季节性求和自回归移动平均模型在北京市房山区感染性腹泻发病趋势预测中的应用

目的 构建北京市房山区感染性腹泻发病的季节性求和自回归移动平均(seasonal autoregressive integrated moving average,SARIMA)模型并进行预测。方法 应用R 3.0.1软件程序包中的TSA对2004-2013年房山区感染性腹泻月发病率构建模型,并对2014年各月感染性腹泻月发病率进行预测和评价。结果 SARIMA(0, 0, 2)(0, 1, 1)12模型较好地拟合既往时间段月发病率,对2014年发病趋势拟合平均相对误差为19.164%,对年发病率拟合平均相对误差为2.303%。结论 SARIMA(0, 0, 2)(0, 1, 1)12模型能够很好拟合感染性腹泻月发病率数据,可用于房山区感染性腹泻发病趋势的短期预测,为下一步采取针对性防控措施提供科学依据。     

Influenza A and B epidemic criteria based on time-series analysis of health services surveillance data

Many countries now have epidemiological surveillance systems using health services-based indicators that allow detection of influenza epidemics. However, there is no accepted criterion for defining an influenza epidemic. An epidemic criterion has been developed, based on a time-series analysis of health services-based indicators collected on a weekly basis by a surveillance network implemented in the Paris region since 1984: the Groupe Régional d'Observation de la Grippe (GROG). For each new season, an epidemic threshold is independently defined for each health services-based indicator as the upper limit of the one-sided confidence interval of the expected value calculated from the weekly differences between the observed number of events and those predicted by a SARIMA model fitted on the non-epidemic data of previous seasons. Epidemic criteria for influenza A and B are then defined from the combination of both viral indicators and epidemic thresholds of individual health services-based indicators. Among health indicators, sick-leave data collected from GP's or the Health Insurance system, emergency home medical visits, and influenza-like-illness reported by GP's are the most sensitive indicators for the early recognition of epidemics. The exceeding of the above mentioned thresholds combined with virological data allows the specific detection of influenza A or B epidemics. This time-series method of analysing surveillance data provides early and reliable recognition of these epidemics.     

气候变化对长沙市肾综合征出血热发病的影响与预警模型

目的 了解气象因素变化对肾综合征出血热(HFRS)发病的影响,探索应用气象因素对HFRS发病进行预警。方法 收集长沙市2000-2009年HFRS病例（共2171例）,同时收集同期气象数据,构建基于气象因素的长沙市HFRS传播预测模型,使用Cochran-Armitage趋势检验分析HFRS年发病率的变化趋势,采用交叉相关分析法计算气象因素[包括月平均温度、相对湿度、降水量及厄尔尼诺南方涛动指数(MEI)]与每月HFRS发病人数之间的时滞周期,最后采用时间序列泊松回归模型分析不同气象因素对HFRS传播的影响。结果 2000-2009年长沙市HFRS年发病率分别为13.09/10万（755例）、9.92/10万（578例）、5.02/10万（294例）、2.55/10万（150例）、1.13/10万(67例)、1.16/10万（70例）、0.95/10万（58例）、1.40/10万（87例）、0.75/10万（47例）、1.02/10万（65例）,整体呈下降趋势（Z= -5.78,P＜0.01）。模型分析显示,月平均气温[18.00℃,r=0.26,P ＜0.01,1个月时滞周期;发病率比(IRR)=1.02,95％CI:1.00～1.03,P＜0.01]、相对湿度（75.50％,r=0.62,P＜0.01,3个月时滞周期;IRR= 1.03,95％CI:1.02～ 1.04,P＜ 0.01）、降水量（112.40mm,r=0.25,P＜0.01,6个月时滞周期;IRR= 1.01,95％CI:1.01～1.02,P=0.02）和MEI（r=0.31,P＜0.01,3个月时滞周期;IRR =0.77,95％ CI:0.67～0.88,P＜0.01）与HFRS月发病人数（18.10例）紧密相关。结论 气象因素对HFRS发病存在明显影响,在控制变量自相关、季节性及长期趋势的影响后,长沙市时间序列泊松回归模型预测精度较高,可以实现对长沙市HFRS的提前预警。     

宁波市大气污染物一氧化碳及臭氧对人群心肌梗死死亡的急性效应研究

目的 探讨宁波市大气污染物一氧化碳（CO）及臭氧（O₃）对人群心肌梗死（心梗）死亡的影响。方法 收集宁波市2011年1月1日至2015年12月31日每日空气质量监测数据、同期气象监测数据及死因监测数据资料,控制时间长期趋势、气象因素及星期几效应等混杂因素,采用广义相加模型进行分析。结果 2011-2015年宁波市CO及O₃日均浓度分别为0.90（0.02～3.31）mg/m³及82.78（4～236）μg/m³,心梗死亡共5 388例,日均死亡3例。单污染物模型显示滞后第6天CO日均浓度每升高0.1 mg/m³造成总人群心梗死亡风险增加1.06%（95% CI：0.29%～1.93%）。≥65岁人群风险增加1.26%（95% CI：0.28%～2.24%）,但对<65岁人群影响的差异无统计学意义;对男性人群影响的差异亦无统计学意义,但对女性人群风险增加1.77%（95% CI：0.44%～3.13%）。O₃对死亡影响的差异无统计学意义。结论 宁波市CO浓度变化与人群心梗死亡风险增加相关,未发现O₃浓度变化与心梗死亡风险的关联。     

淄博市大气污染物浓度与急救人次关联的时间序列分析

目的 探讨淄博市大气污染物［SO₂、NO₂、细颗粒物(PM_2.5)］浓度对急救人次的影响。 方法 收集2016年1月1日至2017年12月31日淄博市的急救医疗数据,按照WHO国际疾病分类(ICD-10)统计每日因非意外总急救及呼吸系统疾病急救的人次,结合同期逐日空气污染数据和气象数据,利用广义相加模型(GAM)控制长期趋势、季节趋势、星期几效应及气象因素的影响后,分析SO₂、NO₂、PM_2.5日均浓度与非意外、呼吸系统疾病急救人次的关系。 结果 淄博市2016年至2017年SO₂日均浓度52.6 μg/m³,NO₂日平均浓度57.0 μg/m³,PM_2.5日平均浓度70.4 μg/m³。2016年至2017年日平均急救366.78人次,其中日均非意外急救63.98人次,日均呼吸系统急救8.04人次。相关性分析表明,日均非意外急救及呼吸系统急救人次均与PM_2.5、SO₂、NO₂、CO浓度呈正相关。时间序列分析单污染物模型显示,SO₂和NO₂浓度均对非意外急救人次的影响存在滞后效应,滞后1 d健康效应最强,其中SO₂浓度每增加10 μg/m³,非意外急救人次增加0.423%(95%CI:0.253%~0.601%),NO₂浓度每增加10 μg/m³,非意外急救人次增加0.412%(95%CI:0.218%~0.621%);PM_2.5浓度均对呼吸系统疾病急救人次的影响存在滞后效应,滞后3 d健康效应最强,其中PM_2.5浓度每增加10 μg/m³,呼吸系统疾病急救人次增加0.314%(95%CI:0.178%~0.533%)。双污染物模型中,在分别引入PM_2.5和O₃后,SO₂浓度每升高10 μg/m³,非意外总急救人次的超额危险度(ER)分别为0.286%(95%CI:0.061%~0.519%)、0.389%(95%CI:0.229%~0.671%),NO₂浓度每升高10 μg/m³,非意外总急救人次的ER分别为0.176%(95%CI:0.117%~0.561%)、0.427%(95%CI:0.287%~0.663%)。双污染物模型中,分别引入SO₂和NO₂后,PM_2.5浓度每升高10 μg/m³,呼吸系统疾病急救人次的ER分别为0.219%(95%CI:0.128%~0.456%)、0.193%(95%CI:0.101%~0.429%)。 结论 大气中SO₂或NO₂浓度的升高可能增加居民非意外急救的风险,PM_2.5浓度的升高可能增加居民呼吸系统疾病急救的风险,应加强环保整治力度,加快能源改革,发展清洁能源,减少大气污染物的排放,促进居民身体健康。     

心脑血管与呼吸疾病就诊人数与气象因素关系

[目的]探讨日均气温、相对湿度和空气压力与医院日心脑血管与呼吸疾病急、门诊就诊人次的相关性.[方法]收集三级甲等医院日呼吸、心脑血管疾病急、门诊就诊人次和该地区日均气温、相对湿度和空气压力的时间序列数据,采用立方平滑样条函数将日均气温、相对湿度和空气压力引入,与医院日呼吸、心脑血管疾病急、门诊就诊人次间建立Poisson广义相加模型,选择最终进入模型的变量,并确定其立方平滑样条函数自由度取值. [结果]医院日呼吸、心脑血管疾病急、门诊就诊人次与气温、相对湿度和空气压力表现出一定的线性相关趋势;当气温＜ 10℃时,气温每升高1℃,就诊总人数增加1.057 217倍,当气温≥10℃,气温每增加1℃,就诊总人数上年同期下降0.990 571 73倍. [结论]气象因素与呼吸、心脑血管疾病急、门诊就诊人次的相关性比较明显.     

Ambient ozone concentration and hospital admissions due to childhood respiratory diseases in New York State, 1991-2001

Background and objective

Respiratory diseases constitute one of the leading causes of ill health among children in New York State (NYS). The current project, a component of the NYS Environmental Public Health Tracking Program, investigated a potential association between ambient ozone (O₃) concentration and childhood respiratory hospital admissions over 11 years in NYS.

Methods

We used a two-stage Bayesian hierarchical model to assess the exposure-disease associations within 11 geographic regions and statewide. The data included total daily hospital admissions due to respiratory diseases for children 0-17 years old from 1991 to 2001 in NYS (N=134,099) and daily ambient O₃ level with different single-day lags. These analyses adjusted for particulate matter ?10 μm in size (PM₁₀), meteorological conditions, day of the week, seasonality, long-term trends, and demographic characteristics.

Results

In 5 of the 11 regions, including the Upper and Lower Adirondacks, Upper Hudson Valley, Staten Island, and New York City, positive associations were found between respiratory hospital admissions and ambient O₃ level 2 days prior to the admission. Applying different statistical methods and sensitivity analysis of PM₁₀ did not alter these findings. When region-specific results were combined, no statewide association was apparent.

Conclusions

Geographic differences were found in the associations between O₃ levels and respiratory hospital admissions among children. In addition, we found that the two-stage model may be an appropriate approach for tracking the health effects of air pollution over time in different geographic areas when heterogeneity of risk factors across regions is present.     

National independence,women's political participation,and life expectancy in Norway

This study investigates the role of national independence and women's political participation on population health using historical lifespan data from Norway. We use time-series methods to analyze data measuring the actual length of time lived by Norwegian birth cohorts spanning a 61 year period surrounding the political emancipation of Norway from Sweden in 1905 and the establishment of a Norwegian monarchy in 1906. The use of a discrete, historical event improves our ability to interpret the population health effects of national independence and women's political participation as causal. We find a large and significant positive effect on the lifespan of Norwegian females born in the 1906 cohort. Interestingly, the effect does not extend to all living females during the Norwegian drive toward sovereignty. We conclude that the beneficial effects were likely conferred through intrauterine biological transfers and/or neonatal investments specific to the first year of life.     

合肥市大气污染与肺炎门诊量的时间序列研究

目的探讨大气污染物SO_2、NO_2和PM_(2.5)浓度与合肥市滨湖医院肺炎日门诊量之间的关系。方法采用时间序列分析的广义相加Poisson回归模型,在控制长期趋势、星期几效应和气象因素等混杂因素的影响后,定量分析2014年安徽省合肥市大气污染物SO_2、NO_2、PM_(2.5)日均浓度与滨湖医院肺炎日门诊量的关系及滞后效应。结果单污染物模型中,在控制了长期趋势、星期几效应和气象因素的影响后,SO_2在滞后3、4、5 d(lag3、lag4、lag5)时对肺炎日门诊量的影响有统计学意义(P0.05),NO_2滞后2、3、4、5 d(lag2、lag3、lag4、lag5)时的影响有统计学意义(P0.01),PM_(2.5)滞后3、4 d(lag3、lag4)时的影响有统计学意义(P0.05);SO_2、NO_2、PM_(2.5)的滞后效应分别在lag3、lag2、lag4时最明显,当SO_2、NO_2、PM_(2.5)浓度每升高10μg/m~3时,肺炎日门诊量分别增加1.54%(95%CI:0.28%~2.81%),1.98%(95%CI:0.89%~3.08%)和0.28%(95%CI:0.06%~0.50%)。多污染物模型中,当模型中引入两种或两种以上的污染物后,各污染物对肺炎日门诊量的效应估计值均较单污染物模型降低,但并不改变各污染物与肺炎日门诊量之间的正向关联。结论合肥市大气污染物SO_2、NO_2、PM_(2.5)浓度升高可能引起医院肺炎日门诊量增加,且有一定的滞后效应。