Method and system for supporting narrative of caregiver
By using semi-structured electronic interviews and NLP technology, combined with multi-dimensional assessment tools, a multi-role collaborative support network was built, which solved the problem of family caregivers lacking professional knowledge, achieved standardized digital narrative support, and improved the quality and efficiency of care.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE FIFTH PEOPLES HOSPITAL OF SHANGHAI
- Filing Date
- 2025-12-01
- Publication Date
- 2026-04-28
AI Technical Summary
In existing technologies, family caregivers lack professional knowledge and skills, resulting in a high incidence of complications such as pressure sores in disabled elderly people. Traditional narrative interventions are costly, have low standardization and are difficult to scale up, and lack digital and intelligent support systems.
By employing semi-structured electronic interviews, natural language processing (NLP) technology, and multi-dimensional assessment tools, combined with intelligent analysis and feedback generation, a multi-role collaborative support network is constructed to provide personalized emotional support and skills guidance.
It has achieved a standardized digital narrative support process, improved intervention efficiency, provided an instant emotional catharsis and skills learning platform, and enhanced care quality and adherence.
Smart Images
Figure CN121938567A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of digital healthcare and health information technology, and specifically relates to a caregiver narrative support method and system. Background Technology
[0002] Family caregivers, as the core of the care system, endure immense physical and psychological stress, commonly experiencing anxiety, depression, and other emotional problems. Furthermore, their general lack of professional nursing knowledge and skills contributes to a persistently high incidence of complications such as pressure sores among the disabled elderly they care for. Narrative medicine, as an emerging medical humanities practice, guides caregivers to recount their own experiences and express their inner emotions, and has been proven to effectively alleviate their psychological stress and promote cognitive transformation, thereby improving the quality of care.
[0003] However, traditional narrative interventions heavily rely on in-home interviews, which suffer from inherent drawbacks such as high implementation costs, low standardization, cumbersome data collection, low efficiency, and difficulty in scaling up. Currently, the market lacks dedicated systems or platforms that deeply integrate narrative medicine theory, caregiver support, and digital technology, thus failing to provide caregivers with immediate, continuous, standardized emotional support and precise skills guidance.
[0004] While existing technologies include systems for mental health support or care management, such as questionnaire-based assessment tools or simple information delivery platforms, they often lack structured guidance based on in-depth narratives. They fail to achieve intelligent perception and dynamic intervention of caregivers' emotional states, nor do they establish a sustainable support ecosystem linking caregivers, professional teams, and the community. Therefore, there is an urgent need in this field to develop a professional system and methodology that integrates digitalization, structuring, and intelligence to address these challenges. Summary of the Invention
[0005] In view of the above-mentioned problems, the present invention provides a caregiver narrative support method and system to solve the problems of insufficient caregiver support, low intervention efficiency and difficulty in scaling up in the prior art.
[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:
[0007] The first aspect of this invention provides a caregiver narrative support method, comprising the following steps:
[0008] S10, Narrative Guidance and Data Collection: Based on a pre-set semi-structured electronic interview outline, guide caregivers to express their narratives through mobile terminals, and collect narrative data and basic information in the form of voice and text from caregivers.
[0009] S20, conduct multi-dimensional assessment: integrate and use the Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), and Pressure Ulcer Knowledge Questionnaire to conduct online assessments of caregivers at baseline and multiple follow-up times, and automatically complete scoring and data management;
[0010] S30 performs intelligent analysis and feedback generation: It uses Natural Language Processing (NLP) technology to perform real-time sentiment analysis on narrative text, automatically identifies the caregiver's stress level and emotional state, and generates personalized feedback reports.
[0011] S40, Resource Push: Based on the analysis of narrative text and assessment results, identify caregivers' knowledge gaps and emotional needs, and automatically match and push relevant skills training videos, graphic guides or professional consultation resources;
[0012] S50, Building a Collaborative Support Network: Constructing an online community and communication platform connecting caregivers, researchers, community doctors, and mental health counselors to form a multi-role collaborative support network.
[0013] In one possible implementation, S10 specifically includes: compiling a standardized electronic interview outline based on the Kallio semi-structured interview method, and guiding the caregiver to conduct an orderly narrative about their caregiving experience, emotional experiences, sources of stress, and coping mechanisms through a speech recognition and text input module.
[0014] In one possible implementation, S30 specifically includes: extracting keywords, performing sentiment polarity analysis, and modeling themes from the narrative text to quantify the caregiver's cognitive stress level and emotional change trends; the personalized feedback report includes an emotional state summary, stressor analysis, and positive suggestions based on cognitive restructuring theory.
[0015] In one possible implementation, the method further includes S60, which involves evaluating the intervention effect: within a preset intervention period, the multidimensional assessment is repeated at time points including the baseline, the 3rd month, and the 6th month, and the changes in the caregiver's psychological status score and the incidence of pressure ulcers in the disabled elderly are compared and analyzed to quantitatively evaluate the intervention effect.
[0016] A second aspect of this invention provides a caregiver narrative support system, comprising a mobile terminal and a cloud service platform. The mobile terminal includes a narrative guidance and data collection module, a multi-dimensional assessment module, and a collaborative support network construction module. The cloud service platform includes an intelligent analysis and feedback generation module and a resource push module.
[0017] The narrative guidance and data collection module is used to guide caregivers to express their narratives via mobile terminals based on a pre-set semi-structured electronic interview outline, and to collect narrative data and basic information in the form of voice and text from caregivers.
[0018] The multi-dimensional assessment module is used to integrate and use the Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), and Pressure Ulcer Knowledge Questionnaire to conduct online assessments of caregivers at baseline and multiple follow-up times, and to automatically complete scoring and data management.
[0019] The intelligent analysis and feedback generation module uses natural language processing (NLP) technology to perform real-time sentiment analysis on narrative texts, automatically identify caregivers' stress levels and emotional states, and generate personalized feedback reports.
[0020] The resource push module is used to identify caregivers' knowledge gaps and emotional needs based on the analysis of narrative text and assessment results, and automatically match and push relevant skills training videos, graphic guides or professional consultation resources.
[0021] The Collaborative Support Network Building Module is used to build an online community and communication platform that connects caregivers, researchers, community doctors, and mental health counselors, forming a multi-role collaborative support network.
[0022] In one possible implementation, the narrative guidance and data collection module specifically works by: compiling a standardized electronic interview outline based on the Kallio semi-structured interview method, and guiding caregivers to conduct an orderly narrative about their caregiving experiences, emotional experiences, sources of stress, and coping mechanisms through a speech recognition and text input module.
[0023] In one possible implementation, the intelligent analysis and feedback generation module specifically works by: extracting keywords, performing sentiment polarity analysis, and modeling themes from the narrative text to quantify the caregiver's cognitive stress level and emotional change trends; the personalized feedback report includes an emotional state summary, stressor analysis, and positive suggestions based on cognitive restructuring theory.
[0024] In one possible implementation, the intervention effect evaluation module is further included, which is used to repeat the multidimensional evaluation at time points including baseline, 3 months and 6 months within a preset intervention period, and compare and analyze the changes in caregiver psychological status scores and the incidence of pressure ulcers in disabled elderly people to quantitatively evaluate the intervention effect.
[0025] The present invention has the following beneficial effects:
[0026] (1) By deeply integrating the theoretical framework of narrative medicine with digital systems, the traditional offline and oral intervention model is transformed into a standardized online process that can be recorded, analyzed, and replicated, thus solving the core pain point that manual intervention is difficult to scale up.
[0027] (2) By introducing NLP sentiment analysis technology, the passive recording is transformed into active perception, which can automatically identify the caregiver's emotional crisis and knowledge needs, and trigger personalized resource push and early warning mechanisms, thus realizing the leap from "generalized support" to "precise intervention".
[0028] (3) A sustainable support ecosystem involving caregivers, families, communities, and professional teams has been built. Through the digital platform, the spatial and temporal boundaries of professional support have been greatly extended, providing caregivers with an emotional outlet and skills learning platform 24 / 7, which has effectively improved intervention efficiency, caregiver compliance, and sense of gain. Attached Figure Description
[0029] Figure 1 This is a flowchart illustrating the steps of a caregiver narrative support method according to an embodiment of the present invention.
[0030] Figure 2 This is a schematic diagram of a caregiver narrative support system according to another embodiment of the present invention. Detailed Implementation
[0031] 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, not all, of the embodiments of the present invention. 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.
[0032] Reference Figure 1 The diagram shows a flowchart of a caregiver narrative support method according to an embodiment of the present invention, including the following steps:
[0033] S10, Narrative Guidance and Data Collection: Based on a pre-set semi-structured electronic interview outline, guide caregivers to express their narratives through mobile terminals, and collect narrative data and basic information in the form of voice and text from caregivers.
[0034] S20, conduct multi-dimensional assessment: integrate and use the Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), and Pressure Ulcer Knowledge Questionnaire to conduct online assessments of caregivers at baseline and multiple follow-up times, and automatically complete scoring and data management;
[0035] S30 performs intelligent analysis and feedback generation: It uses Natural Language Processing (NLP) technology to perform real-time sentiment analysis on narrative text, automatically identifies the caregiver's stress level and emotional state, and generates personalized feedback reports.
[0036] S40, Resource Push: Based on the analysis of narrative text and assessment results, identify caregivers' knowledge gaps and emotional needs, and automatically match and push relevant skills training videos, graphic guides or professional consultation resources;
[0037] S50, Building a Collaborative Support Network: Constructing an online community and communication platform connecting caregivers, researchers, community doctors, and mental health counselors to form a multi-role collaborative support network.
[0038] In a specific application example, S10 includes: compiling a standardized electronic interview outline based on the Kallio semi-structured interview method, and guiding caregivers to conduct an orderly narrative about their caregiving experiences, emotional experiences, sources of stress, and coping mechanisms through a voice recognition and text input module.
[0039] In a specific application example, S30 specifically includes: extracting keywords, analyzing emotional polarity, and modeling themes from the narrative text to quantify the caregiver's level of stress cognition and emotional change trends; the personalized feedback report includes an emotional state summary, stressor analysis, and positive suggestions based on cognitive restructuring theory.
[0040] Based on S10 to S50, a caregiver narrative support method further includes S60, which evaluates the intervention effect: within a pre-defined intervention period, the multi-dimensional assessment is repeated at time points including baseline, 3 months and 6 months, and the changes in caregiver psychological status scores and the incidence of pressure ulcers in disabled elderly are compared and analyzed to quantitatively evaluate the intervention effect.
[0041] The caregiver narrative support method described above begins with caregiver registration and proceeds sequentially with baseline assessment, periodic narrative guidance, and data collection. The collected data triggers intelligent analysis and immediate feedback, leading to precise resource allocation based on the analysis results. Throughout the process, the system continuously monitors the data and initiates collaborative support intervention when necessary. Periodic follow-up assessments are conducted at pre-defined follow-up points, and all data is ultimately used for effectiveness evaluation and program optimization, forming a continuously improving closed-loop intervention model.
[0042] Through the steps outlined above, a specific application example of a caregiver narrative support method includes the following steps: Caregivers register via a mobile app, and the system guides them to complete online assessments of baseline information (age, care duration, etc.), SDS, SAS scales, and a pressure ulcer knowledge questionnaire. The system proactively pushes narrative guidance questions based on the Kallio interview method weekly or bi-weekly (configurable according to needs), such as "Please describe the most stressful thing that happened to you during this week's caregiving process." Caregivers respond via voice or text. The system performs real-time NLP analysis on the submitted narrative text, generating brief emotional state feedback and encouraging words, which are immediately displayed on the app. Simultaneously, a deep analysis process is triggered to generate detailed reports and resource recommendation decisions. The system combines narrative analysis and the latest assessment data to determine the caregiver's current primary needs. If insufficient pressure ulcer knowledge and a high anxiety score are identified, a series of videos on "Early Identification and Prevention of Pressure Ulcers" and audio on "Relaxation Training" are automatically pushed. At preset follow-up time points (e.g., month 3, month 6), the system automatically reminds caregivers to complete the SDS, SAS scales, and pressure ulcer knowledge questionnaire again, and records the skin condition of the disabled elderly. When the system identifies emergency situations such as caregivers showing signs of severe depression (e.g., SDS score exceeding the threshold) or expressing suicidal ideation in their narratives, it automatically generates an alert and notifies designated researchers or psychologists for manual intervention. After the study, the system exports all structured data (scale scores, pressure ulcer incidence, narrative text topic distribution, etc.) for statistical analysis of intervention effects and to provide data support for optimizing the intervention program.
[0043] Corresponding to the method embodiment, another embodiment of the present invention provides a caregiver narrative support system, including a mobile terminal 10 and a cloud service platform 20. The mobile terminal is equipped with a narrative guidance and data collection module 101, a multi-dimensional assessment module 102, and a collaborative support network construction module 103. The cloud service is equipped with an intelligent analysis and feedback generation module 201 and a resource push module 202. The narrative guidance and data collection module 101 guides caregivers to express their narratives based on a preset semi-structured electronic interview outline, and collects narrative data and basic information in voice and text formats from caregivers via the mobile terminal. The multi-dimensional assessment module 102 integrates and uses the Self-Rating Depression Scale (SDS), the Self-Rating Anxiety Scale (SAS), and other data. The pressure ulcer knowledge questionnaire provides online assessments of caregivers at baseline and during multiple follow-ups, automatically completing scoring and data management. The intelligent analysis and feedback generation module 201 uses Natural Language Processing (NLP) technology to perform real-time sentiment analysis on narrative texts, automatically identifying caregivers' stress levels and emotional states, and generating personalized feedback reports. The resource push module 202, based on the analysis of narrative texts and assessment results, identifies caregivers' knowledge gaps and emotional needs, automatically matching and pushing relevant skills training videos, illustrated guides, or professional consultation resources. The collaborative support network construction module 103 builds an online community and communication platform connecting caregivers, researchers, community doctors, and psychologists, forming a multi-role collaborative support network.
[0044] In a specific application example, the narrative guidance and data collection module 101 works as follows: Based on the Kallio semi-structured interview method, a standardized electronic interview outline is prepared, and through the speech recognition and text input module, caregivers are guided to conduct an orderly narrative about their caregiving experience, emotional experience, sources of stress, and coping mechanisms.
[0045] In a specific application example, the intelligent analysis and feedback generation module 201 works by extracting keywords, performing sentiment polarity analysis, and modeling themes from the narrative text to quantify the caregiver's stress cognition level and emotional change trends. The personalized feedback report includes an emotional state summary, stressor analysis, and positive suggestions based on cognitive restructuring theory.
[0046] Another embodiment of the caregiver narrative support system of the present invention further includes an intervention effect evaluation module 203, which is set on a cloud service platform 20. It is used to repeat the multi-dimensional evaluation at time points including baseline, the third month and the sixth month within a preset intervention period, and compare and analyze the changes in caregiver psychological status scores and the incidence of pressure ulcers in disabled elderly people to quantitatively evaluate the intervention effect.
[0047] In a specific application example, the narrative guidance and data collection module 101, based on the Kallio semi-structured interview method, incorporates a standardized electronic interview outline. The outline covers the caregiver's daily care activities, emotional experiences, stress perception, social support, and specific skills challenges (such as pressure ulcer prevention). Caregivers can narrate their stories via the app's voice input function (converted to text via speech recognition) or direct text input. This process ensures the structured and standardized nature of data collection. The multi-dimensional assessment module 202 integrates standardized scales with good reliability and validity, including the Self-Rating Depression Scale (SDS), the Self-Rating Anxiety Scale (SAS), and a knowledge questionnaire on pressure ulcer prevention and care. The system automatically pushes these scales to caregivers upon enrollment (baseline), mid-intervention (e.g., month 3), and end of intervention (e.g., month 6), and automatically scores, stores, and visualizes trends. The intelligent analysis and feedback module 201 uses a pre-trained natural language processing (NLP) model (e.g., a sentiment analysis model based on BERT or a similar architecture) to analyze the narrative text submitted by caregivers. The analysis includes: sentiment analysis (determining the text's sentiment tendency as positive, negative, or neutral, and its intensity), and keyword and topic extraction (identifying high-frequency words such as "fatigue," "helplessness," "red skin," and "difficulty turning over," as well as related topics). Based on the analysis results, a personalized feedback report is automatically generated, which may include: "According to your description, you may have recently experienced a high level of stress and anxiety, mainly related to frequent nighttime awakenings and concerns about pressure sores. We suggest you pay attention to the 'stress reduction techniques' videos and 'pressure sore prevention and postural management' guidelines." The resource push module 202 has a built-in resource library containing videos, images, articles, and other formats, covering pressure sore prevention, psychological adjustment, and care techniques. When the intelligent analysis module identifies a caregiver's specific knowledge gaps (such as repeatedly mentioning but not understanding "pressure sore air mattress") or emotional needs (such as exhibiting significant depressive mood), this module will retrieve the most relevant content from the resource library according to preset rules (keyword matching, sentiment tag matching) and proactively push it to the caregiver through the APP message center. For the collaborative support network module 203, a secure community forum and peer-to-peer communication functions are provided within the APP. Caregivers can anonymously share experiences and seek peer support in the forum; researchers or community nurses can publish popular science knowledge and organize online activities; when caregivers encounter emergency or complex problems, they can directly seek help from designated family doctors or psychological counselors through the communication function, forming a multi-role collaborative support network.
[0048] It should be understood that the exemplary embodiments described herein are illustrative and not restrictive. Although one or more embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art will understand that various changes in form and detail may be made without departing from the spirit and scope of the invention as defined by the appended claims.
Claims
1. A caregiver narrative support method, characterized in that, Includes the following steps: S10, Narrative Guidance and Data Collection: Based on a pre-set semi-structured electronic interview outline, guide caregivers to express their narratives through mobile terminals, and collect narrative data and basic information in the form of voice and text from caregivers. S20, conduct multi-dimensional assessment: integrate and use the Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), and Pressure Ulcer Knowledge Questionnaire to conduct online assessments of caregivers at baseline and multiple follow-up times, and automatically complete scoring and data management; S30 performs intelligent analysis and feedback generation: It uses Natural Language Processing (NLP) technology to perform real-time sentiment analysis on narrative text, automatically identifies the caregiver's stress level and emotional state, and generates personalized feedback reports. S40, Resource Push: Based on the analysis of narrative text and assessment results, identify caregivers' knowledge gaps and emotional needs, and automatically match and push relevant skills training videos, graphic guides or professional consultation resources; S50, Building a Collaborative Support Network: Constructing an online community and communication platform connecting caregivers, researchers, community doctors, and mental health counselors to form a multi-role collaborative support network.
2. The caregiver narrative support method as described in claim 1, characterized in that, S10 specifically includes: compiling a standardized electronic interview outline based on the Kallio semi-structured interview method, and guiding caregivers to conduct an orderly narrative about their caregiving experiences, emotional experiences, sources of stress, and coping mechanisms through a speech recognition and text input module.
3. The caregiver narrative support method as described in claim 1, characterized in that, S30 specifically includes: extracting keywords, analyzing emotional polarity, and modeling themes from the narrative text to quantify the caregiver's cognitive stress level and emotional change trends; the personalized feedback report includes an emotional state summary, stressor analysis, and positive suggestions based on cognitive restructuring theory.
4. The caregiver narrative support method as described in any one of claims 1 to 3, characterized in that, Further, S60 is used to evaluate the intervention effect: within the preset intervention period, the multidimensional assessment is repeated at time points including baseline, 3 months and 6 months, and the changes in caregiver psychological status scores and the incidence of pressure ulcers in disabled elderly are compared and analyzed to quantitatively evaluate the intervention effect.
5. A caregiver narrative support system, characterized in that, This includes mobile terminals and a cloud service platform. The mobile terminals are equipped with narrative guidance and data collection modules, multi-dimensional evaluation modules, and collaborative support network construction modules. The cloud service platform is equipped with intelligent analysis and feedback generation modules and resource push modules. The narrative guidance and data collection module is used to guide caregivers to express their narratives through mobile terminals based on a pre-set semi-structured electronic interview outline, and to collect narrative data and basic information in the form of voice and text from caregivers. The multi-dimensional assessment module is used to integrate and use the Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), and Pressure Ulcer Knowledge Questionnaire to conduct online assessments of caregivers at baseline and multiple follow-up times, and to automatically complete scoring and data management. The intelligent analysis and feedback generation module uses natural language processing (NLP) technology to perform real-time sentiment analysis on narrative texts, automatically identify caregivers' stress levels and emotional states, and generate personalized feedback reports. The resource push module is used to identify caregivers' knowledge gaps and emotional needs based on the analysis of narrative text and assessment results, and automatically match and push relevant skills training videos, graphic guides or professional consultation resources. The Collaborative Support Network Building Module is used to build an online community and communication platform that connects caregivers, researchers, community doctors, and mental health counselors, forming a multi-role collaborative support network.
6. The caregiver narrative support system as described in claim 5, characterized in that, The specific working process of the narrative guidance and data collection module includes: compiling a standardized electronic interview outline based on the Kallio semi-structured interview method, and guiding caregivers to conduct an orderly narrative about their caregiving experience, emotional experiences, sources of stress, and coping mechanisms through a speech recognition and text input module.
7. The caregiver narrative support system as described in claim 5, characterized in that, The intelligent analysis and feedback generation module specifically works by: extracting keywords, analyzing emotional polarity, and modeling themes from the narrative text to quantify the caregiver's cognitive stress level and emotional change trends; the personalized feedback report includes an emotional state summary, stressor analysis, and positive suggestions based on cognitive restructuring theory.
8. The caregiver narrative support system as described in any one of claims 5 to 7, characterized in that, It further includes an intervention effect evaluation module, which is used to repeat the multidimensional evaluation at time points including baseline, 3 months and 6 months within a preset intervention period, and compare and analyze the changes in caregiver psychological status scores and the incidence of pressure ulcers in disabled elderly people to quantitatively evaluate the intervention effect.