A method for analyzing the spatial pattern of out-of-town patient visit rate

Abstract

The embodiment of the present invention relates to an analysis method of the spatial pattern of the out-of-town patient's medical treatment rate, the method comprising: obtaining case data, geographic element data and socioeconomic data of out-of-town patients seeking medical treatment in a target city; determining multiple first influencing factors of out-of-town patients seeking medical treatment based on the case data, geographic element data and socioeconomic data; screening multiple first influencing factors using a multiple regression model to obtain multiple second influencing factors; the outflow of patients from a prefecture-level city to a target city's medical treatment rate is a dependent variable, and multiple second influencing factors are independent variables, and fitting analysis is performed on the dependent variable and the independent variable using an OLS regression model, a GWR regression model and an MGWR regression model to determine the target regression model with the highest degree of fit; based on the significance test variable of the target regression model, determining the target influencing factor from the second influencing factors to analyze the spatial pattern of the medical treatment rate. The technical solution provided by the present invention can accurately analyze the spatial pattern of the out-of-town patient's medical treatment rate.

Description

Technical Field

[0001] The embodiments of the present invention relate to the interdisciplinary fields of geography, health statistics, health management, and mathematics, and in particular to an analysis method for the spatial pattern of out-of-town patient visit rates. Background Art

[0002] Non-local patients seeking medical treatment is a common medical phenomenon, reflecting the status of medical reform, legislative changes and development of the medical care system in a country or region. In China, non-local patients seeking medical treatment usually refers to a medical treatment behavior in which the medical treatment location of the medical insurance participant is separated from the insured area.

[0003] In related technologies, the spatial pattern of out-of-town patient visit rates is usually analyzed based on the Ordinary Least Squares (OLS) method, which implies the assumption of spatial homogeneity and assumes that the regression coefficients of each sample point are equal. However, due to the spatial aggregation characteristics of some regions, the relationship between the independent variables and the dependent variables of each sample point is spatially non-stationary. The global regression model represented by the OLS model cannot measure the differential impact of independent variables in different regions on the dependent variable.

[0004] Therefore, there is an urgent need for an analytical method for the spatial pattern of out-of-town patient visit rates to solve the above technical problems. Summary of the invention

[0005] In order to accurately analyze the spatial pattern of the out-of-town patient consultation rate, an embodiment of the present invention provides a method for analyzing the spatial pattern of the out-of-town patient consultation rate.

[0006] The embodiment of the present invention provides a method for analyzing the spatial pattern of the out-of-town patient visit rate, including:

[0007] Obtain case data, geographic element data, and socioeconomic data of non-local patients seeking medical treatment in the target city; wherein the target city is the prefecture-level city where the patients flow in, and the prefecture-level city where the non-local patients are located is the prefecture-level city where the patients flow out;

[0008] Based on the case data, the geographic element data and the socioeconomic data, determining multiple first influencing factors for non-local patients seeking medical treatment;

[0009] Using a multiple regression model to screen a plurality of the first influencing factors to obtain a plurality of second influencing factors;

[0010] Taking the rate of patients flowing out of the prefecture-level city to the target city as the dependent variable and taking the plurality of the second influencing factors as the independent variables, respectively using the OLS regression model, the GWR regression model and the MGWR regression model to perform fitting analysis on the dependent variable and the independent variable, and determining the target regression model with the highest fitting degree;

[0011] Based on the significance test variable of the target regression model, a target influencing factor is determined from a plurality of the second influencing factors.

[0012] In one possible design, the case data includes the prefecture-level city where the patient flows out, medical insurance payment type, age, and disease type;

[0013] The geographic element data includes the time data and distance data of patients flowing out of the prefecture-level city to the target city by car, train and plane, the distance data to other medical centers, the distance data to the provincial capital city and the administrative division data; wherein the other medical centers are obtained by performing spatial autocorrelation analysis on the distribution of the number of patients flowing out of the prefecture-level city to the target city;

[0014] The socioeconomic data include medical statistics, demographic statistics and regional GDP data. The medical statistics include the number of medical institutions with patients flowing out of the prefecture-level city, the number of actual beds, the number of practicing physicians and the number of tertiary hospitals.

[0015] In a possible design, the spatial autocorrelation analysis uses the global Moran's I index and the local Moran's I index to analyze the spatial correlation and difference of the distribution of the number of patients flowing out of the prefecture-level city to the target city;

[0016] The global Moran's I index is calculated by the following formula:

[0017]

[0018] In the formula, is the total number of regional units analyzed; and are the number of medical visits from the spatial object in the i-th and j-th regional units to the target city, for The average value of is the weight matrix, Used to represent the link relationship between the i-th and j-th regional units of a spatial object; Indicates positive spatial autocorrelation; Indicates negative spatial autocorrelation; Indicates that there is no spatial autocorrelation;

[0019] The local Moran's I index is calculated by the following formula:

[0020]

[0021] In the formula, for The discrete variance of is the total number of regional units analyzed; and are the number of medical visits from the spatial object in the i-th and j-th regional units to the target city, for The average value of is the weight matrix, Used to represent the link relationship between the i-th and j-th regional units of a spatial object; Indicates positive spatial autocorrelation; Indicates negative spatial autocorrelation; Indicates that there is no spatial autocorrelation.

[0022] In a possible design, the first influencing factors include the regional basic medical supply level, the number of tertiary hospitals, patient affordability, transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, regional population density, per capita GDP of the prefecture-level city where patients flow out, the distance from the prefecture-level city where patients flow out to its provincial capital, and the average age of patients;

[0023] The multiple first influencing factors for out-of-town patients seeking medical treatment are determined based on the case data, the geographic element data and the socioeconomic data, including:

[0024] The number of medical institutions, actual number of beds and number of practicing physicians from which patients flow out of the prefecture-level city is weighted and summed to obtain the basic medical supply level of the region;

[0025] Determine the affordability of the patients based on the proportion of medical insurance payments from the prefecture-level cities where the patients flow out;

[0026] The time data of patients leaving the prefecture-level city to the target city by car, train and plane are weighted summed to obtain the traffic accessibility of patients leaving the prefecture-level city to the target city;

[0027] Determine the attractiveness of the other medical centers based on the distance from the prefecture-level city where patients flow out to the target city and the distance from the prefecture-level city where patients flow out to the other medical centers closest to it;

[0028] Based on the demographic data and land area of ​​the prefecture-level city where the patients flow out, determine the population density of the area;

[0029] Based on the demographic data and gross regional product data of the prefecture-level city where patients are flowing out, the per capita GDP of the prefecture-level city where patients are flowing out is determined.

[0030] In a possible design, the second influencing factors include the regional basic medical supply level, the number of tertiary hospitals, the transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, the per capita GDP of the prefecture-level city where patients flow out, the distance from the prefecture-level city where patients flow out to its provincial capital, and the average age of patients.

[0031] In a possible design, the target regression model is a MGWR regression model;

[0032] The target influencing factors include the number of tertiary hospitals, the transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, the per capita GDP of the prefecture-level city where patients flow out, and the distance from the prefecture-level city where patients flow out to its provincial capital;

[0033] The rate of patients visiting the target city out of the prefecture-level city is significantly positively correlated with the traffic accessibility from the prefecture-level city out of the patient to the target city, the per capita GDP of the prefecture-level city out of the patient, and the distance from the prefecture-level city out of the patient to its provincial capital city.

[0034] The rate of patients flowing out of prefecture-level cities to the target cities is significantly negatively correlated with the number of tertiary hospitals and the attractiveness of other medical centers.

[0035] In a possible design, the distance from which the patients flow out of the prefecture-level city to the target city is less than a preset distance, and the preset distance is obtained by analyzing the number of out-of-town medical visits at different distances from the target city.

[0036] In one possible design, it also includes:

[0037] For each type of non-local patient of different age groups and different medical insurance payment types in the target city, multiple preset distance decay models are used to fit the number of visits of each type of non-local patient in the target city and the distance from the prefecture-level city where the patient flows out to the target city, and the regression coefficient and distance decay coefficient of each distance decay model are obtained;

[0038] The distance attenuation model with the largest regression coefficient is used as the target distance attenuation model, and the distance attenuation coefficient of the target distance attenuation model is used as the distance attenuation coefficient of each type of out-of-town patient.

[0039] In one possible design, the distance decay model includes a square root exponential model, an exponential model, a square exponential model, a Pareto model, and a constant logarithm model.

[0040] In a possible design, the target distance attenuation model is a Pareto model.

[0041] The embodiment of the present invention provides an analysis method for the spatial pattern of the out-of-town patient's medical treatment rate. By obtaining the case data, geographic element data and socioeconomic data of the out-of-town patients seeking medical treatment in the target city, multiple first influencing factors of the out-of-town patient's medical treatment can be determined based on the case data, geographic element data and socioeconomic data. The multiple first influencing factors can be further screened using a multivariate regression model to obtain multiple second influencing factors; then, the patient's outflow rate from the prefecture-level city to the target city is taken as the dependent variable and the multiple second influencing factors are taken as independent variables, and the OLS regression model, GWR regression model and MGWR regression model are used to perform fitting analysis on the dependent variable and the independent variable to determine the target regression model with the highest fitting degree; in this way, the target influencing factor can be determined from the multiple second influencing factors based on the significance test variable of the target regression model to analyze the target influencing factor, so as to accurately analyze the spatial pattern of the out-of-town patient's medical treatment rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 It is a flow chart of a method for analyzing the spatial pattern of out-of-town patient visit rate provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] Please refer to Figure 1 , an embodiment of the present invention provides a method for analyzing the spatial pattern of the out-of-town patient visit rate, the method comprising:

[0046] Step 100: Obtain case data, geographic element data, and socioeconomic data of non-local patients seeking medical treatment in a target city; the target city is the prefecture-level city where patients flow in, and the prefecture-level city where non-local patients are located is the prefecture-level city where patients flow out;

[0047] Step 102: determining multiple primary influencing factors for out-of-town patients seeking medical treatment based on case data, geographic element data, and socioeconomic data;

[0048] Step 104: Screening the multiple first influencing factors using a multiple regression model to obtain multiple second influencing factors;

[0049] Step 106: Taking the rate of patients flowing out of the prefecture-level city to the target city as the dependent variable and the multiple second influencing factors as the independent variables, respectively use the OLS regression model, the GWR regression model and the MGWR regression model to perform fitting analysis on the dependent variable and the independent variable to determine the target regression model with the highest fitting degree;

[0050] Step 108: Based on the significance test variable of the target regression model, a target influencing factor is determined from the multiple second influencing factors to analyze the target influencing factor.

[0051] In an embodiment of the present invention, by obtaining case data, geographic element data and socioeconomic data of non-local patients seeking medical treatment in a target city, a variety of first influencing factors for non-local patients seeking medical treatment can be determined based on the case data, geographic element data and socioeconomic data, and a multivariate regression model can be further used to screen the various first influencing factors to obtain a variety of second influencing factors; then, the rate of patients flowing out of the prefecture-level city to the target city is taken as the dependent variable and the various second influencing factors are taken as independent variables, and the dependent variables and the independent variables are fitted and analyzed using the OLS regression model, the GWR regression model and the MGWR regression model respectively, to determine the target regression model with the highest degree of fit; in this way, the target influencing factors can be determined from the various second influencing factors based on the significance test variables of the target regression model to analyze the target influencing factors, so as to accurately analyze the spatial pattern of the non-local patients' medical treatment rate.

[0052] Described below Figure 1 How the various steps are performed.

[0053] For step 100:

[0054] Taking City A as the target city as an example, all prefecture-level cities except City A are called prefecture-level cities with patient outflow, and all patients except City A are called non-local patients. The case data selected in the embodiment of the present invention are the hospitalization case data of non-local patients from 154 hospitals in City A in 2015. After analyzing and processing the data, a total of 599,400 valid data of 341 prefecture-level patients who went to City A for medical treatment were included.

[0055] In some embodiments, the case data includes the prefecture-level city where the patient flows out, the type of medical insurance payment, age, and disease type;

[0056] Geographic element data include the time and distance data of patients leaving prefecture-level cities to reach target cities by car, train and plane, distance data to other medical centers, distance data to provincial capital cities and administrative division data; among them, other medical centers are obtained by performing spatial autocorrelation analysis on the distribution of the number of patients leaving prefecture-level cities to reach target cities;

[0057] Socioeconomic data include medical statistics, demographic statistics and regional GDP data. Medical statistics include the number of medical institutions with patients flowing out of prefecture-level cities, the number of actual beds, the number of practicing physicians and the number of tertiary hospitals.

[0058] See Table 1 for detailed description and sources:

[0059] Table 1 Data description and source

[0060]

[0061] It should be noted that spatial autocorrelation analysis is an analytical method to study whether the observed values ​​of a spatial unit are correlated with the observed values ​​of its adjacent units, and is used to measure the degree of aggregation of the observed values ​​of the spatial unit.

[0062] In some embodiments, the spatial autocorrelation analysis uses the global Moran's I index and the local Moran's I index to analyze the spatial correlation and difference of the distribution of the number of patients flowing out of the prefecture-level city to the target city;

[0063] The global Moran's I index is calculated by the following formula:

[0064]

[0065] In the formula, is the total number of regional units analyzed; and are the number of medical visits to City A by the spatial object in the i-th and j-th regional units, for The average value of is the weight matrix, Used to represent the link relationship between the i-th and j-th regional units of a spatial object; Indicates positive spatial autocorrelation; Indicates negative spatial autocorrelation; Indicates that there is no spatial autocorrelation;

[0066] The local Moran's I index is calculated by the following formula:

[0067]

[0068] In the formula, for The discrete variance of is the total number of regional units analyzed; and are the number of medical visits to City A by the spatial object in the i-th and j-th regional units, for The average value of is the weight matrix, Used to represent the link relationship between the i-th and j-th regional units of a spatial object; Indicates positive spatial autocorrelation; Indicates negative spatial autocorrelation; Indicates that there is no spatial autocorrelation.

[0069] The local Moran's I index subdivides the spatial association pattern into four types, which correspond to the four quadrants in the Moran scatter plot, including high-high aggregation areas (areas with high out-of-town medical visits are surrounded by surrounding areas with high out-of-town medical visits), high-low aggregation areas (areas with high out-of-town medical visits are surrounded by surrounding areas with low out-of-town medical visits), low-high aggregation areas (areas with low out-of-town medical visits are surrounded by surrounding areas with high out-of-town medical visits), and low-low aggregation areas (areas with low out-of-town medical visits are surrounded by surrounding areas with low out-of-town medical visits).

[0070] The global and local Moran's I indexes were calculated to detect the spatial autocorrelation between prefecture-level regions to identify the spatial differences in the flow of inpatients to City A. The results (calculation results omitted) showed that there was positive spatial autocorrelation in the number of patient visits between prefecture-level regions (I=0.57, p<0.01), showing a certain spatial agglomeration pattern.

[0071] The results of local spatial autocorrelation analysis (calculation results omitted) show that a high-high aggregation area appears in the prefecture-level regions around City A. Relatively speaking, low-low aggregation areas are mainly distributed in the southwest, south China, the coastal areas of east China, and parts of the northwest, including 104 prefecture-level regional units. The distance from City A and the local medical resources jointly affect the number of patients seeking medical treatment in City A.

[0072] Regarding step 102:

[0073] In some embodiments, the first influencing factor includes the regional basic medical supply level, the number of tertiary hospitals, patient affordability, transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, regional population density, per capita GDP of the prefecture-level city where patients flow out, the distance from the prefecture-level city where patients flow out to its provincial capital, and the average age of patients;

[0074] Based on case data, geographic element data, and socioeconomic data, we identified a variety of primary influencing factors for out-of-town patients seeking medical treatment, including:

[0075] The weighted sum of the number of medical institutions, actual number of beds, and number of practicing physicians from which patients flow out of prefecture-level cities was taken to obtain the regional basic medical supply level;

[0076] Determine patient affordability based on the proportion of medical insurance payments from patients leaving prefecture-level cities;

[0077] The time data of patients leaving the prefecture-level city to reach the target city by car, train and plane is weighted summed to obtain the traffic accessibility of patients leaving the prefecture-level city to the target city;

[0078] Determine the attractiveness of other medical centers based on the distance from the prefecture-level city where patients flow out to the target city and the distance from the prefecture-level city where patients flow out to the nearest other medical center;

[0079] Determine regional population density based on demographic data and land area of ​​prefecture-level cities where patients flow out;

[0080] Based on the demographic data and regional GDP data of the prefecture-level cities where patients outflow, the per capita GDP of the prefecture-level cities where patients outflow is determined.

[0081] Regarding step 104:

[0082] In some embodiments, the second influencing factors include the regional basic medical supply level, the number of tertiary hospitals, the transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, the per capita GDP of the prefecture-level city where patients flow out, the distance from the prefecture-level city where patients flow out to its provincial capital city, and the average age of patients.

[0083] In this implementation, the rate of non-local medical treatment from various municipal areas to City A is used as the dependent variable of the study to characterize the degree of utilization of medical services in City A by non-local patients. In order to accurately find the optimal influencing variables, a multivariate regression model is used to obtain the overall difference characteristics of the influencing factors, and screening is performed at the significance level (α=0.05). A total of 7 variables (i.e., the second influencing factors) passed the test (p<0.001), namely, GDP per capita, the number of local tertiary hospitals, the level of regional basic medical supply, transportation accessibility to City A, the distance to the provincial capital, the average age of patients, and the attractiveness of other medical centers. Among them, the final dependent and independent variables are shown in Table 2:

[0084] Table 2

[0085]

[0086] Regarding step 106:

[0087] The regression analysis results of the three models of OLS, GWR and MGWR are shown in Table 3. The results show that the performance of the MGWR model is better than that of the OLS and ordinary GWR models, reflecting the change of 82.7% in the rate of out-of-town medical treatment in City A. Compared with the OLS model (AICc: 793.204; SRS: 193.589) and the ordinary GWR model (AICc: 608.351; SRS: 79.098), the MGWR model generated a smaller modified Akaike information criterion value (AICc: 556.80) and residual sum of squares (SRS: 58.888), which indicates that the MGWR model has better fit and accuracy.

[0088] Table 3

[0089]

[0090] Note: ***, **, * are significant at 1%, 5%, and 10% levels, respectively.

[0091] The results show that there are 5 variables that meet the significance test ( ≥1.96), among which the rate of out-of-town medical treatment in City A was significantly positively correlated with GDP per capita, transportation accessibility to City A, and distance to the provincial capital, and was significantly negatively correlated with the number of local tertiary hospitals and the attractiveness of other medical centers.

[0092] In other words, the target regression model is the MGWR regression model;

[0093] The target influencing factors include the number of tertiary hospitals, the transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, the per capita GDP of the prefecture-level city where patients flow out, and the distance from the prefecture-level city where patients flow out to its provincial capital;

[0094] The rate of patients' medical treatment from prefecture-level cities to target cities is significantly positively correlated with the traffic accessibility from prefecture-level cities to target cities, the per capita GDP of prefecture-level cities, and the distance from prefecture-level cities to their provincial capitals.

[0095] The rate of patients flowing out of prefecture-level cities to target cities is significantly negatively correlated with the number of tertiary hospitals and the attractiveness of other medical centers.

[0096] In addition, through the analysis of the number of out-of-town medical treatments at different distances from the target city, it was found that the number of out-of-town medical treatments from prefecture-level areas to City A generally showed a decreasing gradient trend with the increase in the distance from City A, and more than 80% of the patients who went to City A for medical treatment came from prefecture-level areas less than 1,300 kilometers away from City A, indicating that there is a significant distance attenuation effect for out-of-town medical treatment in City A.

[0097] Therefore, in order to reduce analysis errors, in some embodiments, the distance from the prefecture-level city where patients flow out to the target city is less than a preset distance (e.g., 1,300 kilometers), and the preset distance is obtained by analyzing the number of out-of-town medical visits at different distances from the target city.

[0098] In some embodiments, the above method further comprises:

[0099] For each type of non-local patient of different age groups and different medical insurance payment types in the target city, multiple preset distance decay models are used to fit the number of visits of each type of non-local patient in the target city and the distance from the prefecture-level city where the patient flows out to the target city, and the regression coefficient and distance decay coefficient of each distance decay model are obtained;

[0100] The distance attenuation model with the largest regression coefficient is taken as the target distance attenuation model, and the distance attenuation coefficient of the target distance attenuation model is taken as the distance attenuation coefficient of each type of out-of-town patient.

[0101] The centroid of each city area is taken as the patient outflow point, and the distance attenuation calculation is performed with a step length of 100 kilometers. In the calculation process, since the cumulative proportion of the number of visits beyond 3,000 kilometers is only 0.48%, in order to avoid the adverse impact of the long-tail effect on the calculation results, the distance is limited to within 3,000 kilometers. This paper uses the cumulative distribution method to measure the distance attenuation. The results of five distance attenuation fitting functions are shown in Table 4. The results show that the Pareto function has the best fitting effect. Therefore, the embodiment of the present invention selects the Pareto function as the fitting function of the distance attenuation effect. That is, the target distance attenuation model is the Pareto model.

[0102] Table 4

[0103]

[0104] In some embodiments, the distance decay model includes a square root exponential model, an exponential model, a square exponential model, a Pareto model, and a constant logarithm model, and their formulas are:

[0105]

[0106] In the formula, I represents the number of out-of-town patients visiting City A from each city-level regional unit, d is the distance from the prefecture-level city where patients flow out to the target city, is a scalar factor ( is calculated in advance using historical data, that is, is a known quantity), β represents the distance attenuation coefficient. β reflects the speed of distance attenuation. The larger its value, the greater the impact of distance on the amount of medical treatment in other places.

[0107] In addition, the OLS (linear regression) method can be used to fit the regression coefficients of the linear model, and the determination coefficient R 2 Measures the goodness of fit of the model. R 2 is the part of the dependent variable variation explained by the regression model, i.e.

[0108] R 2 =ESS / TSS=1−RSS / TSS

[0109] Where TSS is the total sum of squares and ESS is the explained sum of squares.

[0110] Analyzing the differences in the medical treatment behaviors of different patient groups in other places helps to understand the fairness of medical resource allocation. The embodiment of the present invention analyzes the mobility patterns of different types of patient groups and compares groups of different ages and different types of medical payment. Analysis of different age groups found that (Table 5) showed that the distance attenuation effect showed an overall trend of first decreasing and then increasing with the increase of patient age, and different groups were affected to different degrees by the increase in distance. The elderly (≥75 years old) who went to City A for medical treatment in other places were most affected by the increase in distance (β=-0.633), while the middle-aged group (45-59 years old) was least affected by the increase in distance (β=-0.739). This result shows that the mobility of the elderly (≥75 years old) who went to City A for medical treatment in other places is most susceptible to the impact of distance, which is consistent with the current situation that the elderly are not convenient to move long distances due to their age and physical reasons.

[0111] Table 5

[0112]

[0113] The group analysis of different medical payment types found that (Table 6) the urban-rural differences in patient mobility patterns also reflect the differences in accessibility of medical services between urban and rural areas in my country and the geographical differences in the allocation of medical resources.

[0114] Table 6

[0115]

[0116] Regarding step 108:

[0117] Table 7 shows the differential effect scale of the MGWR regression model on each variable:

[0118] Table 7

[0119]

[0120] Based on the significance test results of the independent variables, the embodiment of the present invention determines five independent variables, such as GDP per capita, traffic accessibility to City A, distance to the provincial capital, number of local tertiary hospitals, and attractiveness of other medical centers, as important influencing factors for the study of the rate of out-of-town medical treatment in City A. In order to further explore the spatial pattern of the degree of influence of each influencing factor on the dependent variable, the embodiment of the present invention further applies the natural breakpoint method to analyze the changes in the dependent variable (the rate of out-of-town medical treatment from various municipal-level areas to City A) and the above five independent variables in different regions and different bandwidths, so as to quantitatively study the spatial heterogeneity of the influence of each influencing factor.

[0121] (1) Analysis of the impact of per capita GDP on the rate of medical treatment in City A

[0122] The results of the MGWR model show that GDP per capita is positively correlated with the rate of out-of-town medical treatment in City A, that is, the higher the GDP per capita, the higher the rate of out-of-town medical treatment in City A. As a measure of regional affluence, GDP per capita reflects the economic payment ability of a region's patient population.

[0123] The effect scale of GDP per capita is 183, indicating that it has an impact on the rate of patients seeking medical treatment in other places in City A at a medium scale. Following the reform direction of my country's medical system from "diseases have access to medical care" to "Healthy China", it is recommended that patients with common and common diseases give priority to local medical treatment, and relevant medical institutions should provide policy guidance, clarify the types of common and common diseases and the proportion of patients, and improve the level of related medical services for basic diseases; for other major diseases that cannot be solved by local medical treatment, it is recommended to increase medical subsidies for patients seeking medical treatment in other places, thereby increasing the right of patients in these areas to choose medical treatment in other places.

[0124] (2) Analysis of the impact of transportation accessibility to City A on the rate of medical treatment in City A from other places

[0125] Traffic accessibility to City A has a positive impact on the rate of out-of-town medical treatment in City A, that is, the higher the traffic accessibility to City A, the higher the out-of-town medical treatment rate in City A. Higher traffic accessibility increases the possibility of patients seeking medical treatment in other places to some extent.

[0126] The impact scale of traffic accessibility to City A is relatively small, with an effect scale of 59, indicating that it has an impact on the rate of patients seeking medical treatment in other places in City A at a relatively small scale. Since patients seeking medical treatment in other places need to spend more travel costs to seek high-quality medical treatment, it is necessary to create better transportation conditions for them, strengthen "telemedicine" cooperation with large hospitals, realize online medical treatment in other places, and provide more convenient medical services for patients who seek medical treatment in other places.

[0127] With the development of China's transportation, especially the construction of the "seven radial, nine vertical and eighteen horizontal" expressway network and the "eight vertical and eight horizontal" high-speed railway network, the flow of factors such as medical treatment and tourism in other places has been greatly promoted. The impact of transportation accessibility on the effect of patients' medical treatment in other places will also be further expanded.

[0128] (3) Analysis of the impact of the distance to the provincial capital on the rate of medical treatment in City A

[0129] The distance to the provincial capital has a positive impact on the rate of out-of-town medical treatment in City A, that is, the farther the distance to the provincial capital, the higher the rate of out-of-town medical treatment in City A. Its effect scale is 206, and the coefficient is relatively stable in space, which is close to affecting the rate of out-of-town medical treatment in City A on a global scale.

[0130] The regression coefficient of the distance to the provincial capital shows a circular outward decreasing gradient trend, and overall this positive impact is weak.

[0131] (4) Analysis of the impact of the number of local tertiary hospitals on the rate of out-of-town medical treatment in City A

[0132] The number of local tertiary hospitals has a negative impact on the rate of out-of-town medical treatment in City A, that is, the fewer local tertiary hospitals there are, the higher the out-of-town medical treatment rate in City A. The number of local tertiary hospitals reflects the high-quality medical resources in the region. The model results show that compared with the supply level of basic medical resources, patients seeking medical treatment in other places are greatly affected by the current status of local high-quality medical resource allocation.

[0133] The effect scale of the number of local tertiary hospitals is 150, indicating that it has an impact on the rate of patients seeking medical treatment in other places in City A at a medium scale. In health planning and medical center construction, the actual medical needs of patients in their mobility of medical treatment in other places should be fully considered, the establishment of tertiary hospitals should be promoted, policy guidance should be strengthened, and more patients should be encouraged to seek medical treatment nearby within the province.

[0134] (5) Analysis of the impact of the attractiveness of other medical centers on the rate of out-of-town medical treatment in City A

[0135] The attractiveness of other medical centers has a negative impact on the rate of out-of-town medical treatment in City A, that is, the stronger the attraction of other medical centers, the lower the out-of-town medical treatment rate in City A. For the same medical center, when patients are closer to other medical centers, they are more inclined to seek medical treatment nearby rather than traveling a longer distance to City A for treatment.

[0136] The effect scale of the attractiveness of other medical centers is 339, and the coefficient is relatively stable in space, indicating that it affects the rate of patients going to City A for medical treatment in other places on a global scale, and there is almost no spatial heterogeneity. The rate of patients going to City A for medical treatment in other places is basically affected by the attractiveness of other medical centers. The medical centers where patients from other places gather can provide important references for the construction of national medical centers and national regional medical centers, so that patients with medical needs in other places can receive effective treatment in a certain area, which can not only guide patients to seek medical treatment reasonably, but also help improve the overall and regional medical service levels in my country.

[0137] In summary, based on the cross-regional hospitalized patient data in City A, the embodiment of the present invention analyzes the spatial pattern of the flow of patients from various municipal-level regional units in China to City A for medical treatment, explores the spatial clustering characteristics and distance attenuation effects in detail, and uses the MGWR model to explore the influencing factors of the rate of out-of-town medical treatment from various municipal-level regional units to City A.

[0138] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a method for analyzing the spatial pattern of the out-of-town patient visit rate in any embodiment of the present invention is implemented.

[0139] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor executes a method for analyzing the spatial pattern of the out-of-town patient visit rate in any embodiment of the present invention.

[0140] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0141] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0142] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0143] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0144] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0145] It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the existence of other identical factors in the process, method, article or equipment including the elements.

[0146] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing the spatial pattern of out-of-town patient visit rate, characterized in that: include: Obtain case data, geographic element data, and socioeconomic data of non-local patients seeking medical treatment in the target city; wherein the target city is the prefecture-level city where the patients flow in, and the prefecture-level city where the non-local patients are located is the prefecture-level city where the patients flow out; Based on the case data, the geographic element data and the socioeconomic data, determining multiple first influencing factors for non-local patients seeking medical treatment; Using a multiple regression model to screen a plurality of the first influencing factors to obtain a plurality of second influencing factors; Taking the rate of patients flowing out of the prefecture-level city to the target city as the dependent variable and taking the plurality of the second influencing factors as the independent variables, respectively using the OLS regression model, the GWR regression model and the MGWR regression model to perform fitting analysis on the dependent variable and the independent variable, and determining the target regression model with the highest fitting degree; Determining a target influencing factor from a plurality of the second influencing factors based on the significance test variable of the target regression model; The case data include the prefecture-level city where the patient left, medical insurance payment type, age, and disease type; The geographic element data includes the time data and distance data of patients flowing out of the prefecture-level city to the target city by car, train and plane, the distance data to other medical centers, the distance data to the provincial capital city and the administrative division data; wherein the other medical centers are obtained by performing spatial autocorrelation analysis on the distribution of the number of patients flowing out of the prefecture-level city to the target city; The socioeconomic data include medical statistics, demographic data and regional GDP data. The medical statistics include the number of medical institutions with patients flowing out of the prefecture-level city, the number of actual beds, the number of practicing physicians and the number of tertiary hospitals; The first influencing factors include the regional basic medical supply level, the number of tertiary hospitals, patient affordability, transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, regional population density, per capita GDP of the prefecture-level city where patients flow out, the distance from the prefecture-level city where patients flow out to its provincial capital, and the average age of patients; The multiple first influencing factors for out-of-town patients seeking medical treatment are determined based on the case data, the geographic element data and the socioeconomic data, including: The number of medical institutions, actual number of beds and number of practicing physicians from which patients flow out of the prefecture-level city is weighted and summed to obtain the basic medical supply level of the region; Determine the affordability of the patients based on the proportion of medical insurance payments from the prefecture-level cities where the patients flow out; The time data of patients leaving the prefecture-level city to the target city by car, train and plane are weighted summed to obtain the traffic accessibility of patients leaving the prefecture-level city to the target city; Determine the attractiveness of the other medical centers based on the distance from the prefecture-level city where patients flow out to the target city and the distance from the prefecture-level city where patients flow out to the other medical centers closest to it; Based on the demographic data and land area of ​​the prefecture-level city where the patients flow out, determine the population density of the area; Based on the demographic data and regional gross product data of the prefecture-level city where the patient flows out, determine the GDP per capita of the prefecture-level city where the patient flows out; The second influencing factors include the level of regional basic medical care supply, the number of tertiary hospitals, the traffic accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, the per capita GDP of the prefecture-level city where patients flow out, the distance from the prefecture-level city where patients flow out to its provincial capital, and the average age of patients; The target influencing factors include the number of tertiary hospitals, the transportation accessibility from the prefecture-level city where patients flow out to the target city, the attractiveness of other medical centers, the per capita GDP of the prefecture-level city where patients flow out, and the distance from the prefecture-level city where patients flow out to its provincial capital; For each type of non-local patient of different age groups and different medical insurance payment types in the target city, multiple preset distance decay models are used to fit the number of visits of each type of non-local patient in the target city and the distance from the prefecture-level city where the patient flows out to the target city, and the regression coefficient and distance decay coefficient of each distance decay model are obtained; The distance attenuation model with the largest regression coefficient is taken as the target distance attenuation model, and the distance attenuation coefficient of the target distance attenuation model is taken as the distance attenuation coefficient of each type of out-of-town patient, so as to analyze the spatial pattern of the out-of-town patient visit rate based on the distance attenuation coefficient of each type of out-of-town patient and the target influencing factors.

2. The method according to claim 1, characterized in that The spatial autocorrelation analysis uses the global Moran's I index and the local Moran's I index to analyze the spatial correlation and difference of the distribution of the number of patients flowing out of the prefecture-level city to the target city; The global Moran's I index is calculated by the following formula: In the formula, is the total number of regional units analyzed; and are the number of medical visits from the spatial object in the i-th and j-th regional units to the target city, for The average value of is the weight matrix, Used to represent the link relationship between the i-th and j-th regional units of a spatial object; Indicates positive spatial autocorrelation; Indicates negative spatial autocorrelation; Indicates that there is no spatial autocorrelation; The local Moran's I index is calculated by the following formula: In the formula, for The discrete variance of is the total number of regional units analyzed; and are the number of medical visits from the spatial object in the i-th and j-th regional units to the target city, for The average value of is the weight matrix, Used to represent the link relationship between the i-th and j-th regional units of a spatial object; Indicates positive spatial autocorrelation; Indicates negative spatial autocorrelation; Indicates that there is no spatial autocorrelation.

3. The method according to claim 1, characterized in that The target regression model is a MGWR regression model; The rate of patients visiting the target city out of the prefecture-level city is significantly positively correlated with the traffic accessibility from the prefecture-level city out of the patient to the target city, the per capita GDP of the prefecture-level city out of the patient, and the distance from the prefecture-level city out of the patient to its provincial capital city. The rate of patients flowing out of prefecture-level cities to the target cities is significantly negatively correlated with the number of tertiary hospitals and the attractiveness of other medical centers.

4. The method according to any one of claims 1 to 3, characterized in that The distance from the prefecture-level city where the patients flow out to the target city is less than a preset distance, and the preset distance is obtained by analyzing the number of out-of-town medical visits at different distances from the target city.

5. The method according to claim 1, characterized in that The distance decay model includes a square root exponential model, an exponential model, a square exponential model, a Pareto model and a constant logarithm model.

6. The method according to claim 1, characterized in that The target distance attenuation model is a Pareto model.

Images