significance in promoting the prosperity of the tertiary industry.^7,8 Because of this, the importance of healthcare in the service sector as a whole cannot be overstated.

Economic growth raised people's material standard of living and spiritual pursuits and their standards for the quality of healthcare services.^9,10 Evaluating the quality of health services involved two aspects, namely, the technical and non-technical aspects of the service.¹¹ The level of technical services mainly refers to the use of their professional knowledge by medical personnel, and then through the introduction of advanced equipment in the medical institutions, the corresponding healthcare services to patients, to achieve the purpose of timely treatment and rehabilitation of the patient's disease.¹² The non-technical aspects were mainly reflected in two aspects: the healthcare organization's external environment and the healthcare staff's internal quality. This meant that a healthcare organization's environment and healthcare providers’ attitudes could significantly impact patients’ willingness to seek medical treatment.^13,14 As the world became more digitized, the popularity and depth of use of cutting-edge digital technology in healthcare were increasing, and patients were demanding more technical services from their healthcare providers.¹⁵

Not only that, with the continuous improvement of people's living standards, patients were concerned about the technical service level of medical institutions at the same time, paid more attention to the healthcare experience, the healthcare service environment, the attitude of healthcare service personnel and other non-technical requirements were also increasing day by day.¹⁶ In the digital information age, all industries were journeying to achieve digital transformation and healthcare services. As a necessity of modern life, digital transformation was inevitably a general trend.^17,18 To catch up with the trend of digitalization, the International Healthcare Membership Organisation (IHMO) was oriented towards the development model of Internet + healthcare, aimed at the effective transmission of digital information, realized the precise treatment of healthcare institutions, enhanced the sense of participation of patients in the process of healthcare, and thus solved the communication problems between doctors and patients and eased the tensions between doctors and patients.^19-21 In addition, many healthcare institutions in China had introduced robots to participate in the process of patient surgery and treatment, the equipment helped health care workers in laparoscopic surgery to obtain a clearer surgical field of vision, improved the quality of technical services in healthcare service institutions.^22,23

Literature review

In the current developmental trajectory of the world, digitalization is a widely discussed topic. Whether in academic research or corporate development, digitalization is pervasive. This is due to the growing awareness that the progress of enterprises and society should be integrated with digitalization. Consequently, there is a vast body of research on digital transformation. Furr et al.²⁴ argued that digital transformation dominated the process of global economic development and that digital transformation under executive decision-making tended to show alignment across the three dimensions of products and platforms, firms and ecosystems, and people and tools. Ahlskog et al.²⁵ explored different perspectives on digital transformation within organizations, using four Swedish manufacturing companies, which ultimately showed that the impact of digital transformation varied within organizations and that the critical factor influencing digital transformation was knowledge. Barthel²⁶ described the metrics for successful digital transformation. Ubiparipovic et al.²⁷ defined digital transformation's content and process structure in their study, identifying 19 projects across six stages that constitute digital transformation activities. Several studies have analysed the advantages of digital transformation. For instance, Vu et al.²⁸ pointed out that successfully achieving digital transformation can enhance a company's financial performance. Dou et al.²⁹ argued that successful digital transformation can also positively change the labour structure of an enterprise. Han et al.³⁰ pointed out that digital transformation enhances a company's market competitiveness, promotes employment and improves the social employment structure. Wade and Shan³¹ and Gabryelczyk³² believe that disasters such as the COVID-19 pandemic can adversely affect a company's digital transformation, ultimately leading to stagnation in development and failure in digital transformation.

To accurately evaluate the quality of medical services, it is necessary to understand the corresponding evaluation methods and influencing factors. Thakkar et al.³³ provided a quantitative framework for assessing service quality in the healthcare industry in their study. Gao et al.³⁴ used review mining and questionnaire surveys to establish a healthcare service quality evaluation system. They concluded that factors such as registration services, operational efficiency, consultation and communication, medication treatment, diagnostic processes and medical equipment all impact patients’ healthcare experiences, affecting healthcare institutions’ service quality. Xu and Yang³⁵ addressed the uncertainty in evaluating elderly healthcare services by proposing a new method based on Interval-Valued Pythagorean Fuzzy Sets and Decision-Making Trial and Evaluation Laboratory to determine indicator weights. Rehman et al.³⁶ combined multiple indicators with uncertainty in their study to develop a model for measuring healthcare service quality. They validated and refined the model using data from various medical specialties and healthcare institutions, ultimately deriving the Hospital Service Quality Fuzzy Index.

Mosadeghrad³⁷ argued that the quality of healthcare services is closely related to personal, organizational and environmental factors. Personal factors mainly refer to the performance of doctors and patients, while environmental factors refer to the external environment of healthcare institutions and the broader social environment. Additionally, numerous studies have indicated that digital transformation also impacts the quality of healthcare services. Basile et al.³⁸ explained in their research that digital transformation significantly enhances the quality of healthcare services. Bhadula and Sharma³⁹ posited that the Internet of Things enables high-quality, sustainable healthcare and nursing services. Gungoren⁴⁰ argued that achieving digital transformation in clinical laboratories can enhance the overall quality of patient care from a technical perspective. Ford et al.⁴¹ further supported the positive impact of digital transformation on healthcare service quality with numerous case studies in their research.

Based on literature identified here, there is a wealth of research on digital transformation and healthcare service quality, but there is still considerable room for expansion. First, existing research lacks quantitative measurements of healthcare service quality in specific regions, making it difficult to accurately assess the level of healthcare service quality in a region. Second, under the backdrop of digital transformation, few studies have evaluated the enabling effects of digital transformation on regional healthcare service quality from a digital perspective. Third, as the digital transformation process accelerates, the enabling effects of digital transformation on healthcare service quality are not simply linear relationships. However, existing research primarily explores these effects through linear and non-linear models, with few studies investigating the spatial spillover effects of digital transformation on healthcare service quality.

Therefore, based on existing research, this study proposes three considerations: first, the spatiotemporal evolution characteristics of digital transformation and healthcare service quality; second, the direct impact characteristics of digital transformation on healthcare service quality; and third, whether there is a spatial spillover effect of digital transformation on healthcare service quality. Based on this, this research uses kernel density estimation (KDE) and Moran's index to explore the spatiotemporal evolution characteristics of digital transformation and healthcare service quality. It utilizes a spatial Durbin model (SDM) to analyse the direct impact of digital transformation on healthcare service quality. It employs partial differential decomposition to decompose the spatial spillover effects, thereby exploring the spatial spillover effects of digital transformation on healthcare service quality.

Methods

Data

Data sources and regional overview

To identify data for all provincial units in China, the EPS data platform and statistical yearbooks were two key data sources used. Several yearbooks were consulted, including the China Statistical Yearbook, China Statistical Yearbook of Medical Construction, China Environmental Statistical Yearbook and China Urban Statistical Yearbook. It should be noted that this research found that the data of Hong Kong, Macao, Taiwan and Tibet were not complete during the statistical data, so they were excluded from the sample, and finally 30 provinces in China were selected as the research object, and finally, based on the update of the data, the period of 2012–2021 was selected as the sample period.

Variable measurements

Dependent variable: quality of medical services

The quality of healthcare service is related to the safety of patients’ lives and is closely associated with the development of the healthcare service industry. When the quality of services provided by healthcare institutions is high enough, both technical and non-technical levels of service level will have an advantage, so that patients will have a higher degree of trust in the medical institutions, that the organization produces a brand effect, the formation of positive interaction.⁴² Therefore, exploring the development status of healthcare service quality and further analysing the factors affecting the quality of medical service can provide the right direction for patients who need medical treatment and a theoretical basis for medical institutions to improve their service quality. Based on the above discussion, this study takes the quality of healthcare service as an explanatory variable and constructs the evaluation index system in Table 1 to measure its actual level.⁴³

Table 1.

The quality of medical services evaluation indicator system.

Primary indicators	Secondary indicators	Tertiary indicators	Unit	Attribute
Healthcare service quality	Medical personnel	Number of practicing (assistant) physicians per 1000 people	Persons/1000 people	+
		Number of registered nurses per 1000 people	Persons/1000 people	+
		Number of health administrators per 1000 people	Persons/1000 people	+
	Medical facilities	Total value of medical facilities per capita	Ten thousand CNY/1000 people	+
		Number of beds in medical institutions per 1000 people	Beds/1000 people	+
		Number of hospitals per 1000 people	Hospitals/1000 people	+
	Treatment capability	Maternal mortality rate	%	−
	Treatment capability	Number of tertiary hospitals		+
	Primary healthcare	Average number of village doctors per village	Persons/1000 people	+
	Primary healthcare	Number of primary healthcare institutions per 1000 people	Institutions/1000 people	+
	Healthcare efficiency	Average number of outpatients per doctor per day	Visits/Person	+
		Average number of inpatients per doctor per day	Days/Person	+
		Bed utilization rate	%	+
		Average length of stay	Days/Person	−

Core independent variable: digital transformation

Digitalization is a crucial topic in our times and a direction for future development. To achieve sustainable development and enhance operational efficiency, enterprises must progress with the times. In the era of the digital economy, it is essential to accelerate the integration of traditional industries with digital technologies, thereby realizing corporate value and securing a competitive position in an increasingly competitive society. Undoubtedly, the level of digitalization in healthcare institutions will significantly impact their development. Increasing the application rate of advanced technologies can promote advancements in medical technology, help address challenges inherent in traditional medical models and positively affect healthcare service quality.⁴⁴ Therefore, this research considers digital transformation the explanatory variable to explore its impact on healthcare service quality. When measuring its level, the evaluation indicator system shown in Table 2 was constructed.⁴⁵

Table 2.

Digital transformation evaluation indicator system.

Primary indicators	Secondary indicators	Tertiary indicators	Unit	Attribute
Digital transformation	Digital infrastructure	Length of fibre optic cable	km	+
		Number of broadband internet access ports	Ten thousand	+
		Number of domain names	Ten thousand	+
	Digital application	Number of computers used per 100 employees	Units	+
		Number of companies engaged in E-commerce	Units	+
		Number of companies with websites (per 100)	Units	+
	Digital innovation	Number of granted patent applications	Units	+
		Number of new product projects	Projects	+
		R&D expenditure	Ten thousand CNY	+
	Digital performance	Revenue from software products in the software industry	Ten thousand CNY	+
		New product sales revenue	Ten thousand CNY	+
		E-commerce sales amount	Hundred million CNY	+

Control variables

Generally speaking, economic development and policy orientation can significantly impact the development of regions and industries. There is also a substantial disparity in digitalization and healthcare levels between urban and rural areas. Consequently, the role of digitalization in healthcare services may be influenced by other factors. Based on this, this study sets control variables from five dimensions, urbanization level, social consumption level, degree of government intervention, environmental regulation level and unemployment level, to explore their roles in enhancing healthcare service quality.⁴⁶ The urbanization level is measured by the ratio of the urban population to the total population. The social consumption level is assessed by the ratio of total retail sales of consumer goods to the regional gross domestic product (GDP). The degree of government intervention is calculated using the ratio of fiscal expenditure to the regional GDP. The investment ratio in pollution control to industrial-added value measures the environmental regulation level. The urban registered unemployment rate assesses the unemployment level.

Model construction

Technique for order preference by similarity to ideal solution

The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is a widely employed intra-group comprehensive assessment method that effectively harnesses the information in the original data, yielding results that precisely represent the distinctions between evaluation alternatives. The fundamental procedure entails employing a cosine approach to ascertain the optimal and suboptimal solutions from a finite array of choices, utilizing a normalized raw data matrix. The distances between each evaluation object and both the optimal and least optimal solutions are calculated to ascertain the relative proximity of each object to the optimal solution, which underpins the assessment of their relative qualities. This research uses the entropy approach to ascertain the weights of each indicator. Subsequently, it applies the distance method between optimal and substandard solutions to calculate the comprehensive index for digital transformation and healthcare service quality.

First, normalize all indicators. The processing method for positive indicators is as follows:

X_{i j} = (x_{i j} - x_{\min}) / (x_{\max} - x_{\min})

(1)

The handling method for negative indicators is as follows:

X_{i j} = (x_{\max} - x_{i j}) / (x_{\max} - x_{\min})

(2)

Second, calculate the weighting coefficient $P_{i j}$ and entropy value $e_{j}$ ：

P_{i j} = X_{i j} / \sum_{i = 1}^{m} X_{i j}, k = 1 / \ln (m), e_{j} = - k \sum_{i = 1}^{m} P_{i j} \ln (P_{i j})

(3)

Third, calculate the coefficient of variation $d_{j}$ and entropy weight $w_{j}$ of the indicators:

d_{j} = 1 - e_{j} (j = 1, 2, \dots, n), w_{j} = d_{j} / \sum_{j = 1}^{n} d_{j}

(4)

Fourth, construct a standardized matrix:

Z_{i j} = X_{i j} / \sqrt{\sum_{k = 1}^{n} {(X_{i j})}^{2}}

(5)

Fifth, calculate the difference between each evaluation indicator and the optimal and worst vectors, and define the distance between the i-th evaluation object and the maximum value:

D_{i}^{+} = \sqrt{\underset{j = 1}{\sum^{m}} w_{j} {(Z_{j}^{+} - Z_{i j})}^{2}}

(6)

Define the distance between the i-th evaluation object and the minimum value:

D_{i}^{-} = \sqrt{\sum_{j = 1}^{m} w_{j} {(Z_{j}^{-} - Z_{i j})}^{2}}

(7)

Sixth, calculate the final score for digital transformation and healthcare service quality:

C_{i} = D_{i}^{-} / (D_{i}^{+} + D_{i}^{-})

(8)

Kernel density estimation

In academia, KDE is a commonly used non-parametric estimation method. Due to its high applicability and lack of prior knowledge and assumptions, it is widely applied in data visualization, anomaly detection and classification problems. The principle behind KDE is to accurately measure the probability density distribution of the entire sample data by overlaying the kernel functions around the data points of the observed area. This study utilized MATLAB R2024b software to develop code implementing a KDE model, with the specific formula shown below.

{\hat{f}}_{h} (x) = \frac{1}{n h} \underset{i = 1}{\sum^{n}} K (\frac{x - x_{i}}{h})

(9)

In the formula mentioned above, x represents the observation point, $x_{i}$ represents the sample points, where the value of i ranges from 0～N; $K (\frac{x - x_{i}}{h})$ denotes the kernel function, which specifies that the value is non-negative, the integral is 1, and conforms to the probability density property, with a mean value of 0. It should be noted that the kernel function formula weighs h to denote the bandwidth parameter affecting the function's range of action, which is usually greater than 0.

Moran index

The Moran index is the most commonly used method when spatial econometric analyses are required for research topics. The technique is mainly used to explain whether there are characteristics such as aggregation discrete or random distributions among independent factors in a spatial context. There is a distinction between the global Moran index and the local Moran index. The former is mainly applied to the judgment of regional spatial autocorrelation, indicating whether or not accumulation occurs in the study sample. The latter, in turn, can add to the locations where accumulation occurs in the region. The value of the Moran index is usually in [−1,1] and 0 as the cut-off point. When the value of the index is in [0,1], it indicates that there is a spatial positive correlation between the index data in this research, and the closer it is to 1, the more significant the correlation is, while the value of the index is in [−1,0], it indicates that there is a spatial negative correlation between the index data and the smaller the value is, the bigger the spatial difference is. This study used Stata 17.0 software to write code to calculate the Moran index, the global Moran index is calculated as follows:

M o r a n^{'} s I = \frac{\underset{i = 1}{\sum^{n}} \underset{j = 1}{\sum^{n}} W_{i j} (Y_{i} - \bar{Y)} (Y_{j} - \bar{Y})}{S^{2} \underset{i = 1}{\sum^{n}} \underset{j = 1}{\sum^{n}} W_{i j}}

(10)

Within the above equation is the sample variance and an element in the inverse distance space weight matrix.

Construction of spatial weight matrix

When constructing a spatial econometric model, it is necessary to consider how changes in regional location may affect variables. Therefore, a method must be employed to represent the geographical relationships between cities. When constructing spatial econometric models, scholars, both domestically and internationally, typically use a spatial weight matrix to measure these spatial relationships. In actual geographical area research, the most commonly used spatial weight matrices are adjacency, k-nearest neighbour and inverse distance spatial weight matrices. This research uses Stata 17.0 software to write code to construct spatial weight matrices. The basic form of a spatial weight matrix is:

W = [\begin{matrix} w_{11} & w_{12} & \dots & w_{1 n} \\ w_{21} & w_{22} & \dots & w_{2 n} \\ ⋮ & ⋮ & ⋱ & ⋮ \\ w_{n 1} & w_{n 2} & \dots & w_{n m} \end{matrix}]

(11)

Among them, the diagonal element $w_{i j} = 0 (i = j)$ in the matrix represents that the distance between a city and itself is zero.

The adjacency weight matrix, the specific calculation formula, is shown below.

W_{i j} = {\begin{matrix} 1, R e g i o n i i s a d j a c e n t t o r e g i o n j \\ 0, R e g i o n i a n d j a r e n o t a d j a c e n t \end{matrix}

(12)

Distance-based k-nearest neighbour spatial weight matrix, with the specific calculation formula shown below.

W_{i j} = {\begin{matrix} 1, T h e d i s t a n c e b e t w e e n r e g i o n i a n d r e g i o n j i s l e s s t h a n d \\ 0, T h e d i s t a n c e b e t w e e n r e g i o n i a n d r e g i o n j i s n o t l e s s t h a n d \end{matrix}

(13)

Distance-based spatial weight matrix form, with the specific calculation formula shown below.

W_{i j} = 1 / d_{i j}^{2}

(14)

In this context, $d_{i j}$ denotes the distance between region i and region j.

The spatial Durbin model

The SDM is a well-established option for spatial econometric analyses. The model can not only elucidate the impact of the development of explanatory variables on local explanatory variables but also clarify the impact of changes in the level of variables on neighbouring regions. Regarding the model's characteristics, the SDM strengthens the spatial autoregressive model by introducing spatially lagged variables and considering the spatial correlation between the variables more deeply. Based on this, this study uses the SDM to carry out this study; Stata 17.0 software is used to construct a spatial Dubin model for verification, which is calculated as follows:

S e r v_{i t} = β_{0} + ρ W \times S e r v_{i t} + β_{1} D i g t r a_{i t} + ρ_{1} W \times D i g t r a_{i t} + η X_{i t} + ρ_{2} W \times X_{i t} + ε_{i t}

(15)

In equation (15), $S e r v_{i t}$ is the healthcare service quality index; $D i g t r a_{i t}$ is the digital transformation index; $X_{i t}$ represents the indices of the control variables, W denotes the spatial weight matrix, $W \times S e r v_{i t}$ is the spatial lag of healthcare service quality, $W \times D i g t r a_{i t}$ is the spatial lag of digital transformation, $W \times X_{i t}$ is the spatial lag of the control variables, $ε_{i t}$ is the spatial lag error term.

Results

Descriptive statistical results analysis

Based on the selection of the above variables, this research uses Stata software to carry out descriptive statistical analysis, aiming at a preliminary understanding of the data performance. The results are shown in Table 3. The data in the table show that this study has selected a sample size of 300, and the data analysis is carried out from the four directions of the mean, the standard deviation, the maximum value, and the minimum value. Firstly, looking at the data on healthcare services quality and digital transformation, it can be seen that the mean value is greater than the standard deviation. There is a significant gap between the maximum and minimum values, which indicates that the degree of dispersion of the data is relatively low. There is variability in the development of the region. The same conclusion can be drawn from analysing the statistics of the control variables. There are no outliers in the overall data, which should be investigated further.

Table 3.

Descriptive statistical analysis.

Variables	Obs	Mean	Std.	Min	Max
Healthcare service quality	300	0.356	0.056	0.242	0.562
Digital transformation	300	0.158	0.128	0.037	0.801
Urbanization level	300	0.602	0.118	0.363	0.896
Social consumption level	300	0.384	0.069	0.222	0.538
Level of government intervention	300	0.251	0.103	0.107	0.643
Level of environmental regulation	300	0.003	0.004	0.000	0.031
Level of unemployment	300	0.032	0.006	0.012	0.046

Spatial characteristics analysis of digital transformation and healthcare service quality

Spatial and temporal clustering characteristics of digital transformation and healthcare service quality

Figure 1 presents the KDE plot of healthcare service quality generated using the KDE model. The analysis of this plot focuses on three main aspects: the movement of the curve's centre, the presence of the tailing phenomenon and the existence of peaks in the curve. The kernel density curve of healthcare service quality indicates that from 2012 to 2014, the centre of the curve shifted to the left. From 2014 to 2021, the centre moved to the right, suggesting that the development of healthcare service quality initially declined and then improved. A closer examination reveals that the curve exhibited slight tailing in specific years, indicating some disparities in healthcare service quality across regions. Additionally, the peaks of the curve narrowed from 2012 to 2014 and widened from 2014 to 2021. This suggests that regional differences in healthcare service quality in the study area decreased between 2012 and 2014 but widened again between 2014 and 2021.

Figure 1.

Kernel density estimation plot of healthcare service quality.

Figure 2 presents the KDE plot of digital transformation generated using the KDE model. The analysis of the curve follows the same standards mentioned earlier. Observing the movement of the curve's centre reveals that from 2012 to 2021, the centre continuously shifted to the right, indicating a yearly improvement in digital transformation. Additionally, the presence of tailing phenomena means that the curve exhibited tailing for several years, suggesting significant regional differences in the level of digital transformation. Lastly, observing the state of the peaks reveals that the peaks of the curve widened over the study period, with a significant difference between 2012 and 2021. This indicates that regional disparities in digital transformation have been widening year by year.

Figure 2.

Kernel density estimation plot of digital transformation.

Spatial distribution characteristics of digital transformation and healthcare service quality

To further analyse the spatial distribution characteristics of healthcare service quality in China, this study utilized ArcGIS software to create Figure 3, the spatial distribution map of healthcare service quality. Using data from 2012 and 2021 as time points, healthcare service quality levels across various regions were divided into five categories. Overall, the number of provinces with low and relatively low healthcare service quality in China has significantly decreased. In 2012, the number of provinces with low and relatively low levels of healthcare service quality was 12; in 2021, this number decreased to 8. The number of provinces with high and relatively high levels of healthcare service quality increased from 8 in 2012 to 17 in 2021, representing a growth rate of 112.5%. This indicates that the quality of healthcare service in China has continuously improved during the sample period. From a local perspective, in 2012, regions with high and relatively high levels of healthcare service quality were primarily located in the Northwest (Xinjiang), Sichuan-Chongqing area (Sichuan), Beijing-Tianjin-Hebei area, East China and Guangdong Province. Regions with low and relatively low levels were mainly concentrated in the Northwest and Yunnan-Guizhou areas. By 2021, with Xinjiang, Beijing-Tianjin-Hebei and Sichuan-Chongqing as the core areas, provinces with high and relatively high healthcare service quality began to exhibit a diffusion trend, spreading to surrounding regions. This led to the Northeast and Yunnan-Guizhou areas reaching high or relatively high levels of healthcare service quality. Based on the above results, it can be preliminarily inferred that education significantly enhances healthcare service quality. Regions with higher levels of healthcare service quality also tend to have more advanced medical education sectors, leading the nation in this regard.

Figure 3.

Spatial distribution of healthcare service quality.

This study aims to explore the impact of digital transformation on healthcare service quality. Therefore, further investigation of the current state of digital transformation development and revealing its spatial distribution characteristics is necessary. Based on the research mentioned above, this study also incorporated the data for digital transformation into ArcGIS software to create Figure 4, the spatial distribution map of digital transformation. The time points and classification levels used are consistent with those in Figure 3. Observing the spatial distribution map for 2012, it is evident that the overall level of digital transformation in China was relatively low. Many provinces were at a medium level or below, with only six provinces achieving relatively high or high levels of digital transformation.

Figure 4.

Spatial distribution of digital transformation.

In contrast, the spatial distribution map for 2021 reveals a less optimistic development of digital transformation. Seventy percent of the provinces remain at a medium level or below, while only 30% have achieved relatively high levels of digital transformation. Notably, in 2012, provinces with high and relatively high levels of digital transformation were mainly concentrated in the Beijing-Tianjin-Hebei region, Guangdong Province and the eastern coastal areas. By 2021, regions with relatively high or high levels of digital transformation corresponded with those having high or relatively high healthcare service quality, including the Beijing-Tianjin-Hebei region, the eastern coastal areas, the Sichuan-Chongqing region and Guangdong Province. Additionally, provinces with relatively low levels of digital transformation were primarily concentrated in the Northeast, Northwest and Yunnan-Guizhou areas. This suggests a possible correlation between changes in healthcare service quality and the progress of digital transformation.

Spatial correlation between digital transformation and healthcare service quality

To verify a spatial correlation between healthcare service quality and digital transformation, the data for these variables were input into the global Moran's I calculation formula, resulting in the test outcomes shown in Table 4. Observing the test results for the explanatory variable, digital transformation, it is evident that from 2012 to 2021, the Moran's I values were within the range [0,1], and the P-values were consistently less than 0.05. This indicates a significant positive spatial autocorrelation in the data for digital transformation. Similarly, observing the test results for healthcare service quality reveals that Moran's I values also remained within the range [0,1], with P-values consistently less than 0.05. This indicates that the data for healthcare service quality also exhibit significant spatial autocorrelation.

Table 4.

Spatial econometric model test results.

Year	Digital transformation			Healthcare service quality
Year	I	Z_value	P_value	I	Z_value	P_value
2012	0.214	3.183	0.001	0.153	2.342	0.010
2023	0.147	2.312	0.010	0.107	1.790	0.037
2014	0.165	2.528	0.006	0.183	2.740	0.003
2015	0.157	2.422	0.008	0.203	3.016	0.001
2016	0.141	2.215	0.013	0.155	2.435	0.007
2017	0.121	1.976	0.024	0.214	3.201	0.001
2018	0.101	1.755	0.040	0.194	2.931	0.002
2019	0.102	1.779	0.038	0.157	2.488	0.006
2020	0.094	1.682	0.046	0.114	1.900	0.029
2021	0.102	1.768	0.038	0.183	2.764	0.003

Spatial trends of digital transformation and healthcare service quality

Spatial trend analysis of healthcare service quality

To comprehensively analyse the current state of healthcare service quality development, this study utilized GIS software to create Figure 5, illustrating the spatial trend characteristics of healthcare service quality. The aim is to validate the trend of healthcare service quality changes along the z-axis by observing the fitted curve variations on the x and y coordinates in the spatial coordinate system. Observing Figure 5(a) and (b), it is evident that the fitted curve variations along the x and y axes in both 2012 and 2021 are generally consistent. The fitted curve along the x-axis exhibits a high-low-high pattern from east to west, while the fitted curve along the y-axis shows a low-high-low pattern from south to north. Specifically, in 2012, the healthcare service quality levels in the eastern and western regions were relatively similar and higher than in the central region. The central region, in turn, had higher healthcare service quality than the southern and northern regions, with the south region outperforming the north region. In 2021, the healthcare service quality in the eastern region began to surpass that of the western region. Similarly, the quality of healthcare service in the central region remained lower than that of the eastern and western regions. However, the northern region's healthcare service quality was higher than the southern regions, and the north-south disparity became more pronounced compared to 2012.

Figure 5.

Spatial trend characteristics of healthcare service quality. (a) 2012 and (b) 2021.

Figure 6 illustrates the spatial trend characteristics of digital transformation created using the abovementioned method. The aim is to clarify the spatial performance of digital transformation, setting a foundation for subsequent regression analysis. In Figure 6, the spatial coordinate system's z-axis represents the explanatory variable's development level. Figure 6(a) and (b) shows that the fitted curve variations along the x and y axes are also generally consistent. From a spatial trend perspective, the fitted curves in both figures exhibit characteristics of being lower in the west and higher in the east, with the central region having higher levels than the southern region, which in turn is higher than the northern region. The level of digital transformation gradually increases from west to east, while from north to south, it first rises and then falls, reaching its peak in the central region. In summary, there are significant differences in the level of digital transformation across eastern and western, as well as northern and southern areas of China, with apparent spatial trend variations.

Figure 6.

Spatial trend characteristics of digital transformation. (a) 2012 and (b) 2021.

Analysis of the spatial spillover effects of digital transformation on healthcare service quality

Testing of the spatial econometric model

The previous tests show that digital transformation and healthcare service quality exhibit significant spatial autocorrelation and have visualizable spatial distribution characteristics. Therefore, this study employs a spatial econometric model to analyse whether there are spatial effects between digital transformation and healthcare service quality. Relevant tests were conducted to determine the appropriate model. The selection of the spatial econometric model was carried out using the LM test, Hausman test and LR test. Table 5 presents the test results of the spatial econometric model. Observing the LM test results, it is evident that both the spatial lag and spatial error terms should be considered when using a spatial econometric model. Therefore, the SDM is the most appropriate choice for this study. According to the Hausman test results, whether random effects exist in the SDM can be verified. The test P-value in the table is 0.021, indicating that fixed effects should be selected. The LR test can ascertain whether the SDM with fixed effects degenerates into a Spatial Lag Model or a Spatial Error Model. The test results in the table provide a negative answer. Simultaneously, the SDM model measures the reciprocal influence between regions via a spatial weight matrix, illustrating both the spatial spillover of the dependent variable and the spatial transmission mechanism of the independent variable. SAR alone addresses the spatial dependence of the dependent variable, whereas SEM pertains solely to the spatial correlation of the error component. Compared to these two models, SDM offers a more thorough and precise framework for analysing spatial effects, making it appropriate for intricate spatial interaction scenarios. Therefore, this study should use the SDM with fixed effects to conduct the analysis.

Table 5.

Test results of the spatial econometric model.

Model test	Statistic	P_value
LM-error	31.69	0.000
Robust LM-error	26.30	0.000
LM-lag	13.15	0.000
Robust LM-lag	7.76	0.005
LR SDM-SEM	14.46	0.025
LR SDM-SEM	48.42	0.000
Hausman	14.96	0.021

Analysis of baseline regression results

Table 6 presents the test results after regression analysis using the SDM. Model (1) explains the impact of digital transformation on local healthcare service quality, while model (2) explains the impact of digital transformation on the healthcare service quality of neighbouring regions. Specifically, the central effect regression coefficient of digital transformation in the model (1) is 0.067, which is significant at the 1% level. This indicates that digital transformation has a considerable positive impact on healthcare service quality. For every one-unit increase in digital transformation, healthcare service quality improves by 0.067%. A 0.067% enhancement in the quality of medical services may appear negligible, although it possesses substantial implications in healthcare environments. Consider a premier hospital with an annual outpatient volume of one million visits; this enhancement may result in 670 fewer misdiagnoses or a decrease of 6700 h in patient waiting time. At the same time, the spatial spillover effect regression coefficient of digital transformation in the model (2) is −0.105, which is significant at the 10% level. This indicates that the development of digital transformation may harm the healthcare service quality of neighbouring regions, inhibiting the improvement of their service levels.

Table 6.

Baseline regression test results.

Variables	Model (1)	Model (2)
Variables	Main effect	Spatial spillover effect
Digital transformation	0.067***	−0.105*
Digital transformation	(3.41)	(−1.75)
Urbanization level	0.236***	0.286***
	(11.08)	(4.54)
Social consumption level	0.110***	−0.066
Social consumption level	(3.36)	(−0.82)
Government intervention degree	0.060**	0.380***
Government intervention degree	(2.25)	(3.86)
Environmental regulation level	−1.570***	2.040
Environmental regulation level	(−2.59)	(1.10)
Unemployment level	0.564*	2.567***
Unemployment level	(1.77)	(2.59)
rho	2.127***
rho	(6.10)
sigma2_e	0.001***
sigma2_e	(12.22)
Observations	300	300
R-squared	0.643	0.643
Number of ids	30	30

Note: z-statistics in parentheses, ***P < 0.01, **P < 0.05, *P < 0.1.

On the one hand, the acceleration of digital transformation helps improve healthcare efficiency, optimize the medical environment and promote academic exchange. For instance, the advent of electronic medical records (EMRs) can eliminate the cumbersome process of handwritten records in traditional settings. Introducing surgical robots can provide surgeons with a more precise surgical view, ensuring the smooth completion of operations. Additionally, the emergence of intelligent ward systems enables real-time monitoring of room temperature and air quality, providing optimal conditions for patient recovery. Moreover, digital platforms can overcome spatial barriers, facilitating the flow of academic information between hospitals and medical universities. This promotes the advancement of the medical field, leading to an overall improvement in healthcare service quality. On the other hand, medical institutions that are early adopters of digital transformation may establish a monopoly in technology integration, thereby gaining advantages in brand reputation and medical technology. This can inevitably impact similar medical institutions in neighbouring regions, causing their healthcare service quality to fall short of patient expectations.

Analysis of spatial spillover effects

To gain a more thorough understanding of the spatial performance of digital transformation and healthcare service quality, this study decomposes the spatial spillover effects discussed earlier. The results of this decomposition using the partial differentiation method are recorded in Table 7. Observing the results in the table, the direct effect coefficient of digital transformation is 0.068, and the indirect effect coefficient is −0.271, both of which are significant at the 1% level. This indicates that digital transformation has a significant positive effect on the healthcare service quality within the region while simultaneously exerting a significant inhibitory effect on the healthcare service quality of neighbouring areas. Since the inhibitory effect of digital transformation on the healthcare service quality of neighbouring regions is more robust than its promotive effect on the local region, and the total effect coefficient also clearly indicates this, it can be concluded that digital transformation has a significant overall inhibitory effect on healthcare service quality across the entire region. Therefore, medical institutions that are early adopters of digital transformation may have significant advantages regarding resource competition and digital technology monopoly. This can promote local healthcare service quality while exerting a more substantial inhibitory effect on neighbouring regions, ultimately hindering the coordinated development of the entire area.

Table 7.

Test results of spatial spillover effects.

Variables	Model (1)	Model (2)	Model (3)
Variables	Direct effect	Indirect effect	Total effect
Digital transformation	0.068***	−0.271***	−0.202**
Digital transformation	(3.83)	(−3.40)	(−2.45)
Urbanization level	0.241***	0.362***	0.603***
Urbanization level	(11.76)	(5.61)	(9.30)
Social consumption level	0.113***	−0.054	0.058
Social consumption level	(3.60)	(−0.61)	(0.65)
Government intervention degree	0.067**	0.444***	0.511***
Government intervention degree	(2.55)	(3.98)	(4.68)
Environmental regulation level	−1.523***	2.081	0.558
Environmental regulation level	(−2.58)	(0.99)	(0.26)
Unemployment level	0.639**	3.108**	3.747***
Unemployment level	(1.99)	(2.39)	(2.70)
Observations	300	300	300
R-squared	0.643	0.643	0.643
Number of ids	30	30	30

Note: z-statistics in parentheses, ***P < 0.01, **P < 0.05, *P < 0.1.

Robustness test

A robustness check should be conducted after deriving the above conclusions to avoid the possibility of coincidental findings in this study. Table 8 presents four models, each representing a different robustness check method. The research initially substituted the second-order inverse distance spatial weight matrix with a spatial adjacency matrix and a spatial k-nearest neighbour matrix, subsequently conducting regression analysis again. The primary regression coefficients for digital transformation were 0.071 and 0.55, while the geographical spillover effects were −0.331 and −0.382, respectively. These findings align with the results shown in Table 6, affirming the robustness of the study's conclusions. Subsequently, this study uses the entropy method to replace the TOPSIS measurement method for digital transformation and healthcare service quality and performs the regressions again. The regression results remain consistent with those in Table 6. In summary, the conclusions of this study have passed the robustness checks and are not coincidental.

Table 8.

Robustness test results.

Variables	Model 1		Model 2		Model 3		Model 4
Variables	Adjacency matrix		K-nearest neighbour matrix		Entropy method for digital transformation		Entropy method for healthcare service quality
Digital transformation	0.071***	−0.331***	0.055***	−0.382***	0.067***	−0.277***	0.059***	−0.396***
Digital transformation	(3.66)	(−4.17)	(2.89)	(−6.51)	(3.71)	(−3.83)	(2.66)	(−4.36)
Urbanization level	0.223***	−0.070	0.255***	−0.061	0.224***	−0.081	0.267***	0.025
Urbanization level	(12.20)	(−0.88)	(14.50)	(−1.08)	(12.24)	(−1.02)	(12.81)	(0.26)
Social consumption level	0.090***	−0.651***	0.137***	−0.301***	0.091***	−0.669***	0.111***	−0.689***
Social consumption level	(3.22)	(−5.18)	(4.57)	(−3.49)	(3.23)	(−5.31)	(3.50)	(−4.85)
Government intervention degree	−0.025	−0.072	−0.067**	0.014	−0.023	−0.051	−0.071**	0.025
Government intervention degree	(−0.97)	(−0.65)	(−2.37)	(0.29)	(−0.88)	(−0.46)	(−2.40)	(0.20)
Environmental regulation level	−1.358**	0.426	−1.569***	2.737**	−1.353**	0.167	−0.990	2.007
Environmental regulation level	(−2.30)	(0.21)	(−2.74)	(1.98)	(−2.28)	(0.08)	(−1.47)	(0.85)
Unemployment level	0.425	3.422**	−0.108	1.624	0.387	3.340**	0.154	3.579**
Unemployment level	(1.33)	(2.14)	(−0.37)	(1.41)	(1.22)	(2.08)	(0.42)	(1.96)
Observations	300	300	300	300	300	300	300	300
R-squared	0.712	0.712	0.716	0.716	0.308	0.308	0.463	0.463
Number of ids	30	30	30	30	30	30	30	30

Note: z-statistics in parentheses, ***P < 0.01, **P < 0.05, *P < 0.1.

Endogeneity test

This research utilizes a two-stage least squares estimation method using panel instrumental variables to address potential omitted factors or measurement errors affecting the relationship between digital transformation and healthcare service quality, hence testing for endogeneity. Two indicators – digital transformation policy and the extent of industrial digitalization – are chosen as instrumental variables for evaluating digital transformation. Digital transformation policies are categorized as virtual variables within the domain of variable selection. In August 2015, the State Council released the ‘Action Plan for Promoting Big Data Development,’ highlighting the use of big data technologies to facilitate significant transformations, integrations and innovations. Consequently, data preceding 2015 are assigned a value of 0, but values for subsequent years are designated as 1. The degree of industrial digitalization is quantified by the number of websites each firm possesses. The correlation between the instrumental and dependent variables indicates that both digital transformation policies and levels of industrial digitalization exert a significant direct influence on regional digital transformation. However, they do not significantly affect the quality of regional healthcare services. Consequently, the instrumental factors chosen in this study are typical. Thus, two variables were selected as instrumental factors for endogeneity assessment, and the two-stage least squares approach was utilized to model and verify the endogeneity. The test outcomes are displayed in Table 9. As shown in Table 9, the regression results indicate that the digital transformation policies and industrial digitalization levels selected in this study passed the significance test at the 1% significance level in the first regression stage, indicating that the selected instrumental variables are representative. Furthermore, the impact of digital transformation on healthcare service quality is significant at the 5% level, indicating a significant positive correlation between digital transformation and healthcare service quality. This is consistent with the regression results of the benchmark model, indicating that the research conclusions are robust.

Table 9.

Endogeneity test.

Variables	Digital transformation policy		Industrial digitalization
Variables	First stage	Second stage	First stage	Second stage
Instrumental variables	0.042***		0.227***
Instrumental variables	(3.99)		(5.03)
Digital transformation		0.804***		0.296**
Digital transformation		(4.14)		(2.33)
Urbanization level	0.332***	0.013	0.325***	0.437***
Urbanization level	(7.61)	(0.16)	(6.97)	(7.92)
Social consumption level	0.307***	−0.189**	0.200***	0.169***
Social consumption level	(4.38)	(−2.02)	(2.62)	(2.74)
Government intervention degree	−0.147***	0.224***	−0.128***	0.092***
Government intervention degree	(−4.41)	(7.59)	(−4.63)	(4.52)
Environmental regulation level	−2.058*	−0.450	−4.890***	−4.380***
Environmental regulation level	(−1.71)	(−0.49)	(−3.42)	(−4.40)
Unemployment level	−5.154***	3.771***	−4.992***	−1.901**
Unemployment level	(−5.55)	(2.98)	(−5.18)	(−2.57)
Constant	0.027*	0.116***	−0.018*	0.129***
Constant	(1.74)	(2.91)	(−1.76)	(4.12)
Observations	300	300	300	300
R-squared	0.445	0.475	0.452	0.596
Number of ids	30	30	30	30

Note: z-statistics in parentheses, ***P < 0.01, **P < 0.05, *P < 0.1.

Discussion

This study utilizes data from 30 provinces in China from 2012 to 2021 as its research focus. First, it used the TOPSIS comprehensive evaluation approach to compute the overall score index for digital transformation and healthcare service quality across each province. It subsequently employs KDE and spatial Moran's I to ascertain that digital transformation and healthcare service quality demonstrate notable geographical evolution characteristics. Gong et al.⁴⁷ discovered substantial disparities in the quality of medical care among various regions in Guangzhou City. The calibre of services in major metropolitan regions surpasses that in outlying and peripheral places. The quality of medical services demonstrates a bimodal distribution, reflecting variation in service quality, accompanied by an increasing trend in areas of low-quality care. Second, the spatial econometric model demonstrates a strong positive relationship between digital transformation and medical service quality. Wang and Shao⁴⁸ asserted that, within the framework of digital transformation, hospitals ought to enhance the development of appointment information platforms, internal physician quality management and the integration of information between self-service terminals and information platforms to elevate the quality of hospital services. Ding et al.⁴⁹ state that hospitals may utilize big data technologies and sophisticated algorithms to collect, process, analyse and apply diverse indications in real-time throughout the medical treatment process. This enhances hospital administration efficiency while bolstering medical service quality and optimizing resource allocation. Third, digital transformation has a significant spillover effect on the quality of healthcare services. Liao et al.⁵⁰ argue that integrating digital health into reengineering healthcare services signifies a pivotal revolution and reorganization of conventional healthcare delivery methods. Digital transformation influences the efficiency and quality of healthcare services, with notable variations in the timing of these effects. Consequently, the analysis indicates that the findings of this study align with existing research, affirming the scientific validity and rationale of this investigation.

Based on the above results, three policy recommendations are proposed: First, advocate for the implementation of EMR systems. Medical institutions must proactively acquire knowledge of and adopt EMR systems to get full coverage and transcend the constraints of conventional medical record systems. The utilization of EMR can enhance the precision and timeliness of medical data while mitigating the human and financial burdens associated with outdated information. Second, enhance digital training for healthcare professionals and enforce governmental legislation and policy directives. Emphasis must be made on enhancing employees’ digital literacy and advancing their digital competencies through training, integrating medical expertise with digital technology to facilitate grassroots digital transformation. Third, enhancing information sharing and collaboration among medical institutions is essential to mitigate regional disparities in development. The government should encourage medical institutions to establish medical big data platforms, mobile medical services and telemedicine services, thereby facilitating the digital transformation of grassroots medical facilities and improving the overall quality of medical services across all areas.

This study explores ways to improve the quality of medical services from the digital transformation perspective. This study may have the following three marginal contributions: First, this study reviews and analyses the quality of healthcare services in 30 Chinese provinces, adding a fresh viewpoint and creativity to previous research that can assist policymakers in gaining a thorough grasp of China's current healthcare service quality. Second, assessing how to increase healthcare service quality via the lens of digital transformation strengthens the present research system and provides theoretical support for substantially improving healthcare services. Third, employing the spatial Dobin model to investigate the spatial spillover effects of digital transformation on healthcare service quality facilitates a more accurate comprehension of their relationship, thereby enhancing the research perspective on digital transformation and healthcare service quality. Although this study has done a great deal of work on digital transformation and explored the direction of evolution and improvement paths for healthcare service quality, it still has the following three limitations and prospects: First, the research sample for this study needs to be refined. This study solely examined provincial-level data to understand the relationship between digital transformation and healthcare service quality from a macro perspective, failing to adequately grasp the relationship between the two at the prefecture or county levels. As a result, in future studies, the sample can be refined and evaluated from a micro perspective using prefecture-level cities or economic zones to assess the influence of digital transformation on healthcare service quality. Second, this study solely looked into digital transformation's direct and spatial spillover impacts on healthcare service quality. However, the influence of digital transformation on healthcare service quality does not follow a straight line. As a result, future studies could use mediation and moderation effect models to investigate the mechanisms by which digital transformation affects healthcare service quality. Third, the study period of this research mainly covers 10 years from 2012 to 2021, which includes the SARS-CoV-2 pandemic. The SARS-CoV-2 pandemic may have caused data quality issues, but this study did not explore this issue. As a result, in future studies, the SARS-CoV-2 pandemic can be used as a virtual variable to acquire a more detailed understanding of the influence of digital transformation on healthcare service quality.

Conclusion

This study employed KDE, Moran's index and SDM s to investigate the influence of digital transformation on the quality of medical services, leading to the following conclusions: First, digital change is spreading throughout all regions, with notable geographical distribution characteristics. There are inequalities in development among regions, and the gap is growing year after year. Second, while the quality of medical services has improved year after year, regional inequalities persist, following a spatial pattern. Higher-level medical services are primarily provided in Xinjiang, Sichuan, the Beijing-Tianjin-Hebei region, Guangdong Province and other locations. Third, digital transformation has a considerable positive effect on the quality of local medical services while also harming surrounding regions.

In addition, this study expands on the current literature by investigating the improvement and evolution of regional healthcare service quality via the lens of digital transformation across 30 Chinese provinces, providing fresh insights and possibilities for future research. Furthermore, this study examines the relationship between digital transformation and healthcare service quality from multiple perspectives, including spatial-temporal evolutionary characteristics and spatial spillover effects. This research helps policymakers understand the relationship between the two, enabling them to develop targeted, regionally tailored development policies while providing practical recommendations.

Footnotes

Acknowledgements

We are very grateful to all authors for their contributions to this manuscript, to Wan Mohd Hirwani Wan Hussain for his suggestions and improving on this manuscript. We also thank the editors and reviewers for their helpful comments.

ORCID iDs

Luxin Zhang

Wan Mohd Hirwani Wan Hussain

Sawal Hamid Md Ali

Author contributions

This work was done collaboratively by all authors. LXZ designed the study, performed the data collection, statistical analysis, wrote the protocol and wrote the first draft of the manuscript. WMHWH suggested changes to the article and contributed as corresponding author. SHMA revised the article's formatting and spelling. All the authors read and approved the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

Data supporting this study's findings are available in public databases, such as the China Statistical Yearbook (https://data.stats.gov.cn/publish.htm？sort=1); EPS Data Platform (https://www.epsnet.com.cn/index.html#/Index); China Medical Construction Statistics Yearbook (https://www.nhc.gov.cn/mohwsbwstjxxzx/tjzxtjsj/tjsj_list.shtml); China Environmental Statistics Yearbook (). Data supporting these findings are also available from the corresponding author.

References

Nie

Wang

. Service sector agglomeration and industrial structure optimisation: evidence from China's resource-based cities. Asia Pac Econ Lit 2024. DOI: 10.1111/apel.12414

Nie

Wang

. Spatial effects of service industry's heterogeneous agglomeration on industrial structure optimization: evidence from China. Systems 2024; 12. DOI: 10.3390/systems12030085

Shu

Chen

Yang

, et al. Investigation and countermeasures research of hospital information construction of tertiary class-A public hospitals in China: questionnaire study. JMIR Form Res 2023; 7. DOI: 10.2196/41820

Zhang

, et al. Research on the market development path of intelligent medical industry under the background of aging. In: Proceedings of the 2018 4th international conference on humanities and social science research (ICHSSR 2018). 2018, pp.685–688.

Huang

Chou

. Examining the dilemma between artificial intelligence techniques and professional medical service: a hybrid balancing perspective. IT Prof 2023; 25: 36–40.

Basile

Carbonara

Panniello

, et al. The role of big data analytics in improving the quality of healthcare services in the Italian context: the mediating role of risk management. Technovation 2024; 133. DOI: 10.1016/j.technovation.2024.103010

Zhao

Hou

. The development strategy of dual-channel supply chain of smart elderly care service from the perspective of time perception. Complexity 2024; 2024. DOI: 10.1155/2024/4957463

Sakurada

Ishikawa

Oba

, et al. Medical AI and AI for medical sciences. JMA J 2025; 8: 26–37.

Nguyen

Tariq

Hinchcliff

, et al. Why do consumers choose private over public health services? Reflective accounts of health providers in Vietnam. BMC Health Serv Res 2023; 23. DOI: 10.1186/s12913-023-09892-9

10.

Leal

. Public health and economic development: the case of Medicaid. J Public Health 2023; 45: E386–E386.

11.

Chen

C-C

Hsiao

C-T

Chang

D-S

, et al. The delivery model of perceived medical service quality based on donabedian's framework. J Healthc Qual: Off Publicat Natl Assoc Healthc Qual 2024; 46: 150–159. Journal Article, Research Support, Non-U.S. Gov't.

12.

Belevitin

Miroshnichenko

Goriachev

, et al. Modernization of organization base of technical service and repair of medical technique. Voen Med Zh 2010; 331: 4–11. English Abstract, Journal Article.

13.

Zhou

Guo

. Research on the doctors’ win in crowdsourcing competitions: perspectives on service content and competitive environment. BMC Med Inform Decis Mak 2023; 23. DOI: 10.1186/s12911-023-02309-x

14.

Rotaru

Edelhauser

. Digital transformation: a challenge for the Romanian health system. Systems 2024; 12. DOI: 10.3390/systems12090366

15.

Meshcheryakova

Udin

Demchenko

, et al. Population inclusion in the digitalization of healthcare. J Soc Policy Stud 2023; 21: 661–676.

16.

Qian

Wang

Huang

, et al. Bounded rationality in healthcare: unraveling the psychological factors behind patient satisfaction in China. Front Psychol 2024; 15. DOI: 10.3389/fpsyg.2024.1296032

17.

Drazen

Hansen

. The electronic patient safety net: the digital transformation of patient care. Healthplan 2004; 45: 52–56. Journal Article.

18.

Barik

Misra

Chockalingam

, et al. Data analytics, digital transformation, and cybersecurity perspectives in healthcare. In: Abie H, Gkioulos V, Katsikas S, (eds) Secure and resilient digital transformation of healthcare, sunrise 2023. Cham: Springer Nature Switzerland, 2024, pp.71–89.

19.

Ren

Jiang

Lou

, et al. Discussion on construction framework and key content of integrated medical internet of things. Zhongguo Yi Liao Qi Xie Za Zhi 2022; 46: 686–690. English Abstract, Journal Article.

20.

Zhi

Yin

Ren

, et al. Running an internet hospital in China: perspective based on a case study. J Med Internet Res 2021; 23. DOI: 10.2196/18307

21.

Inampudi

Rajkumar

Gopi

, et al. Barriers to implementation of digital transformation in the Indian health sector: a systematic review. Humanit Soc Sci Commun 2024; 11. DOI: 10.1057/s41599-024-03081-7

22.

Choi

Kim

Song

, et al. Development of microsurgical robot arm using spherical serial RCM mechanism and commercial laparoscopic surgical robot instrument. In: 2022 19th international conference on ubiquitous robots (UR). 2022, pp.159–164.

23.

Yang

Hou

Tang

, et al. Development of surgical robots in recent years. Zhongguo Yi Liao Qi Xie Za Zhi 2023; 47: 1–12. English Abstract, Journal Article.

24.

Furr

Ozcan

Eisenhardt

. What is digital transformation? Core tensions facing established companies on the global stage. Glob Strat J 2022; 12: 595–618.

25.

Ahlskog

Badasjane

Granlund

, et al. Differing views of the meaning of digital transformation in manufacturing industry. In: Ng AHC, Syberfeldt A, Högberg D, et al. (eds). SPS 2022 Amsterdam: IOS Press, 2022, pp.331–340.

26.

Barthel

What is meant by digital transformation success? Investigating the notion in IS literature. In: Ahlemann F, Schütte R and Stieglitz S (eds) Innovation through information systems, vol III: a collection of latest research on management issues. Cham: Springer Nature, 2021, pp.167–182.

27.

Ubiparipovic

Matkovic

Pavlicevic

. Key activities of digital business transformation process. Strat Manag 2022. DOI: 10.5937/StraMan2200016U

28.

Nguyen

Nhu

, et al. Does increased digital transformation promote a firm's financial performance? New insights from the quantile approach. Financ Res Lett 2024; 64. DOI: 10.1016/j.frl.2024.105430

29.

Dou

Guo

Chang

, et al. Corporate digital transformation and labor structure upgrading. Int Rev Financ Anal 2023; 90. DOI: 10.1016/j.irfa.2023.102904

30.

Han

Yang

Ying

, et al. The impact of corporate digital transformation on labor employment. Financ Res Lett 2024; 60. DOI: 10.1016/j.frl.2023.104888

31.

Wade

Shan

. Covid-19 has accelerated digital transformation, but may have made it harder not easier. MIS Quart Exec 2020; 19: 213–220.

32.

Gabryelczyk

. Has COVID-19 accelerated digital transformation? Initial lessons learned for public administrations. Inf Syst Manage 2020; 37: 303–309.

33.

Thakkar

Thanki

Guru

. A quantitative framework for health-care service quality assessment in India. J Mod Manage 2023; 18: 1064–1092.

34.

Gao

Kou

. Research on the identification of medical service quality factors: based on a data-driven method. Internet Res 2022; 32: 1617–1645.

35.

Yang

. Service quality evaluation of medical caring and nursing combined institutions for the aged based on IVPFS-DEMATEL and two-stage decision model with grey synthetic measures. J Grey Syst 2022; 34: 154–172.

36.

Rehman

Abidi

Usmani

, et al. Development of an intuitive GUI-based fuzzy multi-criteria decision model for comprehensive hospital service quality evaluation and indexing. Axioms 2023; 12. DOI: 10.3390/axioms12100921

37.

Mosadeghrad

. Factors affecting medical service quality. Iran J Public Health 2014; 43: 210–220.

38.

Basile

Carbonara

Panniello

, et al. How can technological resources improve the quality of healthcare service? The enabling role of big data analytics capabilities. IEEE Trans Eng Manage 2024; 71: 5771–5781.

39.

Bhadula

Sharma

Personalized self adaptive Internet-of-Things enabled sustainable healthcare architecture for digital transformation. In: 2023 advanced computing and communication technologies for high performance applications, ACCTHPA. 2023.

40.

Gungoren

. Crossing the chasm: strategies for digital transformation in clinical laboratories. Clin Chem Lab Med 2023; 61: 570–575.

41.

Ford

Tyler

Johnston

, et al. Challenges encountered and lessons learned when using a novel anonymised linked dataset of health and social care records for public health intelligence: the Sussex integrated dataset. Information 2023; 14. DOI: 10.3390/info14020106

42.

Shriedeh

. Strategic factors for building brand equity: Jordan medical tourism. J Messenger 2019; 11: 121–137.

43.

The innovation research of the medical service quality based on big data. In: 2nd international conference on computer engineering, information science and internet technology, CII 2017. 2017, pp.91–97.

44.

Mazilu

Anghel

. Intelligent patient management for improving quality of medical services. Univ Politech Buchar Sci Bull Ser C-Electr Eng Comput Sci 2021; 83: 73–84.

45.

Ellström

Holtstrom

Berg

, et al. Dynamic capabilities for digital transformation. J Strat Manag 2022; 15: 272–286.

46.

Zhou

Chen

, et al. Research on the construction of medical service quality evaluation Index system in the perspective of patients. In: 2010 international conference on management science and engineering (ICMSE). 2010, pp.305–310.

47.

Gong

Zhang

Wang

, et al. Analysis of regional differences, dynamic evolution, and influencing factors of medical service levels in Guangzhou under the health China strategy. Risk Manag Healthc Policy 2024; 17: 2811–2828.

48.

Wang

Shao

. Design of a smart medical service quality evaluation system based on a hybrid multi-criteria decision model. Sci Rep 2024; 14. DOI: 10.1038/s41598-024-71224-6

49.

Ding

Yang

Tian

, et al. Research and application of medical service analysis and evaluation system based on big data analysis in hospitals. In: Proceedings of 2024 international conference on smart healthcare and wearable intelligent devices, SHWID 2024. 2024, pp.104–107.

50.

Liao

, et al. Influence of blood sampling service process reengineering on medical services supply: Quasi-experimental study. J Med Internet Res 2024; 26. DOI: 10.2196/51412