大数据背景下的医疗保险费用挖掘

发布时间：2018-09-06 14:19

【摘要】：随着互联网技术的发展,大数据分析逐渐渗透到各行各业,包括医疗行业。在医疗数字化的过程中,产生了EMR(电子病历)、HIS(医院信息系统)、LIS(医院检验系统)、MI(医学影像)等资料。英特尔预测:到2020年,医疗数据将增至35ZB,是2009年数据量的44倍。医疗行业正在面临着海量数据和非结构化数据的重大挑战和机遇。大数据正在逐渐改变着公共卫生医疗领域。本文在国内外相关研究综述和理论分析的基础上,从医疗保险的角度出发,针对获得的部分住院病历首页数据和整体医疗费用数据进行数据挖掘分析,为相关决策的合理制定提供科学的依据。本文的主要研究工作如下:1.四种高发病的疾病特征研究及费用分析。通过对获取的住院病历首页数据的挖掘分析,总结出四种疾病的基本特征,得到四种疾病DRGs的医疗保险费用偿付标准。通过描述性分析、回归分析等基本统计方法获得四种高发疾病的基本特征;进一步通过决策树分析方法挖掘住院医疗费用内在规律,获得四种疾病DRGs的医疗保险费用偿付标准。基于DRGs的高发病特征及费用分析可以为医疗保险机构进行医疗保险相关费用偿付政策的制定提供数据支持,具有很大应用价值。2.重大疾病的医疗保险费用挖掘分析。我国卫生总费用、城乡居民人均医疗保健支出、住院病人人均医药费用以及基本医疗保险基金支出均呈现快速增长的趋势。通过对西安市居民整体医疗费用数据的挖掘分析,得出居民住院医疗费用和自付费用的分布状况,并在此基础上依据经验频数法定价模型进行高额费用保险保费的测算。本文的研究成果对费用型商业医疗保险保费研究具有很大的应用价值,同时也对我国医疗保险市场的丰富和完善有一定的贡献。日益增长的医疗大数据是医疗行业的一项重大资源优势。医疗保险机构必须注重收集医疗数字化过程中产生的医疗数据资料,通过科学系统地分析与总结,为今后相关政策的制定提供依据。医疗机构应该重点把握这一优势,进行医疗大数据的统计描述、数据挖掘、建模预测,以提供基于海量数据分析的相关决策支持。
[Abstract]:With the development of Internet technology, big data analysis gradually infiltrated into various industries, including the medical industry. In the process of medical digitization, EMR (Electronic Medical record) and his (Hospital Information system) and MI (Medical Imaging) are produced. Intel predicts that medical data will rise to 35ZB by 2020, 44 times the amount of data in 2009. The medical industry is facing great challenges and opportunities of massive data and unstructured data. Big data is gradually changing the field of public health care. Based on the summary and theoretical analysis of relevant research at home and abroad, this paper analyzes the data mining of the first page data and the overall medical cost data from the point of view of medical insurance. To provide scientific basis for the rational formulation of relevant decisions. The main research work of this paper is as follows: 1. Study on the disease characteristics and cost analysis of four kinds of high-incidence diseases. By mining and analyzing the first page data of the inpatient medical records, the basic characteristics of the four diseases are summarized, and the reimbursement standard of the medical insurance costs for the four diseases DRGs is obtained. Through descriptive analysis, regression analysis and other basic statistical methods, the basic characteristics of four kinds of high incidence diseases are obtained, and the internal rules of hospitalization medical expenses are further excavated by decision tree analysis, and the reimbursement standard of medical insurance costs for four diseases DRGs is obtained. The characteristics and cost analysis of high incidence disease based on DRGs can provide data support for the establishment of medical insurance related expense reimbursement policy in medical insurance institutions, which has great application value. Analysis of medical insurance expenses for major diseases. The total expenditure of health care, the per capita medical care expenditure of urban and rural residents, the per capita medical expenditure of inpatients and the expenditure of basic medical insurance fund are all increasing rapidly in China. By mining and analyzing the data of residents' overall medical expenses in Xi'an, the distribution of residents' hospitalization medical expenses and out-of-pocket medical expenses is obtained, and on this basis, the high cost insurance premium is calculated according to the empirical frequency method pricing model. The research results of this paper have great application value to the research of the premium of the commercial medical insurance of the cost type, and also have certain contribution to the enrichment and perfection of the medical insurance market of our country at the same time. The growing medical treatment big data is a major resource advantage of the medical industry. Medical insurance institutions must pay attention to collecting the medical data generated in the process of medical digitization and provide the basis for the formulation of related policies through scientific and systematic analysis and summary. Medical institutions should focus on this advantage, carry on the medical big data's statistical description, data mining, modeling and forecasting, in order to provide the relevant decision support based on massive data analysis.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F842.684;TP311.13

【相似文献】