Search

CN-122000083-A - Clinical treatment safety-based adverse drug reaction reporting data management system

CN122000083ACN 122000083 ACN122000083 ACN 122000083ACN-122000083-A

Abstract

The invention relates to the technical field of drug safety data management, and discloses a drug adverse reaction reporting data management system based on clinical treatment safety. The system comprises a data integration module, an event gathering module, a feature extraction module, a risk assessment module and a queue generation module which are connected in sequence. The system automatically aggregates discrete events with time adjacency and the same drug identification into a unified analysis unit through an event aggregation module, a feature extraction module constructs a structured unit symptom spectrum and extracts multidimensional medication features according to the structured unit symptom spectrum, a risk assessment module judges potential risk levels of the units through analysis of correlation between symptom spectrum evolution modes and medication features, and a queue generation module generates a priority treatment queue according to the risk levels. The system realizes intelligent identification and risk depth mining of the relevance adverse reaction, and effectively improves analysis precision and early warning timeliness of drug safety monitoring.

Inventors

  • SU DAN

Assignees

  • 常州市第二人民医院
  • 医顺通信息科技(江苏)有限公司

Dates

Publication Date
20260508
Application Date
20260130

Claims (10)

  1. 1. The utility model provides a medicine adverse reaction report data management system based on clinical treatment safety which characterized in that includes: The data integration module is used for receiving and integrating multi-dimensional adverse drug reaction event sets of different medical sources, wherein the multi-dimensional adverse drug reaction event sets at least comprise drug identification, event time stamp, adverse reaction symptom code, administration route and treatment period information; The event collecting module is used for identifying a plurality of reporting events which have time adjacency and relate to the same drug identification from the multi-dimensional adverse drug reaction event set and collecting the reporting events into an analysis unit; the feature extraction module is used for extracting adverse reaction symptom codes of all reported events in each analysis unit, constructing a unit symptom spectrum according to a code level, simultaneously acquiring the administration route and treatment period information corresponding to each reported event and generating unit medication features; The risk assessment module is used for judging and analyzing potential risk levels of the unit according to the combination condition of the change mode of the unit symptom spectrum and the unit medication characteristics; and the queue generating module is used for dynamically sequencing and layering the reported events in the analysis unit according to the potential risk level to generate an event handling queue.
  2. 2. The adverse drug reaction reporting data management system based on clinical treatment safety according to claim 1, wherein the identifying of a plurality of reporting events that exist in temporal proximity and relate to the same drug identity from a multi-dimensional adverse drug reaction event set is specifically: For any medicine identifier, searching all reporting events under the medicine identifier in the time dimension; calculating an event time stamp interval between every two reporting events under the medicine identifier; Setting a time window threshold, and merging all reported events with event time stamp intervals smaller than the time window threshold into a temporary event cluster; And checking whether the treatment period information of all reported events in each temporary event cluster has overlapping or continuity, and if so, confirming the temporary event cluster as an analysis unit.
  3. 3. The clinical treatment safety-based adverse drug reaction reporting data management system according to claim 2, wherein the constructing unit symptom spectrum according to the coding hierarchy is specifically: Acquiring an adverse reaction symptom code of each reported event in the analysis unit, wherein the adverse reaction symptom code is a grading code containing system organ classification and specific symptom terms; taking the classification of system organs as class, and counting the occurrence frequency of each level of adverse reaction symptom codes in the analysis unit; According to the statistical result, a tree-like hierarchical structure which takes the system organ classification as a horizontal axis and the occurrence frequency of specific symptom terms as a vertical axis is constructed, wherein the tree-like hierarchical structure is a unit symptom spectrum.
  4. 4. The clinical treatment safety-based adverse drug reaction reporting data management system of claim 3, wherein the generation of the unit medication features is specifically: Extracting the information of the drug administration routes related to all reported events in the analysis unit, and recording the occurrence times of different drug administration routes; Extracting treatment period information corresponding to all reported events in the analysis unit, and calculating the shortest treatment period, the longest treatment period and the average value of the treatment periods; and combining the statistical information of the administration route with the calculation result of the treatment period, and packaging the statistical information and the calculation result of the treatment period into a structured data object, wherein the structured data object is the unit medication characteristic.
  5. 5. The clinical treatment safety-based adverse drug reaction reporting data management system according to claim 4, wherein the determining and analyzing the potential risk level of the unit according to the combination of the variation pattern of the unit symptom spectrum and the unit medication characteristic specifically comprises: traversing the tree hierarchy of the unit symptom spectrum, and identifying specific symptom terms which appear frequently and show increasing trend on a time sequence; Counting the total number of specific symptom terms with increasing trend, and calculating to obtain a symptom dynamic score by combining the severity preset weight of the system organ classification to which the specific symptom terms belong; reading the average value of the treatment period, and comparing the average value with a preset standard treatment period to obtain a period deviation index; and fusing the symptom dynamic scores, the risk adjustment coefficients and the period deviation indexes, and outputting potential risk grades of the analysis unit through a preset rating rule mapping table.
  6. 6. The adverse drug reaction reporting data management system based on clinical treatment safety as claimed in claim 5, wherein the dynamic ordering and layering of reporting events in the analysis unit according to the potential risk level specifically comprises: assigning a base ranking weight to the analysis unit according to its potential risk level; for each reported event in the analysis unit, calculating an individual risk score of each reported event by combining a severity preset level of adverse reaction symptom codes, a new and old degree of event time stamps and a magnitude of deviation of a treatment period from a standard value; Multiplying the basic sorting weight with the individual risk score of the reported event to obtain the comprehensive sorting score of the reported event in the queue; and sequencing the comprehensive sequencing scores of all the reported events in all the analysis units from high to low, and automatically dividing the reported events into different levels of treatment channels according to the score intervals.
  7. 7. The medication adverse reaction reporting data management system based on clinical treatment safety according to claim 6, further comprising a method for managing feedback information of reporting events for which treatment has been completed, specifically: after a reporting event in a certain analysis unit is treated and clinical feedback information is obtained, the feedback information is stored in association with a unit symptom spectrum and unit medication characteristics of the analysis unit; And verifying and calibrating potential risk level judgment results corresponding to the unit symptom spectrum change mode and the unit medication characteristic similar items of the current analysis unit in the rating rule mapping table by using feedback information.
  8. 8. The adverse drug reaction report data management system based on clinical treatment safety according to claim 7, wherein the step of verifying and calibrating the potential risk level determination result corresponding to the unit symptom spectrum change pattern and the unit medication feature similarity item of the current analysis unit in the rating rule mapping table by using feedback information is specifically as follows: Taking a confirmation result about adverse reaction and medication relevance in the feedback information as a verification label; searching all record entries which obtain the same potential risk level as the current analysis unit in a rating rule mapping table; comparing the unit symptom spectrum change mode and the unit medication characteristic of the record item with the similarity between the unit symptom spectrum change mode and the unit medication characteristic of the current analysis unit; And adjusting the rating rule logic corresponding to the record entries with high similarity according to the verification tags.
  9. 9. The clinical treatment safety-based adverse drug reaction reporting data management system of claim 1, further comprising a data preprocessing step prior to receiving and integrating the multi-dimensional adverse drug reaction event sets of different medical sources: carrying out standardized cleaning on the original data reported by different medical sources, and unifying the formats of medicine identification, adverse reaction symptom coding and time stamp; Identifying and eliminating duplicate records reported repeatedly; and checking the integrity of the necessary filling field of the data, and marking and isolating the record with the missing key field.
  10. 10. The adverse drug reaction reporting data management system based on clinical treatment safety according to claim 1, further comprising a queue update and trace back step after generating the event handling queue: Periodically scanning a newly received multi-dimensional drug adverse reaction event set, and judging whether a newly reported event can be classified into an existing analysis unit or form a new analysis unit; for the new reporting events which can be classified into the existing analysis unit, immediately updating the unit symptom spectrum and the unit medication characteristics of the analysis unit, and recalculating the potential risk grade of the analysis unit and the comprehensive sequencing scores of all reporting events in the unit according to the updated information; based on the recalculated result, the ordering and hierarchical attribution of the related reporting events in the event handling queue are dynamically adjusted, and meanwhile, the traceable log of each queue change is recorded.

Description

Clinical treatment safety-based adverse drug reaction reporting data management system Technical Field The invention relates to the technical field of drug safety data management, in particular to a drug adverse reaction reporting data management system based on clinical treatment safety. Background In current drug alert practice, medical information systems generally rely on spontaneous reporting systems to receive scattered adverse drug reaction event reports. These systems typically treat each report as a separate database record focusing on the entry, storage, and query statistics of the underlying fields of medicine, symptoms, time, etc. of a single event. Prior art schemes lack an effective identification mechanism for potential clinical associations between events when processing data, defaulting to all reports to be statistically independent of each other. The prior art has the defects of isolation and statics of analysis modes. Because the system cannot effectively connect multiple associated events of the same patient or the same medicine in series in the treatment period, the cognition of the adverse reaction stays at a single point moment, and the dynamic evolution process of the adverse reaction in the time dimension is difficult to restore. Meanwhile, the conventional risk assessment is mostly based on single reported symptom severity or population incidence calculation, and the structural evolution of the medication characteristics and adverse reaction symptom spectrum cannot be subjected to deep correlation analysis, so that a plurality of gradually developed risk modes which depend on specific medication backgrounds are easily ignored, and the sensitivity and the specificity of early warning are insufficient. The invention aims to solve the problem of how to automatically identify and collect adverse reaction event sequences with clinical time sequence relevance from mass discrete reporting events. The invention also needs to solve the problem of how to integrate medication context information and intelligently analyze the dynamic change mode of the structured symptom spectrum, thereby realizing accurate relevance risk assessment exceeding single-point statistics. Disclosure of Invention The invention aims to provide a clinical treatment safety-based adverse drug reaction reporting data management system so as to solve the problems in the background technology. To achieve the above object, the present invention provides a clinical treatment safety-based adverse drug reaction reporting data management system, the system comprising: The data integration module is used for receiving and integrating multi-dimensional adverse drug reaction event sets of different medical sources, wherein the multi-dimensional adverse drug reaction event sets at least comprise drug identification, event time stamp, adverse reaction symptom code, administration route and treatment period information; The event collecting module is used for identifying a plurality of reporting events which have time adjacency and relate to the same drug identification from the multi-dimensional adverse drug reaction event set and collecting the reporting events into an analysis unit; the feature extraction module is used for extracting adverse reaction symptom codes of all reported events in each analysis unit, constructing a unit symptom spectrum according to a code level, simultaneously acquiring the administration route and treatment period information corresponding to each reported event and generating unit medication features; The risk assessment module is used for judging and analyzing potential risk levels of the unit according to the combination condition of the change mode of the unit symptom spectrum and the unit medication characteristics; and the queue generating module is used for dynamically sequencing and layering the reported events in the analysis unit according to the potential risk level to generate an event handling queue. Preferably, the identifying a plurality of reporting events which have time adjacency and relate to the same drug identifier from the multi-dimensional adverse drug reaction event set specifically includes: For any medicine identifier, searching all reporting events under the medicine identifier in the time dimension; calculating an event time stamp interval between every two reporting events under the medicine identifier; Setting a time window threshold, and merging all reported events with event time stamp intervals smaller than the time window threshold into a temporary event cluster; And checking whether the treatment period information of all reported events in each temporary event cluster has overlapping or continuity, and if so, confirming the temporary event cluster as an analysis unit. Preferably, the building unit symptom spectrum according to the coding hierarchy specifically includes: Acquiring an adverse reaction symptom code of each reported event in the analysis unit, wherein th