A hospital satisfaction survey method, device, medium, program product

Abstract

The application provides a hospital satisfaction survey method, device, medium and program product. First, patient basic information and patient questionnaire information are collected, so that the patient complaint probability and complaint items are predicted based on the collected patient basic information and patient questionnaire information by using a neural network model. Then, the predicted complaint situation and the actual complaint situation are comprehensively predicted, and the satisfaction survey questionnaire items are optimized. In one aspect, the application generates a predicted virtual complaint form by machine learning technology, so that the relevant functional departments can intervene in advance, and the patient satisfaction is improved. In another aspect, the questionnaire items of the subsequent diagnosis and treatment stage can be adaptively optimized and adjusted according to the past complaint situation of each patient, so that a professional, concise and reliable satisfaction survey questionnaire can be developed for different patients, the traditional survey questionnaire is avoided to be formalistic, and the effectiveness of the survey questionnaire is improved.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and specifically relates to a hospital satisfaction survey method, device, medium, and program product. Background Art

[0002] Patient satisfaction is a key indicator of hospital medical quality. For example, the "Performance Assessment Measures for Tertiary Public Hospitals" clearly stipulates that the "satisfaction index" should be weighted at 10% of the assessment of medical institutions. Therefore, patient satisfaction with the quality of hospital medical services is a crucial indicator for evaluating hospital management.

[0003] Currently, patients have numerous channels or methods for expressing their concerns, including convenient service hotlines, official WeChat accounts, mobile apps, and physical complaint mailboxes. Despite these numerous channels for patient feedback, their effective utilization is generally low. For one thing, most patients are reluctant to invest the extra effort to express their concerns. Furthermore, the questionnaires provided by hospitals often contain too many questions and lack specificity, which can lead patients to fill them out haphazardly. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides a hospital satisfaction survey method, device, medium, and program product to solve the problem of optimizing survey items in a questionnaire.

[0005] The present invention achieves the above technical objectives through the following technical means.

[0006] A hospital satisfaction survey method:

[0007] Collect basic patient information;

[0008] Collecting questionnaire information from patients, which includes their evaluation of their satisfaction with the hospital's services;

[0009] Based on the collected basic patient information and patient questionnaire information, a neural network model is used to predict the probability of patient complaints and complaint items.

[0010] Furthermore, a neural network model is used to predict the probability of patient complaints:

[0011] p=F P (C′,A′)

[0012] Where, F P (·) represents the complaint probability prediction function in the neural network model; C′={c′1,c′2,…,c′ N} and A′={a′1,a′2,…,a′ M} are the standardized results of the patient basic information feature set C and the patient questionnaire information feature set A, respectively, where:

[0013] Patient basic information feature set C = {c1, c2, ..., c N} is extracted from the patient's basic information, where c n is the eigenvalue of the nth basic information, n∈{1,2,…N};

[0014] Patient questionnaire information feature set A={a1,a2,…,a M} is extracted from the patient questionnaire information, where a m is the eigenvalue of the mth questionnaire information, m∈{1,2,…M}.

[0015] Furthermore, a neural network model is used to predict patient complaint items:

[0016] R * =F R (p,C′,A′)

[0017] Where, Predict complaint feature sets for patients, represents the predicted feature value of the x-th complaint item, which is used to characterize whether this item is a complaint, x∈{1,2,…X}; F R (·) represents the complaint item prediction function in the neural network model.

[0018] Furthermore, based on patient complaints, a neural network model was used to optimize the next questionnaire items:

[0019] Q=F Q (p,R′)

[0020] Where, Q={q1,q2,…,q M} is the questionnaire item feature set, where q m is the characteristic value of the mth questionnaire item, which is used to indicate whether this questionnaire item is included in the questionnaire; F Q (·) represents the questionnaire item optimization function in the neural network model; R′={r′1,r′2,…,r′ X} is the comprehensive complaint feature set of patients that combines actual complaints and predicted complaints, where:

[0021]

[0022] Where λ * and λ are the weights of the predicted eigenvalue and actual eigenvalue of the complaint item, respectively. x is the actual eigenvalue of the x-th complaint item.

[0023] Furthermore, the standardization method of the patient basic information feature set C and the patient questionnaire information feature set A is:

[0024]

[0025] Where n∈{1,2,…N}, m∈{1,2,…M}, c n_max and c n_min are the maximum and minimum values ​​of the basic information features of the nth patient, respectively. m_max and a m_min are the maximum and minimum values ​​of the information features of the mth patient questionnaire, ω m is the weighting factor of the characteristic value of the mth patient questionnaire information.

[0026] Furthermore, the neural network model is an XGBoost model.

[0027] Furthermore, the basic information of the patient includes: c1 age, c2 gender, c3 diagnosis result, c4 past medical history, c5 number, c6 current medical history, c7 interval between consultations, c8 complaint history, c9 history of diabetes, c 10 History of hypertension, 11 Whether surgery, c 12 Type of anesthesia, c 13 Diagnosis and treatment phase;

[0028] The questionnaire items include: q1 doctor service score, q2 doctor skill score, q3 nurse service score, q4 nurse skill score, q5 treatment process score, q6 registration service score, q7 examination service score, q8 inspection service score, q 10 Service desk rating, q 11 Payment service rating, q 12 Admission process score, q 13 Food rating, q 14 Anesthesia service score, q 15 Inpatient examination service score, q 16 Inpatient laboratory service score, q 17 Discharge settlement service score, q 18 Hospitalization experience score, q 19 Disease guidance score, q 20 Demand feedback score, q 21 Integrity in medical practice rating;

[0029] The complaint items include: r1 charging situation, r2 logistics support, r3 doctor service and technology, r4 nurse service and technology, r5 pharmacy service, r6 hospital management, r7 inspection service, and r8 service process.

[0030] A computer device comprising a memory and a processor;

[0031] The memory is used to store computer programs;

[0032] The processor is used to execute the computer program and implement the above-mentioned hospital satisfaction survey method when executing the computer program.

[0033] A computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to execute the above-mentioned hospital satisfaction survey method.

[0034] A computer program product includes a computer program, wherein the computer program implements the above-mentioned hospital satisfaction survey method when executed by a processor.

[0035] The beneficial effects of the present invention are:

[0036] (1) The present invention provides a hospital satisfaction survey method, device, medium, and program product, which is based on the Internet hospital platform and collects and integrates patient diagnosis and treatment data through the hospital information system, appointment calling system, charging system, and inspection information system, and combines the questionnaire content to generate predicted virtual complaint forms through machine learning technology, so that relevant functional departments can intervene in advance, thereby improving patient satisfaction.

[0037] (2) The present invention adaptively optimizes and adjusts the questionnaire items in the subsequent diagnosis and treatment stage based on each patient's past complaints, thereby formulating a professional, concise and reliable satisfaction questionnaire for each different patient in a more targeted manner, avoiding the formalism of traditional questionnaires and improving the effectiveness of the questionnaire. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a diagram of the satisfaction system architecture built on the Internet hospital platform in this embodiment;

[0039] Figure 2 This is a flow chart of the hospital satisfaction survey method of the present invention;

[0040] Figure 3 This is a flow chart of the satisfaction survey for the three stages of outpatient, hospitalization, and discharge in this embodiment. DETAILED DESCRIPTION

[0041] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0042] 1. Methods

[0043] 1. Patient basic information

[0044] Collect the basic medical information of patients and summarize the basic information feature set C = {c1, c2, ..., c N}, where cn (n∈{1,2,…N}) is the eigenvalue of the nth basic information, and N is the total number of basic patient information. The basic patient information is the data related to medical activities in the patient's personal information, such as height, weight, age, diagnosis results, and previous medical history.

[0045] Note: The characteristic values ​​of each piece of information can be converted from the patient's actual data according to pre-established coding rules. For example, the coding rule for gender characteristics can be set as follows: "Male" characteristic value is "1" and "Female" characteristic value is "2". If a piece of information is not collected, the corresponding characteristic value in the patient basic information feature set C can be assigned to zero.

[0046] For example, the patient basic information feature set includes: c1 age, c2 gender, c3 diagnosis result, c4 past medical history, c5 medical number, c6 current medical history, c7 duration between consultations, and c8 whether there is a history of complaints.

[0047] The corresponding basic information of a patient is shown in Table 1 below:

[0048] Table 1: Examples of patient basic information feature values

[0049]

[0050] Then the basic information feature set of the patient is {72, 2, H33.001, 0, 030309, 1, 65, 0}.

[0051] 2. Patient questionnaire information

[0052] By filling out the satisfaction questionnaire, the patient questionnaire information feature set A={a1,a2,…,a M}, where a m (m∈{1,2,…M}) is the eigenvalue of the mth questionnaire item, where M is the total number of patient questionnaires. The questionnaires contain patient satisfaction ratings of various hospital services, such as doctor service ratings and doctor skill ratings. If a particular item is not collected, the corresponding eigenvalue in the patient questionnaire feature set A can be assigned zero.

[0053] 3. Complaint Probability Prediction

[0054] Based on the patient basic information feature set C and the patient questionnaire information feature set A, the neural network model is used to predict the patient complaint probability p:

[0055] p=F P (C′,A′)

[0056] Where, F P(·) represents the complaint probability prediction function, and its specific calculation process is determined by the neural network model adopted; C′={c′1,c′2,…,c′ N} and A′={a′1,a′2,…,a′ M} are the normalized results of C and A, respectively. The normalized expressions of the two are as follows:

[0057]

[0058] Where n∈{1,2,…N}, m∈{1,2,…M}, c n_max and c n_min are the maximum and minimum values ​​of the basic information features of the nth patient, respectively. m_max and a m_min are the maximum and minimum values ​​of the information features of the mth patient questionnaire, ω m is the weighting factor (weight) of the characteristic value of the mth patient questionnaire information.

[0059] By calculating the patient complaint probability p, we can predict whether the patient has a tendency to complain.

[0060] 4. Complaint Item Prediction

[0061] Based on the predicted patient complaint probability p, the patient basic information feature set C, and the patient questionnaire information feature set Q, the neural network model is also used to predict the patient's possible complaint items:

[0062] R * =F R (p,C′,A′)

[0063] Where, Predict complaint feature sets for patients, Represents the predicted feature value of the xth complaint item, which is used to characterize whether this item is a complaint, for example When complaining about this, = means no complaint about this item, X is the total number of complaint items; F R (·) represents the complaint item prediction function, and its specific calculation process is determined by the neural network model used.

[0064] When the patient complaint probability p is greater than the threshold, the predicted potential complaint situation will be sent to the relevant functional departments for rectification and optimization.

[0065] 5. Questionnaire item optimization

[0066] For the same patient, based on previous complaints, the items of the next questionnaire will be optimized and adjusted to make the questionnaire items simpler and more targeted.

[0067] 1) Collect the actual complaint information of patients and summarize the actual complaint feature set R = {r1, r2, ..., r X}; Take the weighted sum of the predicted eigenvalue and the actual eigenvalue of each complaint item to obtain the patient's comprehensive complaint feature set R′={r′1,r′2,…,r′ X}:

[0068]

[0069] Where λ * and λ are the weights of the predicted eigenvalues ​​and actual eigenvalues ​​of the complaint items, respectively.

[0070] 2) Based on the patient comprehensive complaint feature set R′ and the patient complaint probability p, a neural network model is used to optimize the questionnaire items:

[0071] Q=F Q (p,R′)

[0072] Where, Q={q1,q2,…,q M} is the questionnaire item feature set, where q m is the characteristic value of the mth questionnaire item, which is used to indicate whether this questionnaire item is included in the questionnaire, for example, q m = 0, question m does not need to be investigated, q m =1 when the mth question needs to be investigated; Q (·) represents the questionnaire item optimization function, and its specific calculation process is determined by the neural network model used.

[0073] In summary, refer to Figure 2 As shown, when the patient first visits the hospital, an initial questionnaire Q1 is given, and the patient questionnaire information feature set A is collected. Combined with the patient basic information feature set C, the patient complaint probability p and possible complaint items R are predicted. * The initial questionnaire, Q1, can include all items or a subset of items after manual screening. Complaint items are then assigned to the appropriate department based on the probability of a complaint, providing early warning. Of course, if an actual complaint is received, it will also be assigned to the appropriate department. Finally, by combining predicted and actual complaints, Q2 is optimized for the patient's next visit.

[0074] In this example, the XGBoost (eXtreme Gradient Boosting) model is used to predict complaint probabilities, complaint items, and questionnaire item optimization. Of course, this neural network model requires a training dataset before use. During subsequent use, actual complaints can be used as feedback to continuously optimize the model.

[0075] 2. Example

[0076] 1. System

[0077] Reference Figure 1 As shown in the figure, the current Internet hospital platform includes hospital information system (HIS), appointment calling system, billing system, laboratory information system (LIS), examination information system, pharmacy information management system, admission preparation information system, electronic medical record system (EMR), and mobile nursing system; the above systems are interconnected through the hospital LAN (local area network) to obtain the required basic patient information.

[0078] Based on the above-mentioned Internet hospital platform, a satisfaction survey system, complaint warning module, complaint system, and questionnaire optimization module are added. Among them:

[0079] Satisfaction survey system, used to send satisfaction survey questionnaires to patients to collect corresponding patient questionnaire information;

[0080] The complaint warning module predicts the probability of complaints and complaint items based on the collected basic patient information and questionnaire information. When the complaint probability exceeds the threshold, a virtual complaint ticket is generated and pushed to the relevant functional departments for business supervision through the Internet hospital platform.

[0081] Complaint system, used to collect actual complaints from patients;

[0082] The questionnaire optimization module comprehensively predicts virtual complaints and actual complaints, optimizes the questionnaire items for each patient in the subsequent stage for the satisfaction survey system, and forms an adaptive questionnaire for each patient.

[0083] 2. Process

[0084] The “next visit” mentioned above is not limited to the same type of diagnosis and treatment. Figure 3 The following is an example of the three diagnosis and treatment links of outpatient, hospitalization and discharge:

[0085] Patient basic information includes: c1 age, c2 gender, c3 diagnosis results, c4 past medical history, c5 number, c6 current medical history, c7 interval between consultations, c8 complaint history, c9 history of diabetes, c 10 History of hypertension, 11 Whether surgery, c 12 Type of anesthesia, c 13 Diagnosis and treatment phase.

[0086] The questionnaire items include: q1 doctor service score, q2 doctor skill score, q3 nurse service score, q4 nurse skill score, q5 treatment process score, q6 registration service score, q7 examination service score, q8 inspection service score, q 10Service desk rating, q 11 Payment service rating, q 12 Admission process score, q 13 Food rating, q 14 Anesthesia service score, q 15 Inpatient examination service score, q 16 Inpatient laboratory service score, q 17 Discharge settlement service score, q 18 Hospitalization experience score, q 19 Disease guidance score, q 20 Demand feedback score, q 21 Integrity in medical practice score.

[0087] The complaint items include: r1 charging situation, r2 logistics support, r3 doctor service and technology, r4 nurse service and technology, r5 pharmacy service, r6 hospital management, r7 inspection service, and r8 service process.

[0088] First outpatient visit: The initial questionnaire Q1 is used to collect the patient's questionnaire information feature set A1, and combined with the patient's basic information C1 at this stage, the patient's complaint probability p1 and possible complaint items are predicted Achieve early warning.

[0089] Inpatient stage: Based on the complaints during the outpatient stage, the inpatient questionnaire Q2 is optimized to collect the patient questionnaire information feature set A2; then, combined with the patient's basic information C2 during the inpatient stage, the patient's complaint probability p2 and possible complaint items are predicted. Achieve early warning.

[0090] Discharge stage: Based on the complaints during the hospitalization stage, the discharge stage questionnaire Q3 is optimized to collect the patient questionnaire information feature set A3; then, combined with the basic information C3 of the patient during the discharge stage, the patient complaint probability p3 and possible complaint items are predicted. Achieve early warning.

[0091] Second outpatient visit: Based on the complaints during the previous discharge stage, the questionnaire Q4 is optimized and used for patient surveys during the next outpatient visit.

[0092] III. Devices, Storage Media, and Program Products

[0093] 1. Based on the same inventive concept as the above-mentioned hospital satisfaction survey method, the present application also provides an electronic device, which includes a processor and a memory, in which a computer-readable code is stored. When the computer-readable code is executed by the processor, the hospital satisfaction survey method of the present invention is implemented.

[0094] The memory includes a non-volatile storage medium and internal memory; the non-volatile storage medium can store an operating system and computer-readable code. The computer-readable code includes program instructions that, when executed, cause the processor to perform the hospital satisfaction survey method. The processor is used to provide computing and control capabilities, supporting the operation of the entire electronic device. The memory provides an environment for the execution of the computer-readable code in the non-volatile storage medium. When executed by the processor, the computer-readable code causes the processor to perform the hospital satisfaction survey method.

[0095] It should be understood that the processor may be a central processing unit, other general-purpose processors, digital signal processors, application-specific integrated circuits, field programmable gate arrays or other programmable logic devices, transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or any conventional processor.

[0096] 2. This application also provides a readable storage medium, which can be the internal storage unit of the electronic device described in the aforementioned embodiment, such as the hard disk or memory of the computer device. The readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart memory card, secure digital card, etc. equipped with the electronic device.

[0097] 3. The present application also provides a computer program product, comprising a computer program or instructions, which, when executed by a processor, implements the hospital satisfaction survey method of the present invention.

[0098] The present invention is not limited to the above-mentioned embodiments. Any obvious improvement, replacement or modification that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A hospital satisfaction survey method, characterized by: Collect basic patient information; Collecting questionnaire information from patients, which includes their evaluation of their satisfaction with the hospital's services; Based on the collected basic information of patients and the information from the patient questionnaire, a neural network model is used to predict the probability p of patient complaints and the items of complaints; Based on patient complaints, a neural network model was used to optimize the next questionnaire items: Q=F Q (p,R ′ ) Where, Q={q1,q2,…,q M } is the questionnaire item feature set, where q m is the characteristic value of the mth questionnaire item, which is used to indicate whether this questionnaire item is included in the questionnaire; F Q (·) represents the questionnaire item optimization function in the neural network model; R′={r′1,r′2,…,r′ X } is the comprehensive complaint feature set of patients that combines actual complaints and predicted complaints, where: Where λ * and λ are the weights of the predicted eigenvalue and actual eigenvalue of the complaint item, respectively. x and are the actual and predicted feature values ​​of the x-th complaint item respectively.

2. The hospital satisfaction survey method according to claim 1, characterized in that: Use neural network model to predict the probability of patient complaints: p=F P (C ′ ,A ′ ) Where, F P (·) represents the complaint probability prediction function in the neural network model; C′={c′1,c′2,…,c′ N } and A′={a′1,a′2,…,a′ M } are the standardized results of the patient basic information feature set C and the patient questionnaire information feature set A, respectively, where: Patient basic information feature set C = {c1, c2, ..., c N } is extracted from the patient's basic information, where c n is the eigenvalue of the nth basic information, n∈{1,2,…N}; Patient questionnaire information feature set A={a1,r2,…,a M } is extracted from the patient questionnaire information, where a m is the eigenvalue of the mth questionnaire information, m∈{1,2,…M}.

3. The hospital satisfaction survey method according to claim 2, characterized in that: Use neural network model to predict patient complaint items: R * =F R (p,C′,A′) Where, Predict complaint feature sets for patients, represents the predicted feature value of the x-th complaint item, which is used to characterize whether this item is a complaint, x∈{1,2,…X}; F R (·) represents the complaint item prediction function in the neural network model.

4. The hospital satisfaction survey method according to claim 2, characterized in that: The standardization method of the patient basic information feature set C and the patient questionnaire information feature set A is: Where n∈{1,2,…N}, m∈{1,2,…M}, c n_max and c n_min are the maximum and minimum values ​​of the basic information features of the nth patient, respectively. m_max and a m_min are the maximum and minimum values ​​of the information features of the mth patient questionnaire, ω m is the weighting factor of the characteristic value of the mth patient questionnaire information.

5. The hospital satisfaction survey method according to claim 1, characterized in that: The neural network model is an XGBoost model.

6. The hospital satisfaction survey method according to claim 1, characterized in that: Patient basic information includes: c1 age, c2 gender, c3 diagnosis results, c4 past medical history, c5 number, c6 current medical history, c7 interval between consultations, c8 complaint history, c9 history of diabetes, c 10 History of hypertension, 11 Whether surgery, c 12 Type of anesthesia, c 13 Diagnosis and treatment phase; The questionnaire items include: q1 doctor service score, q2 doctor skill score, q3 nurse service score, q4 nurse skill score, q5 treatment process score, q6 registration service score, q7 examination service score, q8 inspection service score, q 10 Service desk rating, q 11 Payment service rating, q 12 Admission process score, q 13 Food rating, q 14 Anesthesia service score, q 15 Inpatient examination service score, q 16 Inpatient laboratory service score, q 17 Discharge settlement service score, q 18 Hospitalization experience score, q 19 Disease guidance score, q 20 Demand feedback score, q 21 Integrity in medical practice rating; The complaint items include: r1 charging situation, r2 logistics support, r3 doctor service and technology, r4 nurse service and technology, r5 pharmacy service, r6 hospital management, r7 inspection service, and r8 service process.

7. A computer device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the hospital satisfaction survey method according to any one of claims 1 to 6 when executing the computer program.

8. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor is caused to execute the hospital satisfaction survey method according to any one of claims 1 to 6.

9. A computer program product, characterized in that: The method comprises a computer program, which, when executed by a processor, implements the hospital satisfaction survey method according to any one of claims 1 to 6.

Images