The commercial health insurance market is growing rapidly, but at the same time, insurance companies are also facing higher risk control and competitive pressures. Insurance companies urgently need to implement active risk control fee control and personalized health services through health management and managed care. How to build a closed loop of “healthcare + insurance payment” and how “health insurance + medical treatment” drives the linkage of medical insurance has aroused industry attention.
In mid-April, at the “Fifth Future Medical Top 100” conference hosted by Artery.com, Knowledge-VisionCOO Dr. Long Jiao was invited to attend the “Medical Insurance Technology and Commercial Health Insurance Forum” and used “Health Insurance Data Assets and He gave a speech on the theme of “Big Data Driven Medical Insurance Linkage”, introducing in detail Knowledge-Vision’s innovation and exploration in insurance and medical big data.
Health management based on medical big data drives health insurance innovation
Under the background of accelerating aging, advancement in diagnosis and treatment technology, and upgrading of consumption, commercial health insurance has entered a period of rapid development. It is expected to reach a market size of 2 trillion in 2025. At the same time, health insurance products are becoming more homogeneous and competitive. Increasingly fierce. The maximum coverage of health insurance products, insurance liability, and the richness of covered diseases are all approaching their limits, but the room for further decline in the rate is extremely small. In the future, innovative health insurance and supporting health management services based on medical big data, targeted at sick and sub-healthy populations will become the new growth engine of the health insurance market.
Dr. Jiao Long pointed out that health management services have great commercial and social value. For insurance companies, health management can enhance the differentiated competitiveness of products, improve user satisfaction, and effectively improve the premium capabilities of products; through health management, pre-management of risks can be achieved, which can improve the insurance company’s risk identification capabilities, reduce the loss rate, and further Realize operational value-added. Through health management, the efficiency of the use of social medical resources has been improved, thereby improving the health of the whole society and realizing the enhancement of social value.
Health management is mainly divided into two parts: 1) is the management of the healthy life of the healthy body, 2) is the management of the disease of the sick group. There are three key points to form a double closed loop of healthy living and disease management: First, customer segmentation. Only after population segmentation can a broader coverage of sub-healthy, chronically ill, and critically ill people be formed; second, data Analytical ability is the core of the entire health management system, especially the analysis and utilization of effective clinical data is the basis for personalized health management services; the third is in-depth professional cooperation, with basic medical insurance, commercial insurance for serious illnesses To determine the trend, commercial health insurance companies need to conduct in-depth cooperation with pharmaceutical and device companies in key areas, especially oncology, to jointly build results-oriented health management. Data is the support and bridge of the entire health management system, but due to the obstruction of data circulation between medical institutions, an island of medical data is formed. Out-of-hospital data is generally heterogeneous and unstructured data, and it is difficult to extract, process and analyze data. .
Deeply cultivate medical big data to help realize the digitization of medical information
Insurance companies collect a wealth of medical data in the process of underwriting, claim settlement and health management. There are more than 40 kinds of common medical documents, such as prescriptions, expense lists, diagnostic reports, inspection reports, pathology reports, discharge summary, etc., but there are three types of data. Big problems: First, it is difficult to extract and use valuable medical information at low cost because of massive unstructured data; second, poor standardization. These data come from different medical institutions across the country, and there are problems with inconsistencies in templates and terminology. Standardized processing is required at the time; the third is strong professionalism, and the processing of medical data requires a medical background, which is difficult and costly to process.
There are two traditional processing methods for medical data: manual processing and general OCR. Among them, manual processing costs are high, entry takes a long time, accuracy is low, and reliability is poor. General OCR can only solve the problem of textualization, and cannot complete the structure of the data.化. Furthermore, because the medical data obtained by Baosi is not perfect data, there are often problems such as wrong lines and overlapping characters. General OCR cannot solve these problems.Knowledge-Vision aims at the pain points of the industry and independently developed four intelligent systems to help complete medical data processing.
Inphile intelligent medical data text processing system (digitalization of medical document images): Based on real clinical medical scenarios, the unique OCR algorithm can achieve accounting, diagnosis and medical information extraction, and has an exclusive single word error prompt function, with a low error prompt rate Less than 10%, can greatly reduce the workload of manual verification, and can effectively solve the problems of wrong lines, overlapping characters, etc., through the way of man-machine collaboration, the entry time can be shortened by 70%, and the overall accuracy rate is as high as 99%. Inphile intelligent medical data structured processing system (structured extraction of medical knowledge): Supports structured output of common claim application materials such as ID cards, bank cards, expense lists, medical records, image reports, inspection reports, pathology reports, and structured medicine Data can enrich big health data and provide a strong guarantee for smart risk control, cost control, and health management. Inphile intelligent medical data standardization processing system (data cleaning, normalization): It can realize the normalization of standard nouns, aliases, and aliases for medical institutions, disease diagnosis, drugs, medical services, diagnosis and treatment, and consumables, with an automatic encoding rate of up to 98% ; Establishment of patient electronic health records (knowledge): Generate personalized labels for user portraits through user basic information, medication records, inspection reports, claims records and other data, and display user standardized and integrated health file information in the form of a timeline , And based on the user’s follow-up behavior to update and enrich the health file in real time.
Multi-source heterogeneous medical data, through Inphile textualization, structuring, normalization and knowledge processing, forms operable insurance and medical big data, thus completing the assetization of health insurance data, and standardization of medical data. The foundation of medical data fusion.
As an asset, insurance and medical big data can achieve operational value-added through effective operations.
There is a huge information asymmetry between the demanders and providers of medical services, and the mismatch of medical resources is common. Through insurance and medical big data, accurate matching of medical insurance based on accurate portraits of users can be achieved. For example, the medical examination report, inspection report, test report, etc. provided by the customer in the process of insurance, claim settlement, and health management, complete the digital processing of medical information through the Inphile intelligent platform. Based on medical information, health risk assessment can be conducted, and relevant testing recommendations and health management recommendations can be further provided. At the same time, patient-based labels can recommend corresponding insurance products based on customer needs, promote innovation in insurance marketing, and make the marketing and delivery of insurance products more accurate.
For insured persons who are already ill, such as cancer patients, based on their pathology report, examination report, treatment status, etc., locate the tumor segmentation group, and provide patients with a comparative analysis report of the diagnosis and treatment effects of the same type of patients, that is, the clinical performance of different treatment plans As a reference, this innovation has also promoted the transition from experience-driven medical care to data-driven medical care.
Insurance and medical big data, as a kind of real-world data, can empower the entire life cycle of drug research and development, and is of great value to new drug research and development. Insurance medical big data can provide data support for the development of new drug indications, fast-track new drug listing, clinical trial optimization design, and subject recruitment. At the same time, digital marketing can be carried out based on insurance medical big data in the new drug listing stage, and continue Provide follow-up services for pharmaceutical companies.
In the digital operation management model, the insurance company has realized the whole life cycle health management service for the insured through health service, disease management, treatment management and rehabilitation management, and completed the process management of the medical service. Through health management, Baosi is no longer just a payer for medical services, but also a coordinator of medical resources. Driven by insurance and medical big data, the insurance company can realize the health management of thousands of insured persons, optimize the configuration of medical services, and create differentiated health management capabilities.
Knowledge-Vision is committed to the integration of medical big data centered on health insurance, and uses a new generation of artificial intelligence technology to build a big data-driven medical-pharmaceutical-insurance linkage innovative service platform to provide users with diagnosis and treatment decision-making suggestions based on medical big data. Through the textualization, structuring, normalization and knowledgeization of insurance and medical big data, it helps insurance companies to realize the assetization of health insurance data, and at the same time empowers insurance companies to build data as a service (DaaS) capabilities, thereby realizing data Asset operation management and personalized health management services, and in this process, through the integration of the pharmaceutical supply chain, help insurance companies to achieve the three-medicine linkage of pharmaceutical insurance.
In the future, the application of insurance and medical big data will gradually open up a huge market for insurance operations. Knowledge-Vision will continue to work hard in this field, relying on the three major business sectors of “products, services, and operations” to build a trustworthy health management service platform.