Elderly co-disease intervention effect multi-dimensional evaluation index system construction method and data acquisition terminal

Abstract

The invention discloses a multi-dimensional evaluation index system construction method and a data acquisition terminal for an old-age co-disease intervention effect, and relates to the technical field of medical health, and the method comprises the steps: dividing an intervention full cycle into a baseline period, an acute dry expectation period, a functional recovery period and a maintenance consolidation period, and constructing a multi-dimensional staged framework including biomedicine, functional states and the like; establishing a staged initial index pool, and screening core indexes through a Delphi method, factor analysis and clustering analysis; combining an analytic hierarchy process and an entropy evaluation method to determine the differentiation weight of the indexes in each stage; setting a quantization trigger condition of stage conversion; the system is verified and dynamically improved through phased reliability and validity analysis, the problems that an existing evaluation system is single in dimension and lacks dynamic adjustment are solved, the senile co-disease intervention effect can be scientifically and comprehensively evaluated, and support is provided for optimizing an intervention scheme.

Description

Technical Field

[0001] This invention relates to the field of medical and health technology, specifically to a method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly and a data acquisition terminal. Background Technology

[0003] Currently, existing methods for constructing evaluation index systems for interventions targeting comorbidities in the elderly have several shortcomings: First, the evaluation dimensions are too singular, focusing primarily on biomedical disease control indicators such as blood pressure and blood sugar, neglecting important aspects such as the functional status and psychosocial condition of the elderly, and failing to comprehensively reflect the impact of interventions on the overall health of the elderly; second, the indicators lack specificity, mostly using evaluation indicators for single diseases, without fully considering the particularities of comorbidities in the elderly, such as the interaction between diseases and the effects of multiple medications; third, there is a lack of dynamic adjustment mechanisms, and once the indicator system is constructed, it remains fixed, making it difficult to adapt to the evaluation needs of different intervention stages and different types of elderly comorbidity groups; fourth, some indicators are difficult to obtain and have poor operability, hindering their promotion and application in primary healthcare institutions. Summary of the Invention

[0004] To address the aforementioned technical shortcomings, the present invention aims to provide a method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly and a data acquisition terminal.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly, including: Step 1: Based on the division of the entire intervention cycle into stages and constructing a staged dimensional framework, the intervention process is divided into the baseline period, the acute intervention period, the functional recovery period, and the maintenance and consolidation period. For each stage, a staged first-level evaluation dimensional system including biomedical dimension, functional state dimension, psychosocial dimension, medical resource utilization dimension, and quality of life dimension is constructed. The dimensional framework of each stage is dynamically adjusted according to the intervention goal.

[0006] Step 2: Establish initial indicator pools for each stage. Through literature search, clinical practice surveys, and consultations with patients and caregivers, collect specific indicators for each stage and common indicators across stages. After preliminary screening, initial indicator pools for each stage are formed.

[0007] Step 3: Phased indicator optimization and screening. Using the Delphi method combined with factor analysis and cluster analysis, the core indicator set for each phase is obtained by screening the initial indicator pool for each phase.

[0008] Step 4: Determine the weights of indicators in each stage. After initial weighting in each stage using the analytic hierarchy process (AHP), the weights are corrected using the entropy method based on the measured data of the corresponding stage to obtain the differentiated weights of the indicators in each stage.

[0009] Step 5: Establish a phase transition mechanism, and set quantitative triggering conditions for each phase transition and cross-phase indicator connection rules.

[0010] Step Six: Phased Validation and Dynamic Improvement. The indicator system is validated through reliability and validity analyses at each stage, and the system is revised and improved based on feedback, ultimately forming a formal phased multi-dimensional evaluation indicator system.

[0011] Preferably, the construction of the phased dimensional framework specifically includes: the baseline dimensional framework includes basic state indicators, the biomedical dimension covers the list of diagnosed diseases, baseline physiological parameters, and the list of multiple medications, and the functional state dimension includes the baseline ADL / Barthel index and basic cognitive state.

[0012] The acute intervention dimension framework strengthens acute improvement indicators, the biomedical dimension adds symptom control indicators and adverse drug reaction rates, and the medical resource utilization dimension adds the number of emergency medical interventions.

[0013] The functional recovery period dimension framework focuses on functional recovery indicators. The functional status dimension adds rehabilitation training completion rate and muscle strength improvement, while the psychosocial dimension adds social role adaptability score.

[0014] The framework for the maintenance and consolidation period emphasizes maintenance indicators; the quality of life dimension adds annual life satisfaction tracking scores; the biomedical dimension adds the cumulative incidence of complications; and the medical resource utilization dimension adds the annual unplanned readmission rate.

[0015] Preferably, the initial indicator pools formed after preliminary screening include measurability screening, relevance screening, and representativeness screening.

[0016] The measurability screening specifically involves calculating the measurability score of each indicator, comparing it with a preset measurability score threshold in the database, and determining that the indicator is not measurable if the measurability score of an indicator is lower than the preset measurability score threshold in the database, otherwise determining that the indicator is measurable. The indicators that are measurable are then statistically analyzed.

[0017] The correlation screening specifically involves calculating the correlation coefficient of each indicator and comparing it with a preset correlation coefficient threshold in the database. If the correlation coefficient of an indicator is lower than the preset correlation coefficient threshold in the database, the indicator is determined to be uncorrelated; otherwise, the indicator is determined to be correlated. The indicators that are correlated are then statistically analyzed.

[0018] The representativeness screening specifically involves calculating the representative score of each indicator, comparing it with the preset representative scores in the database, and determining that the representativeness of an indicator is not met if the representativeness score of an indicator is lower than the preset representative score threshold in the database, otherwise the representativeness of the indicator is met, and statistically obtaining the representativeness of each indicator.

[0019] Based on the above analysis, indicators that meet the criteria of measurability, relevance, and representativeness were selected and compiled to form the initial indicator pool for each stage.

[0020] Preferably, the process of selecting the core indicator set for each stage from the initial indicator pool specifically involves:

[0021] Based on the indicators in the initial indicator pool at each stage, the Delphi method is used to analyze and obtain the average score and coefficient of variation of each indicator. These are then compared with the preset average score threshold and coefficient of variation threshold in the database. If the average score of an indicator is higher than the preset average score threshold in the database and the coefficient of variation of the indicator is higher than the preset coefficient of variation threshold in the database, then the indicator is preliminarily determined to be a core indicator. The indicators that can be preliminarily used as core indicators are then statistically obtained.

[0022] Factor analysis is performed on the indicators that can be initially used as core indicators to obtain the correlation coefficient between each indicator and the factor. The correlation coefficient is then compared with the preset correlation coefficient threshold between indicators and factors in the database. If the correlation coefficient between an indicator and a factor is higher than the preset correlation coefficient threshold between indicators and factors in the database, the indicator is then determined to be a core indicator. The indicators that can be determined to be core indicators in the second step are then statistically analyzed.

[0023] Cluster analysis was performed on the indicators that could be used as core indicators in the secondary judgment, and each indicator was selected and summarized to obtain the core indicator set for each stage.

[0024] Preferably, obtaining the differentiated weights of indicators at each stage specifically involves: assigning expert weights to the core indicators at each stage using the analytic hierarchy process to obtain a preliminary weight distribution; obtaining actual intervention data for each stage from the database; and correcting the preliminary weights using the entropy method to obtain the differentiated indicator weights for the actual intervention effects at each stage.

[0025] Preferably, the establishment of the phase transition mechanism specifically involves setting quantitative triggering conditions for each phase transition based on the intervention goals and indicator change trends of each phase.

[0026] Preferably, the reliability analysis at each stage specifically involves: calculating the reliability coefficient of each core indicator at each stage based on the core indicator set at each stage, comparing it with a preset reliability coefficient threshold in the database, and determining that the reliability of a core indicator at a certain stage is consistent with the evaluation indicator system at that stage if the reliability coefficient of a core indicator at a certain stage is higher than the preset reliability coefficient threshold in the database. Otherwise, determining that the reliability of the core indicator is inconsistent with the evaluation indicator system at that stage.

[0027] Preferably, the validity analysis at each stage specifically involves: calculating the validity index of each core indicator at each stage based on the core indicator set at each stage, comparing it with a preset validity index threshold in the database, and determining that the validity of a core indicator at a certain stage conforms to the evaluation indicator system at that stage if the validity index of a core indicator at a certain stage is higher than the preset validity index threshold in the database; otherwise, determining that the validity of the core indicator does not conform to the evaluation indicator system at that stage.

[0028] Preferably, the final formal phased multi-dimensional evaluation index system includes: the name, definition, quantification standard, data source, weight and scoring method of the core indicators of each phase and dimension, and is stored in the database in tabular form.

[0029] The second aspect of this invention provides a data acquisition terminal for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly, comprising: a phase division and dimension framework construction module: responsible for dividing the intervention process into the baseline period, acute intervention period, functional recovery period, and maintenance and consolidation period, and constructing a first-level evaluation dimension system containing five dimensions for each stage to ensure that the dimension framework of each stage is adapted to the intervention goal.

[0030] The initial indicator pool establishment module for each stage: Through literature retrieval, clinical surveys and patient consultations, specific and common indicators for each stage are collected. After screening for measurability, relevance and representativeness, an initial indicator pool for each stage is formed.

[0031] Indicator optimization and screening module: Based on the initial indicator pool, the Delphi method is used to analyze the average score and coefficient of variation of the indicators. Combined with factor analysis and cluster analysis, the core indicator set for each stage is screened out.

[0032] The phased indicator weight determination module uses the analytic hierarchy process (AHP) to assign initial weights by experts, and then uses the entropy method to correct the weights based on the measured data from each phase, thus obtaining the differentiated weights of the indicators at each phase and reflecting the differences in the actual intervention effects.

[0033] Phase transition mechanism module: Based on the intervention goals and indicator change trends of each phase, quantitative trigger conditions and cross-phase indicator connection rules are set to achieve precise transition between phases.

[0034] The phased verification and dynamic improvement module calculates the credibility coefficient and validity index of the core indicators at each stage, compares them with preset thresholds to verify the applicability of the indicator system, and dynamically corrects them based on feedback, ultimately forming a formal phased multi-dimensional evaluation indicator system.

[0035] The beneficial effects of this invention are as follows: Comprehensiveness: It breaks through the limitations of a single biomedical dimension, covering five dimensions including biomedicine, functional status, and psychosocial aspects, and comprehensively reflects the impact of intervention for comorbidities in the elderly on overall health;

[0036] Phased Dynamic Adaptation: The intervention is divided into four phases based on the entire intervention cycle. The dimensional framework, indicators, and weights of each phase are dynamically adjusted according to the intervention goals to adapt to the specific needs of different stages of comorbidity in the elderly.

[0037] Scientific screening mechanism: The indicators are screened by combining multiple methods such as Delphi method, factor analysis, and cluster analysis. The weights are determined by combining the analytic hierarchy process and the entropy method to improve the scientific nature and pertinence of the indicator system.

[0038] Precise stage transition: Set quantitative trigger conditions and cross-stage indicator connection rules to achieve precise transition between stages and avoid over-intervention or under-intervention;

[0039] The verification mechanism is sound: the indicator system is verified through phased reliability and validity analysis and is dynamically improved to ensure its applicability and reliability in primary healthcare institutions. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention;

[0042] Figure 2 This is a schematic diagram of the terminal structure connection of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] according to Figure 1As shown, this invention provides a method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly, including: Step 1: Based on the division of the entire intervention cycle into stages and the construction of a staged dimensional framework, the intervention process is divided into the baseline period, the acute intervention period, the functional recovery period, and the maintenance and consolidation period. For each stage, a staged first-level evaluation dimensional system including biomedical dimension, functional state dimension, psychosocial dimension, medical resource utilization dimension, and quality of life dimension is constructed. The dimensional framework of each stage is dynamically adjusted according to the intervention goal.

[0045] In one specific embodiment, the construction of the phased dimensional framework is as follows: the baseline dimensional framework includes basic state indicators, the biomedical dimension covers the list of diagnosed diseases, baseline physiological parameters, and the list of multiple medications, and the functional state dimension includes the baseline ADL / Barthel index and basic cognitive state.

[0046] The acute intervention dimension framework strengthens acute improvement indicators, the biomedical dimension adds symptom control indicators and adverse drug reaction rates, and the medical resource utilization dimension adds the number of emergency medical interventions.

[0047] The functional recovery period dimension framework focuses on functional recovery indicators. The functional status dimension adds rehabilitation training completion rate and muscle strength improvement, while the psychosocial dimension adds social role adaptability score.

[0048] The framework for the maintenance and consolidation period emphasizes maintenance indicators; the quality of life dimension adds annual life satisfaction tracking scores; the biomedical dimension adds the cumulative incidence of complications; and the medical resource utilization dimension adds the annual unplanned readmission rate.

[0049] Step 2: Establish initial indicator pools for each stage. Through literature search, clinical practice surveys, and consultations with patients and caregivers, collect specific indicators for each stage and common indicators across stages. After preliminary screening, initial indicator pools for each stage are formed.

[0050] In one specific embodiment, the initial indicator pool formed after preliminary screening includes measurability screening, relevance screening, and representativeness screening.

[0051] The measurability screening specifically involves calculating the measurability score of each indicator, comparing it with a preset measurability score threshold in the database, and determining that the indicator is not measurable if the measurability score of an indicator is lower than the preset measurability score threshold in the database, otherwise determining that the indicator is measurable. The indicators that are measurable are then statistically analyzed.

[0052] It should be noted that the calculation of the measurability score for each indicator specifically involves setting up a measurability scoring table, having multiple experts score each indicator, and calculating the measurability score for each indicator. The calculation formula is as follows:

[0053] Among them, kc i pf represents the measurability score of the i-th indicator. ni This represents the score given by the nth expert to the i-th indicator, where n represents the expert's number (n = 1, 2, ..., m, where m is a positive integer greater than 2), and i represents the indicator's number (i = 1, 2, ..., j, where j is a positive integer greater than 2). The preset measurability scoring thresholds in the database are set by professionals. These thresholds need to be determined by combining the practical capabilities of primary healthcare institutions (such as the time and cost of data collection per case), patient cooperation (≥50%), and the availability of clinical data (≥60%). The basic thresholds are determined through expert consensus (Delphi method), and then adjusted according to the characteristics of the indicators at each stage (emphasis on short-term measurability in the acute phase and long-term availability in the maintenance phase), and finally finalized.

[0054] The correlation screening specifically involves calculating the correlation coefficient of each indicator and comparing it with a preset correlation coefficient threshold in the database. If the correlation coefficient of an indicator is lower than the preset correlation coefficient threshold in the database, the indicator is determined to be uncorrelated; otherwise, the indicator is determined to be correlated. The indicators that are correlated are then statistically analyzed.

[0055] It should be noted that the formula for calculating the correlation coefficient of each indicator is as follows: Among them xg i Let x represent the correlation coefficient of the i-th indicator. i This represents the value of the i-th indicator. y represents the average value of the index. i This represents the value of the primary outcome variable for the i-th indicator. This represents the average value of the primary outcome variable, which can be understood as the required value of the indicator at a certain stage. The correlation coefficient threshold preset in the database is based on statistical standards (such as a strong correlation threshold of 0.7) and is set differently in stages according to the clinical significance of the indicator (the threshold of the indicator in the acute phase can be slightly higher to ensure consistency, and can be appropriately relaxed in the maintenance phase). The initial value is determined after verification by expert consensus (Delphi method), and then dynamically calibrated based on the actual data distribution and clustering results, and finally determined.

[0056] The representativeness screening specifically involves calculating the representative score of each indicator, comparing it with the preset representative scores in the database, and determining that the representativeness of an indicator is not met if the representativeness score of an indicator is lower than the preset representative score threshold in the database, otherwise the representativeness of the indicator is met, and statistically obtaining the representativeness of each indicator.

[0057] It should be noted that the calculation formula for the representative score of each indicator is the same as the calculation formula for the measurability score.

[0058] Based on the above analysis, indicators that meet the criteria of measurability, relevance, and representativeness were selected and compiled to form the initial indicator pool for each stage.

[0059] Step 3: Phased indicator optimization and screening. Using the Delphi method combined with factor analysis and cluster analysis, the core indicator set for each phase is obtained by screening the initial indicator pool for each phase.

[0060] In one specific embodiment, the process of selecting the core indicator set for each stage from the initial indicator pool for each stage specifically involves: analyzing each indicator in the initial indicator pool for each stage using the Delphi method to obtain the average score and coefficient of variation of each indicator, comparing them with the preset average score threshold and coefficient of variation threshold in the database respectively; if the average score of an indicator is higher than the preset average score threshold in the database and the coefficient of variation of the indicator is higher than the preset coefficient of variation threshold in the database, then the indicator is preliminarily determined to be a core indicator, and the indicators that can be preliminarily used as core indicators are statistically obtained.

[0061] It should be noted that the coefficient of variation is calculated using the following formula: Among them, CV i σ represents the coefficient of variation of the i-th index. i This represents the standard deviation of the score for the i-th indicator. The average score represents the i-th indicator. The preset average score threshold and coefficient of variation threshold in the database need to be combined with statistical standards and clinical significance: the average score threshold is selected based on expert scores (e.g., ≥4 points in the Delphi method 5-point system) to screen indicators that meet the importance criteria, and the coefficient of variation threshold is used as a benchmark to remove indicators with poor stability with a value of ≤0.25; at the same time, it is adjusted according to the stage (the coefficient of variation threshold is more stringent in the acute phase to ensure sensitivity, and appropriately relaxed in the maintenance phase to ensure stability), and dynamically calibrated after being verified by clinical data.

[0062] Factor analysis is performed on the indicators that can be initially used as core indicators to obtain the correlation coefficient between each indicator and the factor. The correlation coefficient is then compared with the preset correlation coefficient threshold between indicators and factors in the database. If the correlation coefficient between an indicator and a factor is higher than the preset correlation coefficient threshold between indicators and factors in the database, the indicator is then determined to be a core indicator. The indicators that can be determined to be core indicators in the second step are then statistically analyzed.

[0063] It should be noted that the correlation coefficients between the indicators and factors, which are used to calculate factor loadings, are existing technologies and will not be elaborated upon here. The correlation coefficient thresholds between the indicators and factors preset in the database are based on statistical conventions (absolute factor loading ≥ 0.5), and are adjusted in stages according to the clinical correlation strength of the indicators (the threshold can be increased to ≥ 0.6 in the acute phase to maintain sensitivity, and relaxed to ≥ 0.45 in the maintenance phase to maintain stability). The initial values ​​are determined after verification by expert consensus, and then dynamically calibrated according to the actual factor analysis results to ensure that the indicators can effectively reflect the potential constructs of the corresponding factors.

[0064] Cluster analysis was performed on the indicators that could be used as core indicators in the secondary judgment, and each indicator was selected and summarized to obtain the core indicator set for each stage.

[0065] It should be noted that the cluster analysis first standardizes the indicator data to eliminate the influence of dimensions. Then, a hierarchical clustering method (or K-means clustering) is used to determine the optimal number of clusters through silhouette coefficients and scree plots. Indicators with high correlation (correlation coefficient > 0.7) are grouped into one class. In each class, the indicator with the smallest coefficient of variation and the strongest representativeness (such as the one with the highest factor loading or the best clinical sensitivity) is selected. Combined with the intervention goals of each stage (the acute phase focuses on highly sensitive short-term indicators, and the maintenance and consolidation phase focuses on highly stable long-term indicators), redundant indicators are eliminated and the results are summarized to finally obtain the core indicator set for each stage.

[0066] It should be noted that the specific process is as follows: The Delphi method involves multiple rounds of anonymous questionnaire scoring by multidisciplinary experts (such as geriatrics, nursing, rehabilitation, psychology, social work, etc.). In each round, experts score each indicator in the initial indicator pool at each stage from dimensions such as importance, measurability, relevance, and representativeness. The average score and coefficient of variation of each indicator are calculated. Indicators below the set threshold are eliminated. Multiple rounds of feedback and correction are conducted until the experts' opinions tend to be consistent. For the indicators that pass the initial screening by the Delphi method, actual sample data are collected, and factor analysis or principal component analysis is performed to extract common factors that can explain most of the variance. Indicators with high factor loadings (e.g., >0.5) are retained, while redundant or low-contribution indicators are removed. Cluster analysis is performed on the remaining indicators to group indicators with high similarity into one category. The most representative indicators in each category are retained first to reduce indicator redundancy.

[0067] During the acute intervention period, indicators with high sensitivity and significant short-term changes (such as symptom scores and acute complications) are prioritized. During the maintenance and consolidation period, indicators with high stability and high long-term follow-up value (such as quality of life and readmission rate) are prioritized. Finally, based on expert consensus and statistical analysis results, the core indicator set for each stage is determined to form a phased core indicator list.

[0068] Step 4: Determine the weights of indicators in each stage. After initial weighting in each stage using the analytic hierarchy process (AHP), the weights are corrected using the entropy method based on the measured data of the corresponding stage to obtain the differentiated weights of the indicators in each stage.

[0069] In one specific embodiment, obtaining the differentiated weights of indicators at each stage involves: assigning expert weights to the core indicators at each stage using the analytic hierarchy process to obtain a preliminary weight distribution; obtaining actual intervention data for each stage from the database; and correcting the preliminary weights using the entropy method to obtain the differentiated indicator weights for the actual intervention effects at each stage.

[0070] The specific process of the Analytic Hierarchy Process (AHP) is as follows: Construct a hierarchical structure model: For each stage, construct a three-layer hierarchical structure of "target layer - criterion layer - indicator layer".

[0071] Target layer: Evaluation of the intervention effect at this stage.

[0072] Criterion layer: Biomedical, functional status, psychosocial, medical resource utilization, quality of life and other dimensions.

[0073] Indicator layer: Specific evaluation indicators under each dimension.

[0074] Constructing the judgment matrix: Experts use the 1-9 scale method to compare elements at the same level pairwise to construct the judgment matrix A.

[0075] Calculate the weight vector: The weight vector W is calculated using the eigenvalue method: AW = λ mx W, where λ mx Let W be the largest eigenvalue and W be the weight vector. The components of the weight vector W are calculated as follows: Among them, a ij To determine the elements of matrix A, nc is the matrix order, and k represents the value of the attribute of the index.

[0076] Consistency check: The formula for calculating the consistency ratio is: in RI stands for Random Consistency Index.

[0077] The specific process of the entropy method is as follows: Standardize the collected measured data from each stage.

[0078] Where X pq Given the original value of the q-th indicator for the p-th sample, calculate the entropy value of the q-th indicator: in Calculate the entropy weight of the q-th index: Combine AHP weights with entropy weights: w″ q =αw q +(1-α)w' q .

[0079] Baseline option re-determination: Focuses on basic state indicators, with higher weighting for the biomedical dimension.

[0080] The weighting of acute intervention is determined by focusing on symptom control and acute complications, with higher weightings for biomedical and medical resource utilization dimensions.

[0081] Functional recovery options are redefined: the focus is on functional recovery and rehabilitation effects, with higher weighting for functional status and psychosocial dimensions.

[0082] Maintaining and consolidating options requires a focus on long-term effects and stability, with higher weighting for dimensions such as quality of life and utilization of medical resources.

[0083] Step 5: Establish a phase transition mechanism, and set quantitative triggering conditions for each phase transition and cross-phase indicator connection rules.

[0084] In one specific embodiment, the establishment of the phase transition mechanism specifically involves setting quantitative triggering conditions for each phase transition based on the intervention goals and indicator change trends of each phase.

[0085] It should be noted that, based on the intervention goals and indicator trends at each stage, the quantitative triggering conditions for the transition between each stage are set by professionals. For example, based on the intervention goals (baseline assessment of baseline status, acute intervention control of acute symptoms, functional recovery promotion of functional recovery, and maintenance and consolidation to maintain stable status) and indicator trends at each stage, after cluster analysis of the core indicators for secondary assessment, the key dimensions of each stage are first identified (e.g., the acute intervention stage focuses on the biomedical and medical resource utilization dimension, and the functional recovery stage focuses on the functional status dimension). Then, based on the indicator distribution characteristics (e.g., clinical warning lines, change range thresholds), quantitative standards are set (e.g., the acute intervention stage requires core biological indicators to meet the standards and remain stable for 72 hours, and the functional recovery stage requires an ADL score that improves by ≥20% from the baseline and lasts for 1 month). Combined with time thresholds (e.g., duration of symptom stabilization) and cross-stage transition rules (the terminal indicator of the previous stage serves as the baseline for the next stage), and by verifying and adjusting the thresholds through clinical data, the quantitative triggering conditions for the transition between each stage are finally formed.

[0086] For example: Baseline period → Acute intervention transition conditions

[0087] The transition is triggered when a patient experiences an acute worsening of symptoms or the onset of new complications. Specific quantitative criteria include: a symptom score ≥7 (out of 10), abnormal blood pressure (>180 / 110 mmHg or <90 / 60 mmHg), abnormal blood glucose (>16.7 mmol / L or <3.9 mmol / L), or the occurrence of an acute cardiovascular or cerebrovascular event.

[0088] Conditions for transition from acute intervention to functional recovery period

[0089] The conversion is triggered when acute symptoms are effectively controlled and vital signs are stable. Specific quantitative conditions include: symptom score ≤3 (out of 10), stable vital signs for ≥48 hours, no new acute complications, and the patient is conscious and can cooperate with rehabilitation training.

[0090] Conditions for transitioning from the functional recovery period to the maintenance and consolidation period

[0091] The conversion is triggered when the main functional indicators reach the expected goals. Specific quantitative conditions include: ADL score ≥ 60 points (out of 100), cognitive function score ≥ 24 points (out of 30), social function score ≥ 70 points (out of 100), rehabilitation training completion rate ≥ 80%, etc.

[0092] Step Six: Phased Validation and Dynamic Improvement. The indicator system is validated through reliability and validity analyses at each stage, and the system is revised and improved based on feedback, ultimately forming a formal phased multi-dimensional evaluation indicator system.

[0093] In one specific embodiment, the reliability analysis at each stage is specifically as follows: based on the core indicator set of each stage, the reliability coefficient of each core indicator in each stage is calculated, and compared with the preset reliability coefficient threshold in the database. If the reliability coefficient of a core indicator in a certain stage is higher than the preset reliability coefficient threshold in the database, it is determined that the reliability of the core indicator conforms to the evaluation indicator system of that stage; otherwise, it is determined that the reliability of the core indicator does not conform to the evaluation indicator system of that stage.

[0094] It should be noted that the reliability coefficients of each core indicator at each stage are calculated using Cronbach's α coefficient, which is existing technology and will not be elaborated here. The preset reliability coefficient thresholds in the database are based on statistical standards (α ≥ 0.7 is acceptable) and are set differently according to the characteristics of each stage (≥ 0.85 in the acute phase to ensure the reliability of emergency intervention, ≥ 0.78 in the maintenance phase). After the initial values ​​are verified by expert consensus, they are dynamically calibrated based on the reliability analysis results of the measured data at each stage to ensure the consistency of the indicator system and its adaptation to the stage evaluation needs.

[0095] In one specific embodiment, the validity analysis at each stage is specifically as follows: based on the core indicator set of each stage, the validity index of each core indicator at each stage is calculated, and compared with the preset validity index threshold in the database. If the validity index of a core indicator at a certain stage is higher than the preset validity index threshold in the database, it is determined that the validity of the core indicator conforms to the evaluation indicator system of that stage; otherwise, it is determined that the validity of the core indicator does not conform to the evaluation indicator system of that stage.

[0096] It should be noted that the validity indices of each core indicator at each stage are calculated and will not be elaborated here. The validity index thresholds preset in the database are based on statistical norms and are set in combination with the characteristics of each stage (e.g., the structural validity threshold for the acute intervention period is ≥70% to ensure accuracy, and the maintenance period is ≥62%). After the initial values ​​are verified by expert consensus, they are dynamically calibrated based on the measured validity analysis results of each stage to ensure that the indicator system can effectively reflect the evaluation objectives of the corresponding stage.

[0097] In one specific embodiment, the final formal phased multi-dimensional evaluation index system includes: the name, definition, quantification standard, data source, weight and scoring method of the core indicators of each phase and dimension, and is stored in the database in the form of a table.

[0098] refer to Figure 2 A data acquisition terminal for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly includes: a phase division and dimension framework construction module: responsible for dividing the intervention process into the baseline period, acute intervention period, functional recovery period, and maintenance and consolidation period, and constructing a first-level evaluation dimension system with five dimensions for each stage to ensure that the dimension framework of each stage is adapted to the intervention goal.

[0099] The initial indicator pool establishment module for each stage: Through literature retrieval, clinical surveys and patient consultations, specific and common indicators for each stage are collected. After screening for measurability, relevance and representativeness, an initial indicator pool for each stage is formed.

[0100] Indicator optimization and screening module: Based on the initial indicator pool, the Delphi method is used to analyze the average score and coefficient of variation of the indicators. Combined with factor analysis and cluster analysis, the core indicator set for each stage is screened out.

[0101] The phased indicator weight determination module uses the analytic hierarchy process (AHP) to assign initial weights by experts, and then uses the entropy method to correct the weights based on the measured data from each phase, thus obtaining the differentiated weights of the indicators at each phase and reflecting the differences in the actual intervention effects.

[0102] Phase transition mechanism module: Based on the intervention goals and indicator change trends of each phase, quantitative trigger conditions and cross-phase indicator connection rules are set to achieve precise transition between phases.

[0103] The phased verification and dynamic improvement module calculates the credibility coefficient and validity index of the core indicators at each stage, compares them with preset thresholds to verify the applicability of the indicator system, and dynamically corrects them based on feedback, ultimately forming a formal phased multi-dimensional evaluation indicator system.

[0104] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly, characterized in that, include: Step 1: Based on the division of the entire intervention cycle into stages and the construction of a stage-specific dimensional framework, the intervention process is divided into the baseline period, the acute intervention period, the functional recovery period, and the maintenance and consolidation period. For each stage, a stage-specific first-level evaluation dimensional system is constructed, which includes biomedical dimensions, functional status dimensions, psychosocial dimensions, medical resource utilization dimensions, and quality of life dimensions. The dimensional framework of each stage is dynamically adjusted according to the intervention goals. Step 2: Establish initial indicator pools for each stage. Through literature search, clinical practice surveys and consultations with patients and caregivers, collect specific indicators for each stage and common indicators across stages, and form initial indicator pools for each stage after preliminary screening. Step 3: Phased indicator optimization and screening. Using the Delphi method combined with factor analysis and cluster analysis, the core indicator set for each phase is obtained by screening the initial indicator pool for each phase. Step 4: Determine the weights of indicators in each stage. After initial weighting in each stage using the analytic hierarchy process (AHP), the weights are corrected using the entropy method based on the measured data of the corresponding stage to obtain the differentiated weights of the indicators in each stage. Step 5: Establish a phase transition mechanism, and set quantitative trigger conditions for each phase transition and rules for cross-phase indicator connection; Step Six: Phased Validation and Dynamic Improvement. The indicator system is validated through reliability and validity analyses at each stage, and the system is revised and improved based on feedback, ultimately forming a formal phased multi-dimensional evaluation indicator system.

2. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 1, characterized in that, The construction of the phased dimensional framework is specifically as follows: The baseline dimension framework includes basic status indicators, the biomedical dimension covers the list of diagnosed diseases, baseline physiological parameters, and the list of multiple medications, and the functional status dimension includes the baseline ADL / Barthel index and basic cognitive status. The acute intervention dimension framework strengthens acute improvement indicators, the biomedical dimension adds symptom control indicators and adverse drug reaction rates, and the medical resource utilization dimension adds the number of emergency medical interventions. The functional recovery period dimension framework focuses on functional recovery indicators, with the functional status dimension adding rehabilitation training completion rate and muscle strength improvement, and the psychosocial dimension adding social role adaptability score. The framework for the maintenance and consolidation period emphasizes maintenance indicators; the quality of life dimension adds annual life satisfaction tracking scores; the biomedical dimension adds the cumulative incidence of complications; and the medical resource utilization dimension adds the annual unplanned readmission rate.

3. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 2, characterized in that, The initial indicator pool for each stage is formed through preliminary screening, including measurability screening, relevance screening, and representativeness screening; The measurability screening specifically involves calculating the measurability score of each indicator, comparing it with a preset measurability score threshold in the database, and determining that the indicator is not measurable if the measurability score of an indicator is lower than the preset measurability score threshold in the database, otherwise determining that the indicator is measurable. The indicators that are measurable are then statistically analyzed. The correlation screening specifically involves calculating the correlation coefficient of each indicator and comparing it with a preset correlation coefficient threshold in the database. If the correlation coefficient of an indicator is lower than the preset correlation coefficient threshold in the database, the indicator is determined to be uncorrelated; otherwise, the indicator is determined to be correlated. The indicators that are correlated are statistically obtained. The representativeness screening specifically involves calculating the representativeness score of each indicator, comparing it with the preset representativeness score in the database, and determining that the representativeness of an indicator is not met if the representativeness score of an indicator is lower than the preset representativeness score threshold in the database, otherwise the representativeness of the indicator is met, and statistically obtaining the representativeness of each indicator. Based on the above analysis, indicators that meet the criteria of measurability, relevance, and representativeness were selected and compiled to form the initial indicator pool for each stage.

4. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 3, characterized in that, The process of selecting the core indicator set for each stage from the initial indicator pool is as follows: Based on the indicators in the initial indicator pool at each stage, the Delphi method is used to analyze and obtain the average score and coefficient of variation of each indicator. These are then compared with the preset average score threshold and coefficient of variation threshold in the database. If the average score of an indicator is higher than the preset average score threshold in the database and the coefficient of variation of the indicator is higher than the preset coefficient of variation threshold in the database, then the indicator is preliminarily determined to be a core indicator. The indicators that can be preliminarily used as core indicators are statistically obtained. Factor analysis is performed on each indicator that can be initially used as a core indicator to obtain the correlation coefficient between each indicator and the factor. The correlation coefficient is then compared with the preset correlation coefficient threshold between the indicator and the factor in the database. If the correlation coefficient between a certain indicator and the factor is higher than the preset correlation coefficient threshold between the indicator and the factor in the database, the indicator is then determined to be a core indicator. The indicators that can be determined to be core indicators in the second step are statistically obtained. Cluster analysis was performed on the indicators that could be used as core indicators in the secondary judgment, and each indicator was selected and summarized to obtain the core indicator set for each stage.

5. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 4, characterized in that, The differentiated weights for the indicators at each stage are obtained as follows: By using the analytic hierarchy process (AHP) to assign expert weights to the core indicators at each stage, a preliminary weight distribution was obtained. Actual intervention data for each stage was then obtained from the database, and the entropy method was used to correct the preliminary weights, resulting in indicator weights that differentiated the actual intervention effects at each stage.

6. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 5, characterized in that, The establishment of the phase transition mechanism is specifically as follows: Based on the intervention goals and indicator trends at each stage, quantitative trigger conditions for the transition at each stage are set.

7. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 6, characterized in that, The reliability analysis at each stage is as follows: Based on the core indicator set for each stage, the credibility coefficient of each core indicator in each stage is calculated and compared with the preset credibility coefficient threshold in the database. If the credibility coefficient of a core indicator in a certain stage is higher than the preset credibility coefficient threshold in the database, it is determined that the credibility of the core indicator meets the evaluation indicator system for that stage; otherwise, it is determined that the credibility of the core indicator does not meet the evaluation indicator system for that stage.

8. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 6, characterized in that, The validity analysis at each stage is as follows: Based on the core indicator set for each stage, the validity index of each core indicator for each stage is calculated and compared with the preset validity index threshold in the database. If the validity index of a core indicator for a certain stage is higher than the preset validity index threshold in the database, the validity of the core indicator is determined to meet the evaluation indicator system for that stage; otherwise, the validity of the core indicator is determined to not meet the evaluation indicator system for that stage.

9. The method for constructing a multi-dimensional evaluation index system for the intervention effect of comorbidities in the elderly according to claim 8, characterized in that, The final formal phased multi-dimensional evaluation index system includes: the name, definition, quantitative standard, data source, weight and scoring method of the core indicators for each phase and dimension, which are stored in the database in tabular form.

10. A data acquisition terminal that uses the method for constructing a multi-dimensional evaluation index system for intervention effects on comorbidities in the elderly as described in any one of claims 1-9, characterized in that, include: Phase Division and Dimensional Framework Construction Module: Responsible for dividing the intervention process into the baseline period, acute intervention period, functional recovery period, and maintenance and consolidation period, and constructing a primary evaluation dimension system with five dimensions for each phase to ensure that the dimensional framework of each phase is adapted to the intervention goals; Phased Initial Indicator Pool Establishment Module: Collect specific and common indicators for each phase through literature retrieval, clinical surveys and patient consultations, and form initial indicator pools for each phase after screening for measurability, relevance and representativeness; Indicator optimization and screening module: Based on the initial indicator pool, the Delphi method is used to analyze the average score and coefficient of variation of the indicators. Combined with factor analysis and cluster analysis, the core indicator set for each stage is screened. Phased indicator weight determination module: Initial weighting by experts is performed using the analytic hierarchy process (AHP), and then the weights are corrected using the entropy method based on the measured data of each phase to obtain the differentiated weights of the indicators at each phase, reflecting the differences in the actual intervention effects. Phase transition mechanism module: Based on the intervention goals and indicator change trends of each phase, set quantitative trigger conditions and cross-phase indicator connection rules to achieve precise transition between phases; The phased verification and dynamic improvement module calculates the credibility coefficient and validity index of the core indicators at each stage, compares them with preset thresholds to verify the applicability of the indicator system, and dynamically corrects them based on feedback, ultimately forming a formal phased multi-dimensional evaluation indicator system.

Images