系统动力学建模在卫生资源配置研究中的应用

文章运用系统动力学建模方法,对医疗卫生服务系统卫生资源配置流向及效率进行因果关系分析,构建系统动力学流图,筛选模型变量,确立动力学方程,并对模型进行仿真模拟,为卫生资源配置研究在方法学探索上进行了有益尝试。     

基于系统动力学的城市医疗卫生服务研究

文章基于系统科学理论提出城市医疗服务体系的复杂系统,并从分析城市医疗卫生服务的参与主体及其相互关系入手,构建了城市医疗卫生服务的系统动力学模型.在此基础上,以山西省为例,使用Vensim PLE5.11软件对模型进行仿真分析.文章一方面探索系统动力学在医疗卫生服务领域的可用性;另一方面定量分析城市医疗卫生服务的现状及未来发展趋势.     

结构方程模型在健康素养分析中的应用

目的 探索结构方程模型在健康素养分析中的应用,为建立符合理论和实际的健康素养评价体系提供依据.方法 采用武汉市居民健康素养调查数据,构建结构方程模型,评估在健康素养分析中应用价值.结果 结构方程模型拟合良好,测量模型诠释了潜在变量与测量变量之间的关系,结构模型反映了潜在变量间的关系,模型选择的测量变量是健康素养的核心组...     

基于CBR的PACS系统智能辅助诊断模型研究

将数据挖掘技术与PACS系统相结合,对基于CBR(案例推理)的智能辅助诊断模型进行研究,给出模型的构造框架,并对模型功效进行分析和评估。     

结构方程模型在放射工作人员压力与健康关系中的应用

目的 探讨放射工作人员压力对健康的影响.方法 根据现代应激理论原理,采用结构方程模型对压力与健康关系进行分析.结果 利用结构方程模型清晰地展现了压力及中间变量与健康之间关系,初步展示了应激理论过程.结论 结构方程模型作为一种分析潜变量的统计方法,应用在放射工作人员压力与健康关系研究中,取得了较好的证实效果,分析了放射工作人员压力对健康的影响.     

AIC结合最优子集法构建logistic回归模型在预测2型糖尿病并发末梢神经病变中的应用

目的 研究如何快速有效地进行变量的筛选,建立起准确可靠的logistic回归预测模型;针对小样本的特性,如何对模型的泛化能力(即预测性能)进行可靠的评价;并在数据集来源于分离抽样时,对模型进行过抽样的调整,使调整后的结果适用于人群预测疾病发生的可能性.方法 以2型糖尿病并发末梢神经病变数据为例,采用最优子集法与AIC信息准则相结合对变量进行快速方便的筛选,并采用Monte Carlo模拟抽样的方法(具体为10～100次的3～10折分层交叉验证法)对模型的泛化能力作出评价和比较.结果 采用最优子集法与AIC信息准则相结合建立的logistic回归模型,准确率为79.6%,ROC面积为0.8802,经分层交叉验证法验证,泛化能力优于用一般筛选变量方法建立的模型;用先验概率对后验概率进行过抽样的调整,使调整后的结果适用于人群预测疾病发生的可能性.结论 建立logistic回归预测模型时,应根据实际情况,尽量尝试多种筛选变量的建模策略,在小样本情况下,若欲对模型的泛化能力做出可靠的评价,可采用分层交叉验证的方法;当样本来源方式为分离抽样时,若研究目的 为建立预测模型,则应采用先验概率对后验概率进行调整.     

基于PLS结构方程模型的细分算法

结构方程模型的PLS算法在满意度研究领域得到广泛应用.著名的瑞典满意度指数(SCSB)、美国顾客满意度指数(ACSI)、欧洲顾客满意度指数(ECSI)都使用了PLS算法[1].结构方程模型的内部关系是由隐变量间的一系列因果关系构成,要在PLS模型基础上对样本进行细分,我们必须考虑一种基于变量间关系的细分算法.本文引进有限混合模型的EM聚类算法,求解基于PLS结构方程模型细分问题.     

用指示变量进行多元回归方程比较

在进行两个多元回归方程比较时,要考虑两个基本问题。1.两多元回归方程的偏回归系数有无差别;2.两多元回归方程常数项有无差别。我们可以使用指示变量回归解决上述问题。     

职业紧张影响因素的结构方程模型研究

目的 阐述结构方程模型(SEM)的原理及在职业紧张影响因素研究中的应用.方法 应用职业紧张量表(OSI-R)对广东省某市2所大学的老师进行问卷调查,并建立职业任务及个体应对资源与个体紧张反应的结构方程模型.结果 职业任务与个体紧张反应以及个体应对资源与个体紧张反应的SEM拟合指数GFI、NFI均大于0.90,RMSEA均小于0.08,说明模型拟合较好.结论 结构方程模型用于职业紧张影响因素的研究,同时考虑显变量与潜变量以及潜变量之间的关系,对数据信息利用较充分,研究结果更接近真实情况.     

上海市医养整合系统的人力与床位供需仿真模拟——基于因果关系模型构建

上海市医疗卫生服务和长期照护服务体系长期分离的状态使得老年人的医养整合性需求难以被满足。文章以系统动力学为理论基础,对上海市医养整合人力与床位供需系统进行了系统性因果分析。围绕模型目标,对服务的供需双方进行分析,并结合数据的可获得性和完整程度,筛选得到供需两方面的主要变量。根据主要变量将总系统细分为相互联结的6个子系统,同时,明确了各子系统主要变量间的关系及各主要变量的来源及流向。文章为进一步建立该系统的系统动力学模型和揭示医养整合系统的人力与床位供需特征提供理论支撑,为上海市医养系统的试点与应用,切实解决健康老龄化需求无法满足的现实问题提供科学依据。     

Simultaneous estimation of parameters in the bivariate Emax model

In this paper, we explore inference in multi‐response, nonlinear models. By multi‐response, we mean models with m > 1 response variables and accordingly m relations. Each parameter/explanatory variable may appear in one or more of the relations. We study a system estimation approach for simultaneous computation and inference of the model and (co)variance parameters. For illustration, we fit a bivariate Emax model to diabetes dose‐response data. Further, the bivariate Emax model is used in a simulation study that compares the system estimation approach to equation‐by‐equation estimation. We conclude that overall, the system estimation approach performs better for the bivariate Emax model when there are dependencies among relations. The stronger the dependencies, the more we gain in precision by using system estimation rather than equation‐by‐equation estimation. Copyright © 2015 John Wiley & Sons, Ltd.     

基于机器学习的肺癌图像辅助诊断应用研究

目的:为满足肺癌临床早期筛查需求,拟在CT图像分析及病理细胞学诊断中建立智能化辅助筛查工具,提高图像分析效率,降低医生工作量。方法:在对肺癌临床早期诊断技术研究基础上,提出基于机器学习建立肺癌CT及病理切片图像辅助分析工具的方案;基于人工智能辅助诊断理念,采用图像模式增强、分割及机器学习分类模型等方法构建肺癌图像辅助诊断系统,以解决肺癌早筛的推广及应用范围受区域医疗资源分布限制的问题。结果:通过构建肺癌图像辅助诊断系统,实现了CT图像肺结节分割、数字病理细胞及细胞核分割、CT肺结节辅助筛查及细胞及细胞核辅助筛查等功能;系统阳性病例的辅助诊断准确率接近临床低年资医生的诊断水平,平均筛查时间缩短58%,为肺癌早期筛查创造条件。结论:通过构建肺癌图像辅助诊断系统,提高了图像分析效率,降低医生工作量,将一定程度上缓解区域医疗资源不平衡,为临床诊断提供辅助筛查支持,有助于提高肺癌早期筛查在临床的应用范围。     

A longitudinal study of mortality and air pollution for São Paulo,Brazil

We study the effects of various air-pollution variables on the daily death counts for people over 65 years in S?o Paulo, Brazil, from 1991 to 1993, controlling for meteorological variables. We use a state space model where the air-pollution variables enter via the latent process, and the meteorological variables via the observation equation. The latent process represents the potential mortality due to air pollution, and is estimated by Kalman filter techniques. The effect of air pollution on mortality is found to be a function of the variation in the sulphur dioxide level for the previous 3 days, whereas the other air-pollution variables (total suspended particulates, nitrogen dioxide, carbon monoxide, ozone) are not significant when sulphur dioxide is in the equation. There are significant effects of humidity and up to lag 3 of temperature, and a significant seasonal variation.     

大型医用设备的辅机故障及维修

大型医用设备主机的故障远低于与其配套的辅助设备故障，而且有些主机故障现象其实也是由辅机故障而引发的。本文重点分析了辅助设备中供配电系统电源故障，主电源稳压电路故障，冷却装置故障，环境因素的干扰，以及温湿度对设备性能的影响，提出在大型设备故障率不断降低的情况下，医疗器械工程师或操作技师应加强对辅助设备的维护和保养，及时将对大型设备的维修转变为对整套设备的预防性维护，重视辅助设备的故障隐患，关注设备的电源、工作接地、安全接地及高频设备对仪器的影响。     

Multiple imputation for interval censored data with auxiliary variables

We propose a non-parametric multiple imputation scheme, NPMLE imputation, for the analysis of interval censored survival data. Features of the method are that it converts interval-censored data problems to complete data or right censored data problems to which many standard approaches can be used, and that measures of uncertainty are easily obtained. In addition to the event time of primary interest, there are frequently other auxiliary variables that are associated with the event time. For the goal of estimating the marginal survival distribution, these auxiliary variables may provide some additional information about the event time for the interval censored observations. We extend the imputation methods to incorporate information from auxiliary variables with potentially complex structures. To conduct the imputation, we use a working failure-time proportional hazards model to define an imputing risk set for each censored observation. The imputation schemes consist of using the data in the imputing risk sets to create an exact event time for each interval censored observation. In simulation studies we show that the use of multiple imputation methods can improve the efficiency of estimators and reduce the effect of missing visits when compared to simpler approaches. We apply the approach to cytomegalovirus shedding data from an AIDS clinical trial, in which CD4 count is the auxiliary variable.     

Analysis of accelerated failure time data with dependent censoring using auxiliary variables via nonparametric multiple imputation

We consider the situation of estimating the marginal survival distribution from censored data subject to dependent censoring using auxiliary variables. We had previously developed a nonparametric multiple imputation approach. The method used two working proportional hazards (PH) models, one for the event times and the other for the censoring times, to define a nearest neighbor imputing risk set. This risk set was then used to impute failure times for censored observations. Here, we adapt the method to the situation where the event and censoring times follow accelerated failure time models and propose to use the Buckley–James estimator as the two working models. Besides studying the performances of the proposed method, we also compare the proposed method with two popular methods for handling dependent censoring through the use of auxiliary variables, inverse probability of censoring weighted and parametric multiple imputation methods, to shed light on the use of them. In a simulation study with time‐independent auxiliary variables, we show that all approaches can reduce bias due to dependent censoring. The proposed method is robust to misspecification of either one of the two working models and their link function. This indicates that a working proportional hazards model is preferred because it is more cumbersome to fit an accelerated failure time model. In contrast, the inverse probability of censoring weighted method is not robust to misspecification of the link function of the censoring time model. The parametric imputation methods rely on the specification of the event time model. The approaches are applied to a prostate cancer dataset. Copyright © 2015 John Wiley & Sons, Ltd.     

Political and welfare state determinants of infant and child health indicators: an analysis of wealthy countries

Economic indicators such as income inequality are gaining attention as putative determinants of population health. On the other hand, we are just beginning to explore the health impact on population health of political and welfare state variables such as political orientation of government or type of medical care coverage. To determine the socially structured impact of political and welfare state variables on low birth weight rate, infant mortality rate, and under-five mortality rate, we conducted an ecological study with unbalanced time-series data from 19 wealthy OECD countries for the years from 1960 to 1994. Among the political/welfare state variables, total public medical coverage was the most significant predictor of the mortality outcomes. The low birth weight rate was more sensitive to political predictors such as percentage of vote obtained by social democratic or labor parties. Overall, political and welfare state variables (including indicators of health policies) are associated with infant and child health indicators. While a strong medical care system seems crucial to some population health outcomes (e.g., the infant mortality rate), other population health outcomes might be impacted by social policies enacted by parties supporting strong welfare states (the low birth weight rate). Our investigation suggests that strong political will that advocates for more egalitarian welfare policies, including public medical services, is important in maintaining and improving the nation's health.     

Reducing the numbers of the uninsured: policy implications from state-level data analysis.

The objective of the study described in this article was to identify a model that best predicts state uninsurance rates and quantifies the contribution of socio-economic factors to enable targeted state programs to reduce uninsurance. Linear regression analysis was carried out using state uninsurance rate as the dependent variable and state-level data on demographic, employment, income, and health care environment data (independent variables). For 2000 data, the model R is 0.77, indicating that 77% of the variation in uninsurance rates is explained by the percentage of immigrant population, the workforce in very small businesses, the Black population, the state's median income, and the Medicare-aged population (model R = 0.77 for 1999 and 0.68 for 1998 data). A 1% increase in immigrant population is associated with 0.18% increase in uninsurance rate. A 1% increase in workforce employed in very small businesses associates with 0.79% increase in uninsurance. The findings indicate substantial potential for reducing uninsurance through targeted state policies. Policy recommendations are made to alleviate the insurance hurdles faced by immigrant and small business employee populations.     

Models for nearly every occasion: Part I - One box models

The standard “well mixed room,” “one box” model cannot be used to predict occupational exposures whenever the scenario involves the use of local controls. New “constant emission” one box models are proposed that permit either local exhaust or local exhaust with filtered return, coupled with general room ventilation or the recirculation of a portion of the general room exhaust. New “two box” models are presented in Part II of this series.

Both steady state and transient models were developed. The steady state equation for each model, including the standard one box steady state model, is augmented with an additional factor reflecting the fraction of time the substance was generated during each task. This addition allows the easy calculation of the average exposure for cyclic and irregular emission patterns, provided the starting and ending concentrations are zero or near zero, or the cumulative time across all tasks is long (e.g., several tasks to a full shift).

The new models introduce additional variables, such as the efficiency of the local exhaust to immediately capture freshly generated contaminant and the filtration efficiency whenever filtered exhaust is returned to the workspace. Many of the model variables are knowable (e.g., room volume and ventilation rate). A structured procedure for calibrating a model to a work scenario is introduced that can be applied to both continuous and cyclic processes. The “calibration” procedure generates estimates of the generation rate and all of remaining unknown model variables.     

Methods and results for small area estimation using smoking data from the 2008 National Health Interview Survey

The National Health Interview Survey, conducted by the National Center for Health Statistics, is designed to provide reliable design‐based estimates for a wide range of health‐related variables for national and four major geographical regions of the USA. However, state‐level or substate‐level estimates are likely to be unreliable because they are based on small sample sizes. In this paper, we compare the efficiency of different area‐level models in estimating smoking prevalence for the 50 US states and the District of Columbia. Our study is based on survey data from the 2008 National Health Interview Survey in conjunction with a number of potentially related auxiliary variables obtained from the American Community Survey, an ongoing large complex survey conducted by the US Census. A major portion of this study is devoted to the investigation of several methods for estimating survey sampling variances needed to implement an area‐level hierarchical model. Based on our findings, a hierarchical Bayesian method that uses a survey‐adjusted random sampling variance model to capture the complex survey sampling variability appears to be somewhat superior to the other considered area‐level models in accounting for small sample behavior of estimated survey sampling variances. Several diagnostic procedures are presented to compare the proposed methods. Copyright © 2014 John Wiley & Sons, Ltd.