Prognosis prediction method based on MDAS time trajectory of patients with delirium in palliative care

Abstract

The present application relates to the cross field of clinical epidemiology and biostatistics, and particularly relates to a prognosis prediction method based on MDAS time trajectory of delirium of patients in palliative care. The present application introduces the MDAS time trajectory of delirium into the prognosis prediction model, and provides a new prognosis prediction method for the clinic. The dynamic classification of delirium can distinguish patients with similar baseline severity but different progression modes. The prognosis prediction model constructed based on the trajectory of the change in the severity of delirium is a dynamic prediction model, which is helpful for more accurate prognosis stratification, provides personalized and dynamic delirium management, helps to improve the quality of palliative care and improve the quality of life of delirium patients in palliative care, and has good application prospect.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of clinical epidemiology and biostatistics, specifically to a prognostic prediction method based on the time trajectory of delirium MDAS in palliative care patients. Background Technology

[0002] Delirium is one of the most common neuropsychiatric syndromes in patients with advanced cancer in hospice wards. It can accelerate the deterioration of the disease and shorten survival time, seriously endangering patients' health and quality of life. The severity of delirium symptoms usually fluctuates over time, and the trajectory of delirium severity varies among patients. Although some studies in intensive care patients and the elderly population have shown that the trajectory of delirium severity has prognostic significance, current evidence remains limited. In particular, patients with advanced cancer in hospice wards have unique clinical characteristics, including a severe symptom burden, shorter survival time, and frequent use of opioids and benzodiazepines, all of which may exacerbate delirium.

[0003] Therefore, the evolution and progression of delirium in hospice patients may differ from those in other populations. Currently, there are few prognostic studies on delirium in this population, and the findings are inconsistent. Prognostic factors identified for delirium in hospice patients include reduced activity delirium, severe cognitive impairment, organ failure, age, delirium severity, and body mass index.

[0004] The Memorial Delirium Assessment Scale (MDAS) is an internationally widely used tool for quantitatively assessing the severity of delirium. It covers dimensions such as attention, orientation, perceptual impairment, and psychomotor activity, using a four-point scale (0-3) with a total score of 0-30. Higher scores indicate more severe delirium symptoms. However, traditional delirium research often uses single-point assessments, failing to capture the fluctuating nature of delirium.

[0005] Furthermore, predictive models based on other prognostic factors, using single-point-of-time data or simple time-window statistics, fail to fully utilize the continuous-time trajectory information of the MDAS score. Therefore, this application incorporates the delirium severity trajectory into the prognostic prediction model, which helps to dynamically predict the prognosis of delirium patients receiving palliative care, thereby guiding clinical decision-making. Summary of the Invention

[0006] To address the problems of existing technologies, this invention provides a prognostic prediction method based on the time trajectory of delirium MDAS in palliative care patients.

[0007] The first aspect of the present invention provides a prognostic prediction method based on the time trajectory of delirium MDAS, comprising the following steps: Step 1: Obtain parameters such as delirium MDAS time trajectory subgroup, age, comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score of palliative care patients; Step 2: Use the parameters obtained in Step 1 to calculate the instantaneous mortality risk of delirious patients receiving palliative care at time t. The calculation formula is as follows: in, This is the baseline risk function, which represents the change in mortality risk over time without the influence of covariates. The calculation formula is: =ln2 / median survival time, where median survival time is the median survival time of delirium patients in the severe remission trajectory group with increased activity; increased activity delirium, decreased activity delirium, and mixed delirium are classifications of motor subtypes of delirium; delirium in the severe remission trajectory group, delirium in the mild exacerbation trajectory group, and delirium in the severe exacerbation trajectory group are subgroups of delirium MDAS time trajectory; Step 3: Based on the calculation results of Step 2, stratify the mortality risk of delirium patients.

[0008] Preferably, the delirium MDAS time trajectory subgroup is determined through the following steps: The total MDAS score of delirium patients was assessed twice a week. The trajectory of the total MDAS score over time was fitted using the group-based trajectory model statistical method. Based on the fitted trajectory of the total MDAS score over time, the delirium MDAS time trajectory subgroup of the patients was determined.

[0009] Preferably, the delirium MDAS time trajectory subgroups are determined according to the following rules: trajectories with an initial MDAS score below 15 and a gradually increasing MDAS score over time represent mildly aggravated delirium; trajectories with an initial MDAS score greater than or equal to 15 and a progressively increasing MDAS score over time represent severely aggravated delirium; and trajectories with an initial MDAS score greater than or equal to 15 and a gradually decreasing MDAS score over time represent severely relieved delirium.

[0010] Preferably, the mortality risk stratification is performed as follows: calculate The results were used to stratify mortality risk. ≤5% is considered low-risk, 5% < ≤10% is considered the medium-risk group. >10% is considered high-risk.

[0011] A second aspect of the present invention provides a prognostic prediction system based on the time trajectory of delirium MDAS, for implementing the above-described method, comprising: The input module is configured to acquire parameters such as delirium MDAS time trajectory subgroup, age, comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score of delirium patients receiving palliative care; The module is configured to use the parameters obtained from the input module to calculate the mortality risk of delirious patients in palliative care at time t, using the following formula: in, This is the baseline risk function, which represents the change in mortality risk over time without the influence of covariates. The calculation formula is: =ln2 / median survival time, where median survival time is the median survival time of delirium patients in the severe remission trajectory group with increased activity; increased activity delirium, decreased activity delirium, and mixed delirium are classifications of motor subtypes of delirium; delirium in the severe remission trajectory group, delirium in the mild exacerbation trajectory group, and delirium in the severe exacerbation trajectory group are subgroups of delirium MDAS time trajectory; The stratification module is configured to stratify delirium patients based on the calculation results of the construction module to determine their mortality risk.

[0012] Preferably, the delirium MDAS time trajectory subgroup is determined through the following steps: The total MDAS score of delirium patients was assessed twice a week. The trajectory of the total MDAS score over time was fitted using the group-based trajectory model statistical method. Based on the fitted trajectory of the total MDAS score over time, the delirium MDAS time trajectory subgroup of the patients was determined.

[0013] Preferably, the delirium MDAS time trajectory subgroups are determined according to the following rules: trajectories with an initial MDAS score below 15 and a gradually increasing MDAS score over time represent mildly aggravated delirium; trajectories with an initial MDAS score greater than or equal to 15 and a progressively increasing MDAS score over time represent severely aggravated delirium; and trajectories with an initial MDAS score greater than or equal to 15 and a gradually decreasing MDAS score over time represent severely relieved delirium.

[0014] Preferably, the mortality risk stratification is performed as follows: calculate The results were used to stratify mortality risk. ≤5% is considered low-risk, 5% < ≤10% is considered the medium-risk group. >10% is considered high-risk.

[0015] A third aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program for implementing the above-described prognostic prediction system based on delirium MDAS time trajectory.

[0016] A fourth aspect of the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described above.

[0017] By adopting the technical solution of the present invention, the following beneficial effects can be obtained: This invention incorporates the MDAS time trajectory of delirium into a prognostic prediction model, providing a novel prognostic prediction method for clinical practice. This dynamic classification of delirium can distinguish between delirium patients with similar baseline severity but different progression patterns. The prognostic prediction model built based on the trajectory of delirium severity changes is a dynamic predictive model, which facilitates more accurate prognostic stratification and provides personalized and dynamic delirium management.

[0018] Obviously, based on the above description of the present invention, and according to common technical knowledge and conventional methods in the field, various other modifications, substitutions, or alterations can be made without departing from the basic technical concept of the present invention.

[0019] The following detailed embodiments further illustrate the above-described content of the present invention. However, this should not be construed as limiting the scope of the present invention to the following examples. All technologies implemented based on the above-described content of the present invention fall within the scope of the present invention. Attached Figure Description

[0020] Figure 1 The optimal trajectory subgroup for fitting the obtained delirium MDAS time trajectory. Detailed Implementation

[0021] The algorithms for data acquisition, transmission, storage, and processing steps not specifically described in the following embodiments can all be implemented using publicly available technologies.

[0022] Example 1: Prognostic prediction system and method based on the time trajectory of delirium MDAS in palliative care patients The system in this embodiment includes: The input module is configured to acquire parameters such as delirium MDAS time trajectory subgroup, age, comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score of delirium patients receiving palliative care; The module is configured to use the parameters obtained from the input module to calculate the mortality risk of delirious patients in palliative care at time t, using the following formula: in, This is the baseline risk function, which represents the change in mortality risk over time without the influence of covariates. The calculation formula is: =ln2 / median survival time, where median survival time is the median survival time of delirium patients in the severe remission trajectory group with increased activity; increased activity delirium, decreased activity delirium, and mixed delirium are classifications of motor subtypes of delirium; delirium in the severe remission trajectory group, delirium in the mild exacerbation trajectory group, and delirium in the severe exacerbation trajectory group are subgroups of delirium MDAS time trajectory; The stratification module is configured to stratify delirium patients based on the calculation results of the construction module, thus stratifying their mortality risk. ≤5% is considered low-risk, 5% < ≤10% is considered the medium-risk group. >10% are classified as high-risk. Delirium is managed in a stratified manner based on the results of mortality risk stratification, and mortality risk is dynamically assessed based on changes in delirium trajectory after intervention, with timely adjustments to the delirium management plan.

[0023] The following practical application case illustrates the method for prognostic prediction using the aforementioned prognostic prediction system based on delirium MDAS time trajectory. The specific steps are as follows: 1. Establish a prognostic prediction model based on the time trajectory of delirium MDAS in palliative care patients. (1) Fitting the time trajectory of delirium MDAS with different subgroup numbers and morphologies The Chinese version of the 3-Minute Diagnostic Confusion Assessment Method (3D-CAM) was used to screen patients with advanced cancer admitted to the palliative care ward for delirium. Patients diagnosed with delirium were assessed for delirium severity twice weekly using the Memorial Delirium Assessment Scale (MDAS). A group-based trajectory model was used to fit the trajectory of the MDAS total score over time. First, an appropriate model was selected based on the variable distribution type. In this study, the MDAS total score was used to assess delirium severity, and as a continuous variable, a censored normal distribution model was selected for fitting. Then, models with different numbers of subgroups and trajectory patterns were fitted: fitting began with fewer subgroups (from 2 to 5 groups), with each subgroup starting with a higher-order function. If the higher-order function was not statistically significant (P>0.05), lower-order functions were fitted (i.e., from order 3, 2 to 1), resulting in a series of trajectory models with different numbers and patterns of subgroups.

[0024] (2) Determine the optimal delirium MDAS time trajectory subgroup for delirium patients receiving palliative care. Subgroups of delirium MDAS time trajectories for delirium patients were determined based on the Bayesian information criterion (BIC) and average posterior probability (AvePP). Specifically, a smaller absolute BIC value indicates a better model fit; AvePP is the probability that each individual is correctly assigned to a specific trajectory, with AvePP > 0.7 considered acceptable for the model, and ensuring that the frequency of individuals in each subgroup is greater than 5%. Ultimately, the optimal number of trajectory subgroups for delirium MDAS time trajectories for delirium patients was determined to be 3 (with the smallest absolute BIC value). See [link to relevant documentation]. Figure 1 Further, the orders of the polynomial functions with the most statistical significance in each subgroup were selected as 2nd, 3rd, and 3rd orders, with corresponding AvePP values ​​of 0.86, 0.86, and 0.85, respectively. Based on the trajectory morphology and clinical significance, the subgroups were classified to intuitively reflect the different dynamic trends in delirium severity. The mild-exacerbation trajectory group refers to delirium that started mildly but gradually worsened over time; the severe-exacerbation trajectory group has more severe initial delirium symptoms, and the severity of delirium progressively worsens over time, with an overall severity higher than the mild-exacerbation trajectory group; the severe-remission trajectory group has more severe initial delirium symptoms, but the delirium symptoms gradually subside over time.

[0025] (3) Establish a dynamic prediction model for the prognosis of advanced cancer patients with delirium in palliative care wards. Based on previous prognostic studies of delirium patients in palliative care, and combined with clinical practice, information on variables that may affect the prognosis of delirium patients was collected. The delirium MDAS time trajectory subgroup was also included in the analysis, and the survival time of delirium patients was followed up. Using a Cox proportional hazards regression model, independent risk factors affecting the prognosis of advanced cancer patients with delirium in palliative care wards were established as follows: delirium MDAS time trajectory, age, Charlson comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score. Using patients with increased activity delirium and severe remission trajectory groups as references, the dynamic prognostic prediction model constructed based on the results of the Cox proportional hazards regression model is as follows: in, This is the baseline risk function, which represents the change of risk over time without the influence of covariates. =ln2 / median survival time. Median survival time is derived from data in the sample set of patients diagnosed with delirium in this embodiment. Median survival time is the median survival time of delirium patients in the severe remission trajectory group who exhibit increased activity. Increased activity delirium, decreased activity delirium, and mixed delirium are classifications of delirium motor subtypes. Severe remission trajectory delirium, mild exacerbation trajectory delirium, and severe exacerbation trajectory delirium are subgroups of delirium MDAS time trajectories. In the prognostic prediction model, parameters such as decreased activity delirium, mixed delirium, age, comorbidity index, mild exacerbation trajectory delirium, severe exacerbation trajectory delirium, and body mass index change over time. change.

[0026] 2. Data collection and calculation As a result, patients with delirium were stratified by mortality risk. The following parameters were collected from delirium patients receiving palliative care: delirium MDAS time trajectory subgroup, age, comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score. These parameters were then substituted into the above... The calculation formula calculates the mortality risk of delirium patients at time t, i.e., the instantaneous mortality risk of the patients. Based on the results, delirium patients are stratified according to their mortality risk. ≤5% is considered low-risk, 5% < ≤10% is considered the medium-risk group. >10% are classified as high-risk. Delirium is managed in a stratified manner based on the results of mortality risk stratification, and mortality risk is dynamically assessed based on changes in delirium trajectory after intervention, with timely adjustments to the delirium management plan.

[0027] In summary, this invention incorporates the delirium MDAS time trajectory into a prognostic prediction model, providing a novel prognostic prediction method for clinical practice. This dynamic classification of delirium can distinguish patients with similar baseline severity but different progression patterns. The prognostic prediction model built based on the delirium MDAS time trajectory is a dynamic predictive model that facilitates more accurate prognostic stratification, provides personalized and dynamic delirium management, and helps improve the quality of palliative care and the quality of life of delirium patients in palliative care wards, demonstrating promising application prospects.

Claims

1. A prognostic prediction method based on the temporal trajectory of delirium MDAS, characterized in that, Includes the following steps: Step 1: Obtain parameters such as delirium MDAS time trajectory subgroup, age, comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score of palliative care patients; Step 2: Use the parameters obtained in Step 1 to calculate the instantaneous mortality risk of delirious patients receiving palliative care at time t. The calculation formula is as follows: in, This is the baseline risk function, which represents the change in mortality risk over time without the influence of covariates. The calculation formula is: =ln2 / median survival time, where median survival time is the median survival time of delirium patients in the severe remission trajectory group with increased activity; increased activity delirium, decreased activity delirium, and mixed delirium are classifications of motor subtypes of delirium; delirium in the severe remission trajectory group, delirium in the mild exacerbation trajectory group, and delirium in the severe exacerbation trajectory group are subgroups of delirium MDAS time trajectory; Step 3: Based on the calculation results of Step 2, stratify the mortality risk of delirium patients.

2. The method according to claim 1, characterized in that, The delirium MDAS time trajectory subgroup was determined through the following steps: The total MDAS score of delirium patients was assessed twice a week. The trajectory of the total MDAS score over time was fitted using the group-based trajectory model statistical method. Based on the fitted trajectory of the total MDAS score over time, the delirium MDAS time trajectory subgroup of the patients was determined.

3. The method according to claim 2, characterized in that, The delirium MDAS time trajectory subgroups were determined according to the following rules: a trajectory with an initial MDAS score below 15 and a gradually increasing MDAS score over time indicates mild aggravated delirium; a trajectory with an initial MDAS score greater than or equal to 15 and a progressively increasing MDAS score over time indicates severely aggravated delirium; and a trajectory with an initial MDAS score greater than or equal to 15 and a gradually decreasing MDAS score over time indicates severely remitted delirium.

4. The method according to claim 1, characterized in that, The mortality risk stratification is performed as follows: calculate The results were used to stratify mortality risk. ≤5% is considered low-risk, 5% < ≤10% is considered the medium-risk group. >10% is considered high-risk.

5. A prognostic prediction system based on the temporal trajectory of delirium MDAS, characterized in that, To implement the method according to any one of claims 1-4, comprising: The input module is configured to acquire parameters such as delirium MDAS time trajectory subgroup, age, comorbidity index, delirium motor subtype, body mass index, and initial MDAS total score of delirium patients receiving palliative care; The module is configured to use the parameters obtained from the input module to calculate the mortality risk of delirious patients in palliative care at time t, using the following formula: in, This is the baseline risk function, which represents the change in mortality risk over time without the influence of covariates. The calculation formula is: =ln2 / median survival time, where median survival time is the median survival time of delirium patients in the severe remission trajectory group with increased activity; increased activity delirium, decreased activity delirium, and mixed delirium are classifications of motor subtypes of delirium; delirium in the severe remission trajectory group, delirium in the mild exacerbation trajectory group, and delirium in the severe exacerbation trajectory group are subgroups of delirium MDAS time trajectory; The stratification module is configured to stratify delirium patients based on the calculation results of the construction module to determine their mortality risk.

6. The system according to claim 5, characterized in that, The delirium MDAS time trajectory subgroup was determined through the following steps: The total MDAS score of delirium patients was assessed twice a week. The trajectory of the total MDAS score over time was fitted using the group-based trajectory model statistical method. Based on the fitted trajectory of the total MDAS score over time, the delirium MDAS time trajectory subgroup of the patients was determined.

7. The system according to claim 6, characterized in that, The delirium MDAS time trajectory subgroups were determined according to the following rules: a trajectory with an initial MDAS score below 15 and a gradually increasing MDAS score over time indicates mild aggravated delirium; a trajectory with an initial MDAS score greater than or equal to 15 and a progressively increasing MDAS score over time indicates severely aggravated delirium; and a trajectory with an initial MDAS score greater than or equal to 15 and a gradually decreasing MDAS score over time indicates severely remitted delirium.

8. The system according to claim 5, characterized in that, The mortality risk stratification is performed as follows: calculate The results were used to stratify mortality risk. ≤5% is considered low-risk, 5% < ≤10% is considered the medium-risk group. >10% is considered high-risk.

9. A computer-readable storage medium, characterized in that, It stores a computer program for implementing the prognostic prediction system based on delirium MDAS time trajectory as described in any one of claims 5-8.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-4.

Images