健保制度下病患就診模式的建構及探討

Modeling Patient's Behavior of Taking Medical Treatment under National Health Insurance System

張羢琦(Jung-Chi Chang)

指導教授 ： 蔡榮發

國立臺北科技大學/管理學院/商業自動化與管理研究所/碩士(2005年)

https://doi.org/10.6841/NTUT.2005.00261

摘要

我國自民國八十四年實施全民健保以來，可以由健保局的財務收支表發現，醫療支出不斷地上升。究其原因，健保資源的浪費及龐大的門診申請費用有著相當程度的影響。雖然全民健保於實施之際，便採門診分級部份負擔制度，試圖藉此鼓勵民眾生小病時先至基層醫療單位就診，以促進醫療資源的配置更有效率，但由施行的結果來看，分級醫療制度發揮的功效似乎有限，政府應要了解民眾的就醫行為及習性，依此補強基層醫療院所不足之處，讓民眾信任基層醫療院所。
　　本研究特別從病患的角度，嘗試建構出一病患就診之行為模式來探討其面臨就診時的決策行為。此外我們也將異質性(heterogeneity)的觀念納入模式中，把每一位病患視為具有個別差異的不同個體。希望透過此模式來解釋每個病患的就醫行為並找出影響其跨級就診的因素，更期望能進一步的分析出病患每次就診的效用函數值來針對其選擇醫療層級的預測。在本研究中我們根據傳統的probit模式及層級貝氏方法來進行模式的建構及參數的估計，並利用MCMC方法估計貝氏模式中的參數值。
　　為說明模式的有效性，我們針對台灣全民健保研究資料庫的資料進行實證分析，希望能提供健保相關單位在進行決策制定時的參考。根據實證分析結果，我們除了成功地找出影響病患選擇就醫層級的關鍵因素外，也利用層級貝氏模型清楚地解析了病患個人的人口統計變數對個別影響就診選擇因子的影響，建構出一個更一般化的就醫層級選擇的行為模型。

關鍵字

就診層級選擇 ； 層級貝氏 ； probit模式 ； 馬可夫鏈蒙地卡羅(MCMC) ； 異質性

並列摘要

Since the national health insurance was enforced in 1995, we found that the medical treatment costs continually arise. According to the investigation, we understand that this situation is mainly caused by the waste of health insurance resource and the huge amount of hospital application costs. Though the national health insurance has tried to encourage the patient seeing a doctor in the basic level hospital (ex. clinic) in term of partial payment system, the effect of the system is obviously limited. In order to run the system more effective, understanding the patient’s behavior of taking medical treatment become a crucial task.
In this research, a traditional Probit model was constructed to illustrate patient’s decision behavior. The main objective of the developed model is to explore the identification of patient heterogeneity and to understand why patient would choose their specific hospital by using the Hierarchical Bayesian Model (HBM). To demonstrate the effectiveness of the proposed model, the data from Taiwan National Health Insurance Research Database was used. Based on actual application results, we have discovered that the key factors which affect patients choosing hospital level can be successfully identified.

並列關鍵字

Choosing the level of taking medical treatment ； Hierarchical Bayesian Model ； Probit model ； Markov Chain Monte Carlo ； Heterogeneity

參考文獻

[12] Albert, J.H. and Chib, S. (1993), “Bayesian Analysis of Binary and Polychotomous Response Data,” Journal of the American Statistical Association, 88, 669–679.

[13]　Allenby, Greg M. and Ginter, James L. (1995), “Using Extremes to Design Products and Segment Markets,” Journal of Marketing Research, 32, 392-403.

[14]　Allenby, Greg M. and Lenk , Peter J. (1994), “Modeling Household Purchase Behavior with Logistic Normal Regression,” Journal of American Statistical Association, 89, 1218-1231.

[15] Chib, S. and Greenberg, E. (1995), “ Understanding the Metropolis-Hastings algorithm,” The American Statistician, Vol. 49, No.4, 327-335.

[16] Gelfand, Alan E. and Smith, Adrian F. M. (1990), “Sampling-Based Approaches to Calculating Marginal Densities,” Journal of the American Statistical Association, 85, 398-409.