Adverse drug reaction case database system and method

Abstract

The invention relates to the technical field of databases, in particular to an adverse drug reaction case database system and method, and the system comprises a case recording module, a correlation analysis module, a sensitivity index extraction module, a drug screening and marking module and a case filing module. According to the method, the case data is classified in detail and subjected to in-depth correlation analysis, so that the connection between the medicine and the adverse reaction is clearly described, the accuracy of sensitivity analysis is enhanced, the rapid identification of the key risk reaction caused by the medicine is ensured, the risk management process is optimized, and the risk management efficiency is improved. The process of screening and marking problematic drugs enhances the response capability of the monitoring system, improves the management efficiency of medical institutions on drug safety, and increases the visualization of potential problems through systematic processing of adverse drug reaction data, so that medical providers can timely adjust therapeutic schedules and reduce the health risk of patients.

Description

Technical Field

[0001] The present invention relates to the field of database technology, and in particular to a drug adverse reaction case database system and method. Background Art

[0002] The field of database technology covers the storage, retrieval, management and optimization of information, enabling large amounts of data to be effectively organized, stored in computer systems, and managed through the use of database management systems (DBMS). Database technology can support various types of database models, such as relational, non-relational, and distributed databases. Key functions include adding, modifying, deleting, and querying data to ensure data consistency, integrity, and security. It also involves data mining, online analytical processing (OLAP), and transaction management. It is an integral part of modern IT structures and is widely used in many industries, including business, scientific research, and healthcare.

[0003] Among them, the adverse drug reaction case database system is an information system designed specifically for recording, storing and analyzing adverse reactions that occur in elderly patients after using drugs. Unlike ordinary adverse drug reaction databases, this system pays special attention to the special reactions and risks that may occur in the elderly during the use of drugs. Because the elderly often have multiple chronic diseases, their body functions deteriorate, their metabolic capacity weakens, and they often need to use multiple drugs for a long time (i.e., multiple medications), they are more likely to have adverse drug reactions or drug interactions.

[0004] Although existing database technologies have a wide range of applications in information storage and management, they often lack sufficient specialization and depth when dealing with medical-specific data such as adverse drug reactions. In particular, in terms of dynamic monitoring and real-time analysis of complex medical data, existing systems may fail to effectively identify and quickly respond to potential risks of drugs, resulting in an untimely response to drug safety issues. In addition, existing technologies generally lack the necessary mechanisms for automated extraction of sensitivity indicators and risk prediction, which hinders the discovery of key safety risks from large-scale data. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a drug adverse reaction case database system and method.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A drug adverse reaction case database system comprises:

[0007] The case record module collects adverse drug reaction cases, records drug types, user reactions and time of occurrence, organizes case data, marks the severity of adverse reactions, and generates adverse reaction case tables;

[0008] The correlation analysis module extracts the key content of the case according to the adverse reaction case table, statistically analyzes the relationship between the drug and the adverse reaction, determines the key correlation mode, ranks the number of cases, reaction types and their occurrence frequency of each drug according to the determination results, analyzes the main and secondary correlation paths, records the relationship between the drug and the adverse reaction, and generates a case correlation diagram;

[0009] The sensitivity index extraction module uses the case correlation diagram to determine the frequency and impact of adverse reactions, and simultaneously performs sensitivity analysis on the case data to identify the most sensitive reaction index, reveal key issues of drug safety, and obtain sensitivity analysis results;

[0010] The drug screening and marking module identifies and marks problematic drugs based on the sensitivity analysis results, conducts case review and data review on problematic drugs, determines the existing problems and potential impacts of problematic drugs, and constructs a list of problematic drugs;

[0011] The case archiving module marks the adverse reaction cases of problematic drugs according to the problematic drug list, adjusts the database search function, updates the accessibility of the data, and generates an optimized case database.

[0012] As a further solution of the present invention, the steps for obtaining the adverse reaction case table are:

[0013] Collect basic information of drug users, drug types and feedback data, screen case records describing adverse reactions, extract users' adverse reactions and time points of occurrence, and classify them by drug type to generate drug type adverse reaction matching results;

[0014] According to the adverse reaction matching results of the drug type, the severity classification standard of the adverse reaction is set, combined with the user feedback data, and the formula is used:

[0015] ;

[0016] Calculate the comprehensive score of adverse drug reactions , and graded according to the classification standards to obtain the adverse reaction severity grading results, among which, Represents the intensity of the adverse reaction description, Represents the time when the adverse reaction occurred. Represents the user's physiological characteristic value, represents the duration of the adverse reaction, Represents the total number of cases included in the statistics;

[0017] Combined with the adverse reaction severity grading results and the drug type adverse reaction matching results, the adverse reaction trends of different drug types are statistically analyzed according to time distribution, the adverse reaction categories are statistically analyzed, and an adverse reaction case table is integrated and output.

[0018] As a further solution of the present invention, the step of determining the key association pattern is:

[0019] According to the drug name, adverse reaction description, occurrence time and user characteristic data in the adverse reaction case table, the case content is parsed and the core information is extracted. Data mapping is performed according to the drug name and adverse reaction category, and the number of adverse reactions of each drug is counted to obtain preliminary statistical results of drugs and adverse reactions.

[0020] Based on the preliminary statistical results of the drugs and adverse reactions, the probability of adverse reactions corresponding to each drug was calculated using the formula:

[0021] ;

[0022] Calculating drugs Strength of association with adverse events , and obtain the statistical analysis results of drug-adverse reaction relationship, among which, Representative The severity of the adverse reaction in each case, represents the time span of the case, Represents the frequency of adverse reactions reported for a drug in different user groups, represents the number of cases that meet the screening criteria, Represents the number of adverse reaction categories involved in the drug;

[0023] According to the statistical analysis results of the drug-adverse reaction relationship, the common correlation characteristics of the different drugs in the cases were screened, the cross-interaction trends between the drugs were analyzed, and the main correlation patterns and secondary correlation patterns were extracted to obtain the key correlation pattern analysis results.

[0024] As a further solution of the present invention, the steps of obtaining the case correlation graph are:

[0025] According to the analysis results of the key association patterns, the adverse reaction types associated with each drug are extracted, and the drugs are sorted according to the incidence rate, total number of cases and distribution uniformity of adverse reactions to obtain the ranking results of the number of drug cases and reaction types;

[0026] Based on the ranking results of the number of drug cases and reaction types, the direct and indirect associations between drugs and adverse reactions are analyzed, the overlapping ratio of adverse reactions shared between drugs is determined, and the cross-influence between drugs is identified in combination with the time trend of adverse reactions, the transmission path of adverse reactions of different drugs is determined, and the results of the adverse reaction association path analysis between drugs are obtained;

[0027] According to the results of the adverse reaction association path analysis between the drugs, the direct and indirect association information between the drugs and the adverse reactions is recorded, the key adverse reaction nodes are determined, the adverse reaction propagation relationship is revealed according to the association patterns of the different drugs, and a case association diagram is generated.

[0028] As a further solution of the present invention, the steps for obtaining the sensitivity analysis results are:

[0029] Based on the case correlation diagram, the frequency of adverse reactions of each drug is calculated, and the changing trend of adverse reactions in different time periods is counted, and the statistical results of the frequency of adverse reactions are obtained in combination with the severity of the adverse reaction type;

[0030] Based on the statistical results of the adverse reaction frequency, the impact of the adverse reaction was analyzed using the formula:

[0031] ;

[0032] Calculate the impact value of each adverse reaction , and combined with the characteristics of different populations to analyze the differences in the impact of adverse reactions, and obtain the results of the adverse reaction impact assessment, among which, Represents the intensity score of adverse reactions. represents the duration of the adverse reaction, Represents the average severity of the same adverse reaction caused by different drugs, Represents the degree of variation of adverse reactions among different individuals;

[0033] Based on the adverse reaction impact assessment results, analyze the sensitivity of differential adverse reactions to drug safety, screen the most sensitive adverse reaction types, identify the reaction indicators that have the greatest impact on drug safety, determine the most critical adverse reaction issues, and obtain sensitivity analysis results.

[0034] As a further solution of the present invention, the steps for obtaining the problematic drug list are:

[0035] According to the sensitivity analysis results, drugs with highly sensitive adverse reactions are extracted through thresholds, and the case occurrence frequency of the drugs is counted, and the performance differences of the drugs in different user groups are analyzed to obtain the problem drug identification results;

[0036] Based on the problem drug identification results, review all case data involving the problem drugs, analyze the distribution of key parameters in the difference cases, and compare with the case data of similar drugs to evaluate the performance of the problem drugs in different usage environments, and obtain the problem drug case review and data review results;

[0037] Based on the case review and data review results of the problematic drugs, the potential impact of the problematic drugs was analyzed, the names, associated adverse reactions, key parameters and impact ranges of all problematic drugs were recorded, and they were classified and sorted according to the degree of difference in impact to construct a list of problematic drugs.

[0038] As a further solution of the present invention, the steps for obtaining the optimized case database are:

[0039] Based on the list of problematic drugs, classify them according to drug name, adverse reaction type and severity, mark cases with symptom description, medication time and dosage information, check data integrity, and obtain adverse reaction case marking results;

[0040] According to the adverse reaction case marking results, the existing database search rules are analyzed, the indexing method of the problematic drug cases is adjusted, matching is performed according to the drug name, adverse reaction type and severity, and the data access rights are adjusted to obtain the database search function optimization results;

[0041] Based on the optimization results of the database retrieval function, all marked adverse reaction cases are checked, the data storage structure is analyzed, the data is reorganized, the adverse reaction case retrieval and access of all problematic drugs are recorded, and an optimized case database is generated.

[0042] A drug adverse reaction case database method comprises the following steps:

[0043] S1: Collect basic information, drug types and feedback data of drug users, classify them according to drug types, set classification standards for the severity of adverse reactions, calculate the comprehensive score of adverse reactions of drugs, statistically analyze the adverse reaction trends of different drug types according to time distribution, and output adverse reaction case tables;

[0044] S2: parsing the case content according to the adverse reaction case table, performing data mapping according to drug name and adverse reaction category, calculating the association strength between drugs and adverse reactions, extracting the main association pattern and the secondary association pattern and sorting the drugs, determining the overlapping ratio of adverse reactions shared between drugs, determining the key adverse reaction nodes, and constructing a case association graph;

[0045] S3: Based on the case correlation diagram, the changing trend of adverse reactions in different time periods is counted, the impact value of each adverse reaction is calculated, the difference in the impact of adverse reactions is analyzed in combination with the characteristics of different populations, the sensitivity of the impact of different adverse reactions on drug safety is analyzed, the most critical adverse reaction problems are determined, and the sensitivity analysis results are obtained;

[0046] S4: Based on the sensitivity analysis results, extract drugs with high sensitivity to adverse reactions through thresholds, analyze the performance differences of drugs in different user populations, analyze the distribution of key parameters in different cases, evaluate the performance of problem drugs in different use environments, analyze the potential impact of problem drugs, and build a list of problem drugs;

[0047] S5: Based on the list of problematic drugs, classify them according to drug name, adverse reaction type and severity, mark the cases, adjust the indexing method of problematic drug cases, match them according to drug name, adverse reaction type and severity, analyze the data storage structure, reorganize the data, and generate an optimized case database.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are:

[0049] In the present invention, by exhaustively classifying case data and conducting in-depth correlation analysis, the connection between drugs and adverse reactions is clearly depicted, the accuracy of sensitivity analysis is enhanced, and the rapid identification of key risk reactions caused by drugs is ensured, thereby optimizing the risk management process. The process of screening and marking problem drugs enhances the responsiveness of the monitoring system and improves the management efficiency of drug safety by medical institutions. At the same time, through the systematic processing of drug adverse reaction data, the visualization of potential problems is increased, so that medical providers can adjust treatment plans in a timely manner and reduce patients' health risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a system flow chart of the present invention;

[0051] Figure 2 The following is a flowchart for obtaining the adverse reaction case table of the present invention;

[0052] Figure 3 It is a flow chart for determining the key association mode of the present invention;

[0053] Figure 4 A flowchart for obtaining a case correlation diagram of the present invention;

[0054] Figure 5 The flowchart for obtaining the sensitivity analysis results of the present invention is as follows;

[0055] Figure 6A flowchart for obtaining a list of problematic drugs of the present invention;

[0056] Figure 7 This is a flow chart for obtaining the optimized case database of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0059] See also Figure 1 , a drug adverse reaction case database system includes:

[0060] The case record module collects adverse drug reaction cases, records drug types, user reactions and time of occurrence, organizes case data, marks the severity of adverse reactions, and generates adverse reaction case tables;

[0061] The correlation analysis module extracts the key content of the case based on the adverse reaction case table, statistically analyzes the relationship between the drug and the adverse reaction, determines the key correlation mode, ranks the number of cases, reaction types and their frequency of occurrence for each drug based on the determined results, analyzes the main and secondary correlation paths, records the relationship between the drug and the adverse reaction, and generates a case correlation diagram;

[0062] The sensitivity index extraction module uses the case correlation diagram to determine the frequency and impact of adverse reactions, and simultaneously conducts sensitivity analysis on the case data to identify the most sensitive reaction indicators, reveal key issues in drug safety, and obtain sensitivity analysis results;

[0063] The drug screening and marking module identifies and marks problematic drugs based on the results of sensitivity analysis, conducts case reviews and data reviews on problematic drugs, determines the existing problems and potential impacts of problematic drugs, and builds a list of problematic drugs;

[0064] The case archiving module marks the adverse reaction cases of problematic drugs based on the list of problematic drugs, adjusts the database search function, updates the accessibility of the data, and generates an optimized case database.

[0065] The adverse reaction case table specifically includes drug identification information, reaction event description and event severity index; the case correlation diagram includes frequency ranking, correlation strength score and main correlation path; the sensitivity analysis results specifically refer to key sensitivity indicators, reaction impact assessment results and key safety issues; the problem drug list includes problem drug identification records, potential impact analysis results and case review details; the optimized case database includes retrieval function optimization results and data accessibility update records.

[0066] See also Figure 2 , the steps to obtain the adverse reaction case table are:

[0067] Collect basic information of drug users, drug types and feedback data, screen case records describing adverse reactions, extract users' adverse reactions and time points of occurrence, and classify them by drug type to generate drug type adverse reaction matching results;

[0068] The basic information of drug users, drug types and feedback data are collected. First, the data is screened and standardized to ensure the integrity and availability of the records. For the information of drug users, including age, gender, weight, medical history, allergy history, etc., the field matching method is used for classification and storage, and different categories of data are indexed for subsequent query. For drug types, they are classified and sorted according to drug names, active ingredients, manufacturers, batches, etc., and the same type of drugs are merged to avoid repeated calculations caused by different naming methods. For the data fed back by users, keyword matching is used to extract adverse reaction descriptions, and combined with common adverse reaction symptoms in medical literature, a A keyword thesaurus is built to accurately identify the symptom descriptions reported by users, and feedback records that cannot be classified or are vaguely described are filtered out. After cleaning the data, the frequency of adverse reactions of each drug in different populations is calculated. Based on the physiological characteristics of the users, the users are divided into different groups, such as children, adults, and the elderly, and the probability of adverse reactions in each group is calculated. Conditional screening is used to find the types of drugs with higher frequencies of adverse reactions in specific groups, and preliminary classification is performed according to the severity of the adverse reactions. On this basis, the distribution of the time of occurrence of adverse reactions is further analyzed, the incidence of adverse reactions at different time points is calculated, and the adverse reaction matching results of drug types are obtained.

[0069] According to the adverse reaction matching results of drug types, the severity classification standard of adverse reactions is set, combined with user feedback data, and the formula is used:

[0070] ;

[0071] Calculate the comprehensive score of adverse drug reactions , and graded according to the classification standards to obtain the adverse reaction severity grading results, among which, Represents the intensity of the adverse reaction description, Represents the time when the adverse reaction occurred. Represents the user's physiological characteristic value, represents the duration of the adverse reaction, Represents the total number of cases included in the statistics;

[0072] (Intensity of adverse reaction description): The user feedback text is matched with keywords and the description is quantitatively scored based on medical standards. The score range is set from 1 to 5, where 1 represents a mild reaction and 5 represents a severe reaction. If the user feedback description shows "severe headache, nausea, and vomiting", it is matched to the high severity level and set .

[0073] (Time of adverse reaction): The time interval is calculated by comparing the time of taking the medicine with the time of adverse reaction, in hours. If the user has adverse reaction 8 hours after taking the medicine, set .

[0074] (User's physiological characteristics): Based on age, weight, BMI, underlying diseases and other characteristics, the weighted average method is used to calculate the comprehensive score, with a score range of 0-5. The higher the value, the more susceptible the user is to the effects of the drug. The user is 65 years old, has a high BMI, and suffers from hypertension. .

[0075] (Duration of adverse reactions): Extract the duration of adverse reactions from user feedback data in hours. If the user's symptoms last for 20 hours, set .

[0076] (Total number of cases included in the statistics): Count the number of adverse reaction cases of this drug type. There are 75 adverse reaction cases of this drug in the database. Set .

[0077] Calculate the weighted intensity term:

[0078] ;

[0079] Calculate the physiological characteristics influence:

[0080] ;

[0081] Request the final comprehensive score:

[0082] ;

[0083] The calculation results It indicates the comprehensive score of the adverse reactions of the drug, which can be used to evaluate the severity of the adverse reactions of the drug in different users. The higher the score, the more obvious the adverse reaction effect of the drug in the statistical case. 1.2747 indicates that the adverse reactions of the drug are mild overall, but combined with the scores of other drugs, further analysis is still needed. This score can be used for subsequent classification of the severity of adverse reactions. At the same time, combined with time distribution data, the risk level of the drug in different populations can be analyzed, and high-risk drug types can be further screened.

[0084] Combine the adverse reaction severity classification results and drug type adverse reaction matching results, statistically analyze the adverse reaction trends of different drug types according to time distribution, statistically analyze the adverse reaction categories, and integrate and output the adverse reaction case table;

[0085] According to the adverse reaction severity classification results, combined with the drug type adverse reaction matching results, the time distribution of adverse reactions was statistically analyzed. First, the time information of user feedback was extracted from the data records, including the time of taking the medicine, the time of occurrence of the adverse reaction, the duration of the adverse reaction, etc., and the delay time of the adverse reaction was calculated, that is, the time interval from taking the medicine to the occurrence of the adverse reaction. For cases with incomplete records, the missing time information was filled by interpolation method to ensure the integrity of the time distribution. All adverse reaction cases were arranged in chronological order, the frequency of occurrence of adverse reactions in different time periods was calculated, and a time curve was drawn to identify the time intervals of adverse reactions with high frequency, such as Some drugs may cause adverse reactions within 1 hour after taking them, while the adverse reactions of some drugs may not appear until 24 hours after taking them. For the occurrence of adverse reactions in different time periods, we further analyze the corresponding drug types and user characteristics, and screen out drug categories with higher incidences of adverse reactions. We calculate the proportion of each type of adverse reaction in the overall cases, find out the main types of adverse reactions, such as nausea, headache, rash, etc., and analyze their occurrence trends. For drugs with high frequency of adverse reactions, we further combine them with medical database comparison to verify whether they are consistent with known medical research, screen high-risk drug types, and finally integrate all data to generate an adverse reaction case table.

[0086] See also Figure 3 , the steps to determine the key association pattern are:

[0087] According to the drug name, adverse reaction description, occurrence time and user characteristic data in the adverse reaction case table, the case content is parsed and the core information is extracted. Data mapping is performed according to the drug name and adverse reaction category, and the number of adverse reactions of each drug is counted to obtain preliminary statistical results of drugs and adverse reactions.

[0088] According to the drug name, adverse reaction description, occurrence time and user characteristics data in the adverse reaction case table, the case content is parsed. First, the data fields are formatted, the field structure in different case sources is unified, and the core fields are extracted, including drug name, symptom description, reaction occurrence time, user characteristics, etc. For incomplete case data, a filling or elimination strategy is used to ensure data quality. Cluster analysis is performed on the adverse reactions of different drugs, and data mapping is established according to drug name and adverse reaction category. The number of adverse reactions of each drug is counted, and the distribution characteristics of adverse reactions are calculated based on different cases of the same drug. For the occurrence time field, the time interval between adverse reactions is calculated, and it is analyzed whether there is a centralized trend in a specific time period. Finally, the preliminary statistical results of drugs and adverse reactions are obtained.

[0089] Based on the preliminary statistical results of drugs and adverse reactions, the probability of adverse reactions corresponding to each drug is calculated using the formula:

[0090] ;

[0091] Calculating drugs Strength of association with adverse events , and obtain the statistical analysis results of drug-adverse reaction relationship, among which, Representative The severity of the adverse reaction in each case, represents the time span of the case, Represents the frequency of adverse reactions reported for a drug in different user groups, represents the number of cases that meet the screening criteria, Represents the number of adverse reaction categories involved in the drug;

[0092] (Adverse reaction severity): Based on the adverse reaction severity classification in medical literature, the adverse reaction type of each case is classified and scored, with a score range of 1-5. The larger the value, the more severe the adverse reaction. According to user feedback on "severe allergic reaction", the .

[0093] (Case time span): Calculate the interval between the time when the adverse reaction occurred and the time when the medication was taken in this case, in hours. In this case, the adverse reaction occurred 6 hours after taking the medication, so set .

[0094] (Frequency of adverse reaction reports): Count the frequency of adverse reaction reports of the drug in all cases and calculate the corresponding proportion. In this example, the drug has 300 reports with 90 adverse reactions. .

[0095] (Number of cases that meet the screening criteria): Count the number of valid cases for the drug. In this example, .

[0096] (Number of adverse reactions to the drug): Count the different types of adverse reactions to the drug. In this example, .

[0097] Calculate the weighted severity term:

[0098] ;

[0099] Calculate the adverse reaction frequency term:

[0100] ;

[0101] Take the square root:

[0102] ;

[0103] Find the final correlation strength:

[0104] ;

[0105] The calculation results Represents the comprehensive strength of association between the drug and the adverse reaction. The larger the value, the closer the relationship between the drug and the adverse reaction. In actual analysis, this value can be used to screen high-risk drugs and conduct comparative analysis in combination with other drugs. The results show that the adverse reaction association strength of this drug is low and it is not a major adverse reaction-associated drug, but its potential impact still needs to be further evaluated in combination with other drug comparative data.

[0106] According to the statistical analysis results of drug-adverse reaction relationships, the common correlation characteristics of different drugs in the cases were screened, the cross-effect trends between drugs were analyzed, and the main correlation patterns and secondary correlation patterns were extracted to obtain the key correlation pattern analysis results;

[0107] Based on the statistical analysis results of the drug-adverse reaction relationship, the adverse reaction patterns of highly correlated drugs are calculated. First, the adverse reaction types of each drug are sorted, and the high-frequency adverse reaction categories are screened out. Trend analysis is performed based on the time of case occurrence. The incidence of different drugs in each time period is statistically analyzed, and their distribution characteristics in different populations are calculated. For common adverse reactions between multiple drugs, their cross-action trends are extracted, and their mechanisms of action are analyzed. Combined with historical data, drug adverse reaction combinations with significant patterns are found. Finally, the main and secondary correlation patterns are screened, and their characteristics are recorded to finally determine the key correlation patterns.

[0108] See also Figure 4 , the steps to obtain the case correlation graph are:

[0109] According to the results of key association pattern analysis, the adverse reaction types associated with each drug were extracted, and the drugs were sorted according to the incidence rate, total number of cases and distribution uniformity of adverse reactions to obtain the ranking results of drug case quantity and reaction type;

[0110] According to the results of key association pattern analysis, the number of adverse reaction cases involved in each drug was counted. First, the core fields such as drug name, adverse reaction description, and case occurrence time in the adverse reaction case table were extracted and classified according to the drug name to ensure that the cases of the same drug were attributable to the same. For the adverse reaction description content, the natural language description of user feedback was converted into standardized adverse reaction categories by keyword extraction, and a unified category coding table was established to ensure that data from different sources were consistent during analysis. For adverse reaction descriptions that could not be directly classified, the symptom similarity comparison method was used to match them to the closest standardized category. For the case occurrence time, the interval between the adverse reaction occurrence time and the medication time was extracted and recorded in hours. The time information was converted into numerical form to facilitate the subsequent calculation of the frequency of adverse reactions. For each adverse reaction category of each drug, its number of occurrences in all cases was calculated, and the proportion of each adverse reaction type in the drug case was counted. Combined with the total number of cases, the overall adverse reaction incidence of the drug was calculated. All drugs were sorted, and drugs with a high number of adverse reaction cases were screened. The adverse reaction types involved and their occurrence frequencies were recorded, and finally the ranking results of the number of drug cases and reaction types were obtained.

[0111] Based on the ranking results of drug case quantity and reaction type, the direct and indirect associations between drugs and adverse reactions are analyzed, the overlapping proportion of adverse reactions shared between drugs is determined, and the cross-influence between drugs is identified in combination with the time trend of adverse reactions, the transmission path of adverse reactions of different drugs is determined, and the results of the adverse reaction association path analysis between drugs are obtained;

[0112] The ranking results of the number of drug cases and reaction types were obtained, and the direct and indirect associations between drugs and adverse reactions were analyzed. First, the drug-adverse reaction combinations with high frequency associations were screened. For the top-ranked drugs, their main adverse reaction types were counted, and the overlapping proportion of adverse reactions shared by the drug and other drugs was calculated. For adverse reactions co-occurring with different drugs, the number of cases was recorded, their proportion in all cases was calculated, and it was determined whether the adverse reaction occurred frequently in the cases of multiple drugs. For drugs with shared adverse reactions, their active ingredients, metabolic mechanisms and indications were analyzed, and drugs with similar structures or similar pathways of action were screened. For adverse reactions of the same category, their occurrence time in different drugs was calculated, the distribution interval of their occurrence time was extracted, and the mean time interval of the adverse reaction in different drugs was calculated. It was analyzed whether there was a time concentration effect on adverse reactions. For drugs with short time intervals and high co-occurrence frequency, it was determined whether there was an interaction between drugs, and the potential causes of adverse reactions were analyzed in combination with historical medical data. The transmission paths of adverse reactions of different drugs were screened, and the main and secondary association paths between drugs were calculated according to the co-occurrence frequency of adverse reactions, the total number of cases and time synchronization, and finally the analysis results of the adverse reaction association paths between drugs were obtained.

[0113] According to the results of the adverse reaction association path analysis between drugs, the direct and indirect association information between drugs and adverse reactions is recorded, the key adverse reaction nodes are determined, the adverse reaction propagation relationship is revealed according to the association pattern of different drugs, and a case association diagram is generated;

[0114] According to the results of the adverse reaction association path analysis between drugs, the direct and indirect association information between drugs and adverse reactions was recorded. First, the association information of drug-adverse reaction combinations was sorted out, and drugs were classified according to the type of adverse reaction. For adverse reactions co-occurring with multiple drugs, the transmission intensity of the adverse reaction between different drugs was calculated. For high-frequency adverse reactions, the main associated drugs were screened, and their impact range in different populations was calculated. The differences in adverse reactions of drugs under different ages, genders and health conditions were analyzed. For adverse reactions co-occurring with different drugs, their cross-action patterns were extracted, and the adverse reaction transmission links of drugs were recorded. For drugs that may have indirect associations, their action pathways were analyzed, and it was determined whether there was a cross-influence on the metabolic or excretion pathways. The adverse reaction transmission patterns between different drug types were extracted, the degree of association between different drug categories was calculated, the most representative drug-adverse reaction transmission relationship was screened, and a drug-adverse reaction association network was constructed. According to the drug's mechanism of action, metabolic pathway and adverse reaction type, the impact path of drugs and adverse reactions was drawn, and finally a case association diagram was generated.

[0115] See also Figure 5 , the steps to obtain the sensitivity analysis results are:

[0116] Based on the case correlation diagram, the frequency of adverse reactions of each drug is calculated, and the changing trend of adverse reactions in different time periods is statistically analyzed. Combined with the severity of the adverse reaction type, the statistical results of the adverse reaction frequency are obtained;

[0117] Based on the drug name, adverse reaction type, occurrence time and association path information in the case association graph, we first screened out case data with clear records between drugs and adverse reactions, extracted the occurrence time of all adverse reactions, and calculated the frequency of adverse reactions of the corresponding drugs. For each type of adverse reaction of each drug, we classified them according to the symptom category to ensure that the same type of adverse reactions can be classified into the same category. The frequency of occurrence of each adverse reaction in different drugs was counted, and the trend changes of various types of adverse reactions in different time periods were calculated. The time of adverse reaction occurrence was divided into cycles, the number of cases in each time period was counted, and the proportion of the total number of cases was calculated. The severity of the adverse reaction type was scored, and the adverse reaction propagation relationship of the drug was analyzed through the association path information in the association graph. Combined with the frequency of adverse reactions in different time periods, the changing trend of high-frequency adverse reactions was calculated, and finally the statistical results of the frequency of adverse reactions were obtained.

[0118] Based on the statistical results of the frequency of adverse reactions, the impact of adverse reactions was analyzed using the formula:

[0119] ;

[0120] Calculate the impact value of each adverse reaction , and combined with the characteristics of different populations to analyze the differences in the impact of adverse reactions, and obtain the results of the adverse reaction impact assessment, among which, Represents the intensity score of adverse reactions. represents the duration of the adverse reaction, Represents the average severity of the same adverse reaction caused by different drugs, Represents the degree of variation of adverse reactions among different individuals;

[0121] (Intensity score of adverse reactions): Based on the symptom description reported by the user, the severity of the adverse reaction is scored in combination with medical standards. The score range is 1-5. The higher the value, the more severe the symptom. In this example, the user reported "severe headache, nausea, and vomiting", which is matched to a higher severity level. .

[0122] (Duration of adverse reaction): Calculate the time from the onset to the relief of the adverse reaction in this case, in hours. In this case, the adverse reaction lasts for 18 hours. Set .

[0123] (Average severity of the same adverse reaction caused by different drugs): Calculate the average severity score of the adverse reaction in different drug use cases (the average value of the same adverse reaction caused by different drugs), and get .

[0124] (The degree of variation of the adverse reaction in different individuals): The standard deviation of the severity of the adverse reaction in different users is used to measure the differences between different individuals. The standard deviation is calculated .

[0125] Calculate the sum of impact factors:

[0126] ;

[0127] Calculate individual differences in adverse reactions:

[0128] ;

[0129] Take the square root:

[0130] ;

[0131] Find the final impact value:

[0132] ;

[0133] Calculation results Represents the comprehensive impact value of the adverse reaction. The higher the value, the greater the impact of the adverse reaction on multiple factors, and the more attention should be paid to it in drug safety assessment. The results show that the adverse reaction has a greater impact and will be used to identify highly sensitive reaction indicators in subsequent sensitivity analysis to further reveal key issues in drug safety.

[0134] Based on the adverse reaction impact assessment results, analyze the sensitivity of differential adverse reactions to drug safety, screen the most sensitive adverse reaction types, identify the reaction indicators that have the greatest impact on drug safety, determine the most critical adverse reaction issues, and obtain sensitivity analysis results;

[0135] Based on the results of the adverse reaction impact assessment, the sensitivity of different adverse reactions to drug safety was analyzed. First, the impact values ​​of all adverse reactions were counted and classified according to the adverse reaction types. Adverse reactions with higher impact values ​​were screened, and their distribution characteristics in the population were analyzed. The differences in adverse reactions in different ages, genders and health status were calculated. For adverse reactions with higher impact values, their sensitivity index in the population was calculated, the fluctuations of the adverse reactions among different individuals were analyzed, and it was determined whether they were affected by specific physiological characteristics. For adverse reactions with higher sensitivity indexes, the key issues of drug safety were screened in combination with their severity, duration and scope of impact, and the sensitivity analysis results were obtained.

[0136] See also Figure 6 , the steps to obtain the list of problematic drugs are:

[0137] According to the sensitivity analysis results, drugs with highly sensitive adverse reactions are extracted through thresholds, and the case frequency of drugs is counted, and the performance differences of drugs in different user groups are analyzed to obtain the problem drug identification results;

[0138] According to the drug name, adverse reaction type and impact index in the sensitivity analysis results, all drugs involving highly sensitive adverse reactions are extracted, and the drugs are sorted according to the adverse reaction impact index. Drugs with a higher impact index (compared with the preset threshold) are screened, and the case frequency of drugs is counted. All case records are obtained, and the drugs used, the time of adverse reaction occurrence and the symptom manifestation of each case are analyzed one by one. The case data are structured so that they can be classified according to the type of adverse reaction, basic patient information and medication status. According to the population information involved in the case data, different groups are divided according to age, gender and basic health status, and the incidence of adverse reactions to the same drug in various populations is calculated. The sensitivity changes under different ages, genders and health conditions are compared, and the distribution of adverse reactions in the population is counted. It is analyzed whether there are certain specific populations with a higher incidence of adverse reactions to specific drugs. Potential high-risk groups are evaluated, and high-risk drugs are further screened in combination with the time of adverse reaction occurrence and the severity of symptoms. By analyzing the case data of high-risk groups, the historical adverse reaction frequency of the corresponding drugs is calculated, and finally the high-risk problematic drugs are determined to obtain the problematic drug identification results.

[0139] Based on the problem drug identification results, review all case data involving the problem drugs, analyze the distribution of key parameters in the difference cases, and compare with the case data of similar drugs to evaluate the performance of the problem drugs in different usage environments, and obtain the problem drug case review and data review results;

[0140] Based on the results of problem drug identification, all historical case data involving problem drugs were reviewed, and the time of occurrence, dosage, symptom severity and duration of adverse reactions in each case were extracted. The case data were segmented according to the dosage, duration of use and individual characteristics of the drug, and the differences in adverse reactions of the same drug at different dosages and durations of use were compared. All cases involving problem drugs were classified according to factors such as symptom manifestations, patient background, and drug usage. Cases with abnormal symptom manifestations or inconsistent with the usual adverse reaction pattern were screened out, and the proportion of abnormal cases in each problem drug was counted. In combination with the medication background in each case, it was analyzed whether the adverse reactions were affected by specific usage environments or combined medication factors. All abnormal case data were extracted and compared according to the similarity of adverse reaction symptoms. Cases with obvious deviations from other cases were screened out. In combination with long-term drug monitoring records, the adverse reaction occurrence pattern of the drug was analyzed to determine whether there was a long-term change in adverse reaction trends. Finally, the problem drug case review and data review results were obtained.

[0141] Based on the review of problematic drug cases and data review results, analyze the potential impact of problematic drugs, record the names, associated adverse reactions, key parameters and impact ranges of all problematic drugs, classify and sort them according to the degree of difference in impact, and build a list of problematic drugs;

[0142] Based on the review of problem drug cases and data review results, the potential impact of each problem drug was analyzed. First, the relevant parameters of all problem drugs were extracted, including name, adverse reaction type, scope of impact, patient background, dosage and long-term trend. The problem drugs were classified according to the adverse reaction pattern, and the adverse reaction categories with a wide impact range or severe symptoms were screened out. The adverse reaction manifestations of each drug in this category were analyzed, and the differences in the impact of different drugs in the same population were statistically analyzed. The drugs with abnormal performance were extracted, and their historical adverse reaction trends were further reviewed. The changes in their adverse reaction incidence rates in different time periods were calculated to determine whether there were long-term cumulative effects or potential drug metabolism problems. Combined with the existing adverse reaction data, the historical use of the drug and the corresponding adverse reaction records were analyzed, the symptom duration and recovery of different patients were compared, and cases with higher symptom severity and slower recovery were extracted to determine possible drug safety risks. Finally, the names, associated adverse reactions, key parameters and scope of impact of all problem drugs were sorted out, and they were classified and sorted according to different degrees of impact to finally construct a list of problem drugs.

[0143] See also Figure 7 , the steps to obtain the optimized case database are:

[0144] Based on the list of problematic drugs, classify them according to drug name, adverse reaction type and severity, mark cases with symptom description, medication time and dosage information, check data integrity, and obtain adverse reaction case marking results;

[0145] Call up the list of problematic drugs, obtain all drug cases related to adverse reactions, extract key information from the cases, including drug name, patient medication status, adverse reaction time, adverse reaction type, severity, patient age, gender, past medical history and medication dosage, screen the data, exclude invalid or duplicate cases, ensure that the selected cases have complete key information, analyze the common characteristics between cases, extract high-frequency adverse reaction types, identify the main adverse reaction symptoms involved in different categories of drugs, cross-check these marked information with the original case data, confirm the accuracy of the marking, if there are marking errors, re-screen and correct the relevant cases to ensure the authenticity and reliability of the marking results, after completing the marking of all cases, classify and organize the data, sort by drug category, adverse reaction type and frequency of occurrence, and classify all marked cases into corresponding categories so that subsequent data queries can quickly locate related cases and finally obtain the adverse reaction case marking results.

[0146] According to the adverse reaction case marking results, the existing database search rules are analyzed, the indexing method of problematic drug cases is adjusted, matching is performed according to drug name, adverse reaction type and severity, and data access permissions are adjusted to obtain the optimization results of the database search function;

[0147] Based on the adverse reaction case marking results, the case data structure in the database is reviewed to identify key factors affecting data retrieval, including case storage format, field naming method, index structure and query logic. Check whether the data index can quickly match the drug name, adverse reaction type and severity. If the index field is redundant or missing, adjust the index method to ensure that all cases can be accurately retrieved according to relevant conditions. Compare the query logic of the current database, analyze the query execution efficiency, determine whether there are problems with duplicate data, missing data or query delays, verify the data query results item by item, check whether the query results under different screening conditions are consistent, adjust data access rights, ensure that different user categories can access case data within a specific range according to their permissions, and prevent unauthorized users from obtaining sensitive information. While optimizing the index storage method, update the database query rules to make the retrieval process more efficient. At the same time, track the data access log to ensure the rationality of data calls, and finally obtain the optimization results of the database retrieval function.

[0148] Based on the results of database search function optimization, all marked adverse reaction cases are checked, the data storage structure is analyzed, the data is reorganized, and the adverse reaction case retrieval and access of all problematic drugs are recorded to generate an optimized case database;

[0149] According to the optimization results of the database retrieval function, check all marked adverse reaction cases, analyze the storage format in the database, ensure that the data classification method conforms to the optimized retrieval rules, organize and adjust the data index, check whether each case can be quickly queried according to the drug name, adverse reaction type and severity, ensure that the database storage structure can efficiently support query requests, optimize the classification method of case data, adjust the data storage level to reduce redundant data and improve data access efficiency, re-execute data integrity check for each optimized case data, confirm that all cases contain complete field information, and check the logical relationship between data to ensure that adverse reaction cases and drug information can accurately correspond. After completing the data organization, review the overall storage structure of the database to confirm that all optimization content has been correctly applied, and perform a final data consistency check to exclude possible duplicate data or misclassified cases to ensure the accuracy and completeness of database data, and finally generate an optimized case database.

[0150] A drug adverse reaction case database method comprises the following steps:

[0151] S1: Collect basic information, drug types and feedback data of drug users, classify them according to drug types, set classification standards for the severity of adverse reactions, calculate the comprehensive score of adverse reactions of drugs, statistically analyze the adverse reaction trends of different drug types according to time distribution, and output adverse reaction case tables;

[0152] S2: Analyze the case content according to the adverse reaction case table, map the data according to the drug name and adverse reaction category, calculate the association strength between the drug and the adverse reaction, extract the main and secondary association patterns and sort the drugs, determine the overlap ratio of shared adverse reactions between drugs, determine the key adverse reaction nodes, and construct a case association diagram;

[0153] S3: Based on the case correlation diagram, the changing trend of adverse reactions in different time periods is counted, the impact value of each adverse reaction is calculated, the difference in the impact of adverse reactions is analyzed in combination with the characteristics of different populations, the sensitivity of the impact of different adverse reactions on drug safety is analyzed, the most critical adverse reaction problems are determined, and the sensitivity analysis results are obtained;

[0154] S4: Based on the results of sensitivity analysis, extract drugs with high sensitivity to adverse reactions through thresholds, analyze the performance differences of drugs in different use populations, analyze the distribution of key parameters in different cases, evaluate the performance of problem drugs in different use environments, analyze the potential impact of problem drugs, and build a list of problem drugs;

[0155] S5: Based on the list of problematic drugs, classify them according to drug name, adverse reaction type and severity, mark the cases, adjust the indexing method of problematic drug cases, match them according to drug name, adverse reaction type and severity, analyze the data storage structure, reorganize the data, and generate an optimized case database.

[0156] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A drug adverse reaction case database system, characterized by: The system comprises: The case record module collects adverse drug reaction cases, records drug types, user reactions and time of occurrence, organizes case data, marks the severity of adverse reactions, and generates adverse reaction case tables; The correlation analysis module extracts the key content of the case according to the adverse reaction case table, statistically analyzes the relationship between the drug and the adverse reaction, determines the key correlation mode, ranks the number of cases, reaction types and their occurrence frequency of each drug according to the determination results, analyzes the main and secondary correlation paths, records the relationship between the drug and the adverse reaction, and generates a case correlation diagram; The sensitivity index extraction module uses the case correlation diagram to determine the frequency and impact of adverse reactions, and simultaneously performs sensitivity analysis on the case data to identify the most sensitive reaction index and obtain sensitivity analysis results; The drug screening and marking module identifies and marks problematic drugs based on the sensitivity analysis results, conducts case review and data review on problematic drugs, determines the existing problems and potential impacts of problematic drugs, and constructs a list of problematic drugs; The case archiving module marks the adverse reaction cases of problematic drugs according to the problematic drug list, adjusts the database search function, updates the accessibility of the data, and generates an optimized case database.

2. The adverse drug reaction case database system according to claim 1, characterized in that: The steps for obtaining the adverse reaction case table are as follows: Collect basic information of drug users, drug types and feedback data, screen case records describing adverse reactions, extract users' adverse reactions and time points of occurrence, and classify them by drug type to generate drug type adverse reaction matching results; According to the adverse reaction matching results of the drug type, the severity classification standard of the adverse reaction is set, combined with the user feedback data, and the formula is used: ； Calculate the comprehensive score of adverse drug reactions , and graded according to the classification standards to obtain the adverse reaction severity grading results, among which, Represents the intensity of the adverse reaction description, Represents the time when the adverse reaction occurred. Represents the user's physiological characteristic value, represents the duration of the adverse reaction, Represents the total number of cases included in the statistics; Combined with the adverse reaction severity grading results and the drug type adverse reaction matching results, the adverse reaction trends of different drug types are statistically analyzed according to time distribution, the adverse reaction categories are statistically analyzed, and an adverse reaction case table is integrated and output.

3. The adverse drug reaction case database system according to claim 2, characterized in that: The steps for determining the key association mode are: According to the drug name, adverse reaction description, occurrence time and user characteristic data in the adverse reaction case table, the case content is parsed and the core information is extracted. Data mapping is performed according to the drug name and adverse reaction category, and the number of adverse reactions of each drug is counted to obtain preliminary statistical results of drugs and adverse reactions. Based on the preliminary statistical results of the drugs and adverse reactions, the probability of adverse reactions corresponding to each drug was calculated using the formula: ； Calculating drugs Strength of association with adverse events , and obtain the statistical analysis results of drug-adverse reaction relationship, among which, Representative The severity of the adverse reaction in each case, represents the time span of the case, Represents the frequency of adverse reactions reported for a drug in different user groups, represents the number of cases that meet the screening criteria, Represents the number of adverse reaction categories involved in the drug; According to the statistical analysis results of the drug-adverse reaction relationship, the common correlation characteristics of the different drugs in the cases were screened, the cross-interaction trends between the drugs were analyzed, and the main correlation patterns and secondary correlation patterns were extracted to obtain the key correlation pattern analysis results.

4. The adverse drug reaction case database system according to claim 3, characterized in that: The steps for obtaining the case correlation graph are: According to the analysis results of the key association patterns, the adverse reaction types associated with each drug are extracted, and the drugs are sorted according to the incidence rate, total number of cases and distribution uniformity of adverse reactions to obtain the ranking results of the number of drug cases and reaction types; Based on the ranking results of the number of drug cases and reaction types, the direct and indirect associations between drugs and adverse reactions are analyzed, the overlapping ratio of adverse reactions shared between drugs is determined, and the cross-influence between drugs is identified in combination with the time trend of adverse reactions, the transmission path of adverse reactions of different drugs is determined, and the results of the adverse reaction association path analysis between drugs are obtained; According to the results of the adverse reaction association path analysis between the drugs, the direct and indirect association information between the drugs and the adverse reactions is recorded, the key adverse reaction nodes are determined, the adverse reaction propagation relationship is revealed according to the association patterns of the different drugs, and a case association diagram is generated.

5. The adverse drug reaction case database system according to claim 4, characterized in that: The steps for obtaining the sensitivity analysis results are as follows: Based on the case correlation diagram, the frequency of adverse reactions of each drug is calculated, and the changing trend of adverse reactions in different time periods is counted, and the statistical results of the frequency of adverse reactions are obtained in combination with the severity of the adverse reaction type; Based on the statistical results of the adverse reaction frequency, the impact of the adverse reaction was analyzed using the formula: ； Calculate the impact value of each adverse reaction , and combined with the characteristics of different populations to analyze the differences in the impact of adverse reactions, and obtain the results of the adverse reaction impact assessment, among which, Represents the intensity score of adverse reactions. represents the duration of the adverse reaction, Represents the average severity of the same adverse reaction caused by different drugs, Represents the degree of variation of adverse reactions among different individuals; Based on the adverse reaction impact assessment results, analyze the sensitivity of differential adverse reactions to drug safety, screen the most sensitive adverse reaction types, identify the reaction indicators that have the greatest impact on drug safety, determine the most critical adverse reaction issues, and obtain sensitivity analysis results.

6. The adverse drug reaction case database system according to claim 5, characterized in that: The steps for obtaining the problematic drug list are as follows: According to the sensitivity analysis results, drugs with highly sensitive adverse reactions are extracted through thresholds, and the case occurrence frequency of the drugs is counted, and the performance differences of the drugs in different user groups are analyzed to obtain the problem drug identification results; Based on the problem drug identification results, review all case data involving the problem drugs, analyze the distribution of key parameters in the difference cases, and compare with the case data of similar drugs to evaluate the performance of the problem drugs in different usage environments, and obtain the problem drug case review and data review results; Based on the case review and data review results of the problematic drugs, the potential impact of the problematic drugs was analyzed, the names, associated adverse reactions, key parameters and impact ranges of all problematic drugs were recorded, and they were classified and sorted according to the degree of difference in impact to construct a list of problematic drugs.

7. The adverse drug reaction case database system according to claim 6, characterized in that: The steps for obtaining the optimized case database are: Based on the list of problematic drugs, classify them according to drug name, adverse reaction type and severity, mark cases with symptom description, medication time and dosage information, check data integrity, and obtain adverse reaction case marking results; According to the adverse reaction case marking results, the existing database search rules are analyzed, the indexing method of the problematic drug cases is adjusted, matching is performed according to the drug name, adverse reaction type and severity, and the data access rights are adjusted to obtain the database search function optimization results; Based on the optimization results of the database retrieval function, all marked adverse reaction cases are checked, the data storage structure is analyzed, the data is reorganized, the adverse reaction case retrieval and access of all problematic drugs are recorded, and an optimized case database is generated.

8. A method for a drug adverse reaction case database, characterized in that: The system according to any one of claims 1 to 7 is implemented, comprising the following steps: S1: Collect basic information, drug types and feedback data of drug users, classify them according to drug types, set classification standards for the severity of adverse reactions, calculate the comprehensive score of adverse reactions of drugs, statistically analyze the adverse reaction trends of different drug types according to time distribution, and output adverse reaction case tables; S2: parsing the case content according to the adverse reaction case table, performing data mapping according to drug name and adverse reaction category, calculating the association strength between drugs and adverse reactions, extracting the main association pattern and the secondary association pattern and sorting the drugs, determining the overlapping ratio of adverse reactions shared between drugs, determining the key adverse reaction nodes, and constructing a case association graph; S3: Based on the case correlation diagram, the changing trend of adverse reactions in different time periods is counted, the impact value of each adverse reaction is calculated, the difference in the impact of adverse reactions is analyzed in combination with the characteristics of different populations, the sensitivity of the impact of different adverse reactions on drug safety is analyzed, the most critical adverse reaction problems are determined, and the sensitivity analysis results are obtained; S4: Based on the sensitivity analysis results, extract drugs with high sensitivity to adverse reactions through thresholds, analyze the performance differences of drugs in different user populations, analyze the distribution of key parameters in different cases, evaluate the performance of problem drugs in different use environments, analyze the potential impact of problem drugs, and build a list of problem drugs; S5: Based on the list of problematic drugs, classify them according to drug name, adverse reaction type and severity, mark the cases, adjust the indexing method of problematic drug cases, match them according to drug name, adverse reaction type and severity, analyze the data storage structure, reorganize the data, and generate an optimized case database.

Images