本文選題：貝葉斯傾向性評分 + 傾向性評分法　； 參考：《第二軍醫(yī)大學(xué)》2014年碩士論文

【摘要】：研究背景： 目前，我國藥品不良反應(yīng)自發(fā)呈報系統(tǒng)的報告數(shù)量飛速增長。如何借助恰當(dāng)?shù)臄?shù)據(jù)挖掘和統(tǒng)計分析方法利用這些數(shù)據(jù)，高效、準(zhǔn)確的發(fā)現(xiàn)藥品安全問題，關(guān)系到廣大患者的生命健康，也是公共衛(wèi)生研究的重點問題。但自發(fā)呈報系統(tǒng)的數(shù)據(jù)中存在著較多的混雜因素，若在信號挖掘中忽略或者未采用適當(dāng)?shù)慕y(tǒng)計方法均衡混雜因素，會影響到信號檢測結(jié)果的準(zhǔn)確性。本課題組已經(jīng)實現(xiàn)了傳統(tǒng)的Logistic傾向性評分法與信號挖掘方法的結(jié)合以控制混雜因素的干擾。然而，傳統(tǒng)的logistic回歸傾向性評分法在應(yīng)用中存在的問題也不容忽視。如忽略了連續(xù)型協(xié)變量與logit(y)呈線性關(guān)系的條件、未考慮到傾向性評分值的不準(zhǔn)確性對處理效應(yīng)的影響、無法利用先驗分布的信息、結(jié)局事件為罕見事件或存在較多協(xié)變量時不適用等問題。 研究目的： 本研究探索貝葉斯思想與傾向性評分分層法、匹配法、加權(quán)法及回歸法的結(jié)合以構(gòu)建各種貝葉斯傾向性評分模型。同時，通過模擬研究比較在不同樣本量、不同處理效應(yīng)強度以及不同先驗分布精度情況下各種貝葉斯傾向性評分模型估計處理效應(yīng)的準(zhǔn)確性及精確性。最后，選出最優(yōu)模型應(yīng)用于藥品不良反應(yīng)信號檢測中，以提高藥品不良反應(yīng)信號檢測的準(zhǔn)確性。 研究方法： 本研究采用蒙特卡洛法模擬數(shù)據(jù)集。模型的建立采用兩次建模，先建立處理因素與協(xié)變量的關(guān)系，再建立結(jié)局變量與處理因素、協(xié)變量的關(guān)系。根據(jù)協(xié)變量與結(jié)局變量和處理因素的關(guān)系，分別模擬只和處理因素有關(guān)的協(xié)變量、與處理因素和結(jié)局變量都有關(guān)的協(xié)變量、只和結(jié)局變量有關(guān)的協(xié)變量。根據(jù)協(xié)變量的分布類型分別模擬二分類與連續(xù)型變量。研究設(shè)置了5種不同強度的處理效應(yīng)（t1.5,1.2,0.8,0.5,0.2）、四種先驗信息的精度（B0,1,10,100）及三種樣本量情況（N=50、100、250）。 本研究先構(gòu)建貝葉斯Logistic回歸模型估計傾向性評分值，再與不同的傾向性評分利用方式相結(jié)合（如傾向性評分匹配法、傾向性評分分層法、傾向性評分加權(quán)法及傾向性評分回歸調(diào)整法）及不同的處理效應(yīng)估計方法相結(jié)合，共衍生七種貝葉斯傾向性評分模型。研究從處理效應(yīng)的點估計值、標(biāo)準(zhǔn)差、偏倚（絕對偏倚、相對偏倚）、均方誤差及置信區(qū)間的覆蓋率等方面進行結(jié)果的穩(wěn)健性及準(zhǔn)確性評價。并借助SAS及R軟件，編寫程序完成方法的軟件自動化實踐。 最后，研究根據(jù)評價結(jié)果選出準(zhǔn)確性及穩(wěn)健性較好的貝葉斯傾向評分匹配法（設(shè)置B100）并將其應(yīng)用于上海市食品藥品監(jiān)督管理局2009年自發(fā)呈報系統(tǒng)的數(shù)據(jù)中。通過與國家藥品-不良反應(yīng)字典比對、查閱國內(nèi)外文獻及藥品說明書，以驗證貝葉斯傾向性評分模型、Logistic傾向性評分模型及常規(guī)方法的信號檢測的準(zhǔn)確性。 研究結(jié)果： 在小樣本的情況下（N=50）,貝葉斯傾向性評分匹配法及貝葉斯傾向性評分分層法+CMH受先驗分布信息精度的影響較大。當(dāng)先驗分布的精度設(shè)置為100時，這兩種方法的準(zhǔn)確性及穩(wěn)定性較傳統(tǒng)的傾向性評分法有著明顯的提高。尤其是貝葉斯傾向性評分匹配法，其處理效應(yīng)的點估計及置信區(qū)間長度較傳統(tǒng)的傾向性評分模型明顯改善。 當(dāng)樣本量設(shè)置為100時，貝葉斯傾向性評分匹配法及貝葉斯傾向性評分分層法+CMH的結(jié)果較傳統(tǒng)的傾向性評分法較為接近，其穩(wěn)定性略微提高，但偏倚略有增大。當(dāng)進一步增大樣本到250時，貝葉斯傾向性評分匹配法及貝葉斯傾向性評分分層法+CMH的結(jié)果與傳統(tǒng)的傾向性評分結(jié)果極為接近。同時，樣本量較大的情況下，處理效應(yīng)的點估計也更加接近預(yù)設(shè)的真值。 無論是貝葉斯傾向性評分還是傳統(tǒng)的傾向性評分，傾向性評分值的處理方式的選擇對處理效應(yīng)估計的準(zhǔn)確性及穩(wěn)性影響較大。匹配法及分層法較好，可作為首選。而加權(quán)法及回歸法的穩(wěn)定性較差，受處理因素與結(jié)局變量關(guān)聯(lián)強度影響較大。 實際數(shù)據(jù)應(yīng)用中，，在未采用貝葉斯傾向性評分法及傾向性評分法對基線協(xié)變量進行均衡之前，發(fā)現(xiàn)“阿奇霉素-局部麻木”的組合在四種信號挖掘方法中均檢測為陽性信號，通過與國家藥品-不良反應(yīng)字典比對、查閱國內(nèi)外文獻及藥品說明書判定該信號為假陽性信號。單因素分析發(fā)現(xiàn)年齡、體重、性別、合并用藥及不良反應(yīng)發(fā)生季節(jié)等混雜因素在組間的分布不均衡。而采用貝葉斯傾向性評分匹配法及傳統(tǒng)的Logistic傾向性評分匹配法對基線協(xié)變量進行均衡后，發(fā)現(xiàn)采用前種方法后所有基線斜變量在組間的分布均衡，而采用后者仍舊有部分基線協(xié)變量（如體重、不良反應(yīng)發(fā)生季節(jié)）在組間的分布不均衡。從而可見貝葉斯傾向性評分法混雜因素的均衡效果較傳統(tǒng)的Logistic傾向性評分匹配法有優(yōu)勢。雖然貝葉斯傾向性評分法與Logistic傾向性評分法在最終結(jié)果判定時均得到“阿奇霉素-局部麻木”假陽性的結(jié)論，但Logistic傾向性評分法估計的信號值與閾值較為接近，更容易得到陰性的結(jié)論。 研究結(jié)論： 當(dāng)先驗信息的精度較高（B100）且樣本量較小（N=50）時，貝葉斯傾向評分匹配法及貝葉斯傾向性評分分層法+CMH較傳統(tǒng)的傾向性評分法有著較大的優(yōu)勢。在藥品不良反應(yīng)信號檢測的實際數(shù)據(jù)應(yīng)用中，貝葉斯傾向評分匹配法（設(shè)置B100）較傳統(tǒng)的傾向性評分匹配法的均衡性好，可以降低混雜因素導(dǎo)致的偏倚，在一定程度上可去除可疑的假陽性信號。
[Abstract]:Background of Study :

At present , the number of reports of spontaneous reporting system of drug reactions in China has increased rapidly . How to use these data with proper data mining and statistical analysis methods is a key problem in the study of public health . However , there are many problems in the data of spontaneous reporting system .

Purpose of study :

Finally , the optimal model was applied to the detection of adverse drug reaction signals to improve the accuracy of drug reaction signal detection .

Study method :

In this study , Monte Carlo method was used to simulate the data set . The relationship between treatment factors and covariables was established , and the relationship between outcome variables and treatment factors and covariables was established . Based on the relationship between covariables and outcome variables and processing factors , the covariables related to treatment factors and outcome variables were simulated respectively . Five different intensities of processing effects ( t1 . 5 , 1.2 , 0.8 , 0.5 , 0.2 ) were established .

In this study , a Bayesian logistic regression model was constructed to estimate the value of tendentious score , and then combined with different methods of propensity score utilization ( such as propensity score matching method , propensity score layering method , propensity score weighting method and propensity score regression adjustment method ) and different treatment effect estimation methods . Seven kinds of Bayesian propensity score models were derived . The results were evaluated from the point estimate , standard deviation , bias ( absolute bias , relative bias ) , mean square error and coverage rate of confidence interval .

Finally , the Bayesian propensity score matching method ( B100 ) was selected and applied to the data of Shanghai Food and Drug Administration in 2009 .

Results of the study :

In the case of small samples ( N = 50 ) , the Bayesian propensity score matching method and the Bayesian propensity score hierarchy method + CMH are affected by the accuracy of the prior distribution information . When the accuracy of the prior distribution is set to 100 , the accuracy and stability of the two methods are obviously improved compared with the conventional propensity score method . Especially , the Bayesian propensity score matching method improves the point estimation and confidence interval length of the processing effect obviously .

When the sample size is set to 100 , the Bayesian propensity score matching method and the Bayesian propensity score stratification method + CMH are relatively close to the traditional propensity score method , but the stability is slightly increased . When the sample is further increased to 250 , the Bayesian propensity score matching method and the Bayesian propensity score stratification method + CMH results are very close to the traditional propensity score results . In the meantime , when the sample size is large , the point estimation of the processing effect is closer to the preset true value .

The selection of the treatment mode of the propensity score value has a great influence on the accuracy and stability of the treatment effect estimation , whether it is a Bayesian propensity score or a traditional propensity score , and the matching method and the stratification method are better and can be used as the first choice .

It was found that the combination of azithromycin and local anaesthesia was detected as a positive signal in four kinds of signal mining methods without using the Bayesian propensity score method and the propensity score method .

Conclusions of the study :

When the accuracy of the prior information is high ( B100 ) and the sample size is small ( N = 50 ) , the Bayesian propensity score matching method and the Bayesian propensity score stratification method + CMH have great advantages . In the actual data application of the drug reaction signal detection , the Bayesian propensity score matching method ( setting B100 ) is better than the traditional propensity score matching method , and the bias caused by confounders can be reduced , and the suspected false positive signal can be removed to a certain extent .
【學(xué)位授予單位】：第二軍醫(yī)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：R95

【參考文獻】

相關(guān)期刊論文 前2條

1 錢維;葉小飛;王超;賀佳;;藥品不良反應(yīng)信號檢測中混雜因素的控制方法[J];中國藥物警戒;2010年03期

2 田春華;杜曉曦;;論我國藥品不良反應(yīng)監(jiān)測工作幾點進展[J];藥物流行病學(xué)雜志;2014年01期

本文編號：2086980