The impact of informatization development on healthcare services in China
Variables and measurements
Dependent variable: healthcare service level (medi)
Measuring the level of healthcare services is a complex process that not only reflects the quality and efficiency of services provided by the healthcare system at a specific time, place, and under certain conditions but also serves as an indicator of the rationality of healthcare resource allocation and the effectiveness of healthcare work. Focusing solely on a single dimension may fail to comprehensively reflect the true level of healthcare services.
To measure healthcare service levels, Kruk et al. introduced a framework with 18 core indicators covering aspects such as safety, effectiveness, and a people-centered approach22. Similarly, Erdebilli et al. proposed frameworks that include factors like patient satisfaction, medical staff expertise, equipment quality, and financial performance23. Additionally, scholars like Pan et al. and Tan et al. have assessed healthcare service levels in terms of human, material, and financial resources, respectively24,25. Drawing on the relevant literature and considering the unique context of China’s healthcare system, this study integrates these indicators into a comprehensive framework and evaluates healthcare services across three key dimensions: human resources, material resources, and financial resources.
In the realm of human resources, this study focuses on several key indicators, including the number of health technicians per 10,000 individuals, the frequency of visits to healthcare institutions, and the volume of inpatient surgeries conducted. These indicators are instrumental in evaluating not only the quantity but also the professional competencies and service capabilities of healthcare personnel. By analyzing these factors, we can gain insights into the rationality and adequacy of human resource allocation within the healthcare system.
In terms of material resources, this study examines several key indicators, including the number of healthcare institutions, the number of hospital beds per 10,000 individuals, and the hospital bed utilization rate. These metrics provide valuable insights into the distribution of medical facilities, the capacity of services offered, and the efficiency with which resources are utilized. By evaluating these indicators, we can better understand the overall infrastructure of healthcare services and identify areas for improvement.
Regarding financial resources, local fiscal healthcare expenditure is chosen as its key indicator. By focusing on this indicator, we can evaluate how funding levels impact healthcare outcomes, efficiency, and the overall effectiveness of community services. This approach can provide a clearer understanding of the relationship between financial investment and health system performance, enabling the identification of best practices and highlighting areas for improvement.
Independent variable: informatization development level (info)
Although previous studies on the evaluation system for China’s informatization development level reveal some variations, most reference the framework established in the national “12th Five-Year Plan” for informationization26. This system, which originated from the “11th Five-Year Plan,” was designed to comprehensively assess and reflect the overall level of informatization development in a country or region27. It takes into account various factors, including the construction of information infrastructure, the degree of information application, environmental constraints, and residents’ information consumption patterns. During the preparation of the “12th Five-Year Plan,” these indicators were further refined, resulting in the selection of metrics across several dimensions: infrastructure, industrial technology, application consumption, knowledge support, and development outcomes.
In this study, an evaluation system was developed based on the framework established in China’s “12th Five-Year Plan” for informationization. This system encompasses three dimensions: information infrastructure, information service consumption, and the information industry. Indicators for information infrastructure include the mobile phone penetration rate and the number of broadband Internet access users per 10,000 people. Information service consumption is assessed through per capita telecommunications service volume. The level of the information industry is evaluated using two indicators: the number of urban employed personnel in software, information transmission, and information technology services per 10,000 people, as well as the number of domestic patent applications accepted per 10,000 people, which reflect human capital and information technology involvement, respectively.
Control variables
To ensure a comprehensive analysis of healthcare services, several control variables were included in this study to account for the influence of other factors. These variables encompass government expenditure, the level of urbanization, population density, and the degree of extroversion to the outside world. By incorporating these indicators, we aim to provide a more nuanced understanding of the various elements that affect healthcare service delivery.
Government Expenditure (Gove) was measured by the proportion of local fiscal healthcare expenditure compared to the total government spending. This ratio highlights the commitment of local governments to healthcare services relative. A higher proportion may indicate a stronger focus on improving healthcare access and quality, while a lower proportion could suggest competing fiscal demands or underinvestment in healthcare. Analyzing this measure can help identify disparities in healthcare funding and inform strategies for optimizing resource allocation to enhance service delivery.
Urbanization Level (Urban) was evaluated by the percentage of a population living in urban areas compared to the total population. Regions with higher urbanization rates typically boast more developed healthcare infrastructure, a greater number of healthcare providers, and broader coverage of healthcare services. As a result, urban residents generally enjoy easier access to high-quality healthcare compared to their rural counterparts. However, rapid urbanization can place significant strain on urban healthcare systems, particularly in the realms of public health and emergency services. This dynamic underscores the need for effective planning and resource allocation to ensure that healthcare systems can adequately meet the demands of growing urban populations.
Population Density (Popu) was measured by dividing the total population of an area by the land area it occupies. Areas with higher population density often experience greater demands for healthcare services, which can pose significant challenges to healthcare systems. This increased density may lead to resource constraints in healthcare, heighten the risk of infectious disease transmission, and contribute to environmental pollution, all of which can indirectly impact residents’ health. Consequently, effective management of healthcare services in high-density areas is essential for enhancing service quality and ensuring the health and well-being of the population.
The Extroversion Degree (Exter) was measured by the ratio of foreign investment converted into RMB to GDP. Extroversion indicates the degree of engagement with the outside world, including international trade, foreign investment, and technology exchange. A higher degree of extroversion can facilitate the introduction and updating of medical technology and knowledge, thereby enhancing local healthcare service levels. For example, international cooperation projects can help introduce advanced medical equipment and management expertise, train healthcare professionals, and consequently improve the quality of healthcare services.
Model construction
Before constructing the model, we tested for endogeneity among the variables to ensure appropriate model selection and specification. Specifically, we conducted the Durbin-Wu-Hausman (DWH) test, which yielded a P-value of 0.273, indicating no significant endogeneity issues. We also assessed spatial effects across provinces using Moran’s I and Geary’s C indices, revealing significant spatial correlations among the variables across regions (Moran’s I = 0.562, p < 0.01; Geary’s C = 0.438, p < 0.01).
It is proposed that healthcare services are evolving dynamically, with the impact of informatization on these services continuously changing, and that both informatization development and healthcare services exhibit strong spatial characteristics. In this context, relying solely on a static panel model may distort the effects, making a dynamic panel model more suitable. Additionally, to better understand the influence of informatization development in other regions on local healthcare services, spatial lag terms for the informatization development indicators are incorporated to assess these spatial interactions. Consequently, the model specification for this study is set as follows:
$$\begingathered Medi_it = \beta Medi_it – 1 + \rho \sum\nolimits_j = 1^n W_ij Medi_it + \gamma Info_it + \eta W_ij Info_it + \delta X_ijt + \alpha_i + \nu_it + \varepsilon_it \hfill \\ \varepsilon_it = \lambda \sum\nolimits_j = 1^n W_ij \varepsilon_it + \mu_it \hfill \\ \endgathered$$
In the equations provided, \(Info_it\) represents the explanatory variable of informatization development level, \(Medi_it\) represents the explained variable of healthcare services, \(X_ijt\) denotes the selected control variables, \(W_ij\) denotes the spatial weight matrix, where weights are represented by the reciprocal of geographical distances, and \(\varepsilon_it\) represents the random disturbance term.
The evaluation of both informatization development level and healthcare service level indicators employs the entropy weight-TOPSIS method. This method utilizes information entropy to calculate the weights of each indicator, thereby mitigating the subjective weighting flaws and enhancing objectivity and fairness. The entropy weight-TOPSIS method provides a comprehensive reflection of performance across various system layers. Furthermore, this method is computationally straightforward, without necessitating complex mathematical derivations, and has been extensively applied across multiple research domains.
Before conducting the analysis with the dynamic spatial panel model, we evaluated the model’s suitability through such standard procedures as LM, LR, and Wald tests. The LM test statistics were significant at the 5% level, confirming the appropriateness of the spatial panel model for this study. The LR test statistics were significant at the 1% level, indicating that the dynamic spatial panel model could not be simplified to a SAR or SEM model. The Wald test also demonstrated significance at the 1% level, underscoring the superiority of the dynamic spatial panel model over both SAR and SEM models.
Data collection
To ensure the accessibility and reliability of the research data, this study primarily utilized publicly available data from the “China Statistical Yearbook” published by the National Bureau of Statistics, as well as data provided by provincial statistical bureaus. The research sample includes 31 provinces in mainland China, with the study period covering the years 2010 to 2022, accounting for potential lags in data publication. During data processing, some gaps were identified and addressed using interpolation techniques to maintain data integrity.
Regional divisions in the study followed the classification outlined in the “China Statistical Yearbook 2023,” categorizing the country into Eastern, Central, Western, and Northeastern regions. The Eastern region comprises 10 provinces, including municipalities directly under the central government; the Central region consists of 6 provinces; the Western region includes 12 provinces, encompassing autonomous regions and municipalities directly under the central government; and the Northeastern region consists of 3 provinces.
link
