评估和监测 无烟立法及其影响 李强 加拿大滑铁卢大学心理学系 中国疾控中心控烟办 2012年7月 天津

监测（Monitoring）的定义 规律地收集、分析和解读对制定、实施和评估公共卫生实践 至关重要的卫生数据及信息（例如态度、行为、项目及政策 的状态），并及时地把这些数据提供给决策者以及时出台政 策。监测是一个连续的数据收集过程，可向相关管理者和重 要的利益相关者提供项目进展和取得成果的早期指标。一 般来说，监测包括对执行过程、策略及结果的数据收集和分 析。有时也向决策者提供正确方案的建议 We talked about a lot of terms such as monitoring, evaluation, and surveillance. These terms might be a bit confusing for some of you. Here are the definitions for them. Monitoring refers to the regular collection and analysis of data and information (e.g., attitudes, behaviours, status of programs and policies) to assist in timely decision-making, to ensure accountability and to provide the basis for evaluation and program planning. It is an ongoing function that involves the methodical collection of data to provide management and primary stakeholders of an ongoing project/program with early indications of progress and achievement of objectives. Monitoring generally involves collecting and analyzing data on implementation processes, strategies and results. It sometimes includes recommending corrective measures.

评估（Evaluation）的定义 The systematic assessment of the operation and/or outcomes of a program or policy, compared to explicit or implicit standards, in order to contribute to the improvement of a program or policy. 对项目或政策的执行及效果的系统评估。测量结果与直接 或间接指标进行比较，以改进项目或政策 In contrast, evaluation refers to the systematic assessment of the operation and/or outcomes of a program or policy, compared to explicit or implicit standards, in order to contribute to the improvement of a program or policy.

恰当的评估和监测，需要 要知道监测什么和为什么要监测 需要基线信息的特定指标 从其它地区的成功案例学习好的监测方法 宣传结果的策略 Finally, we come to monitoring the success of legislation. Monitoring is a subject on which we could spend an entire workshop. We will cover it very briefly in this presentation because implementation and enforcement efforts provide important monitoring indicators. The first question is, obviously, what you want to monitor and why. Why do you want to monitor your campaign? You may want to counter pressure to weaken or roll back the law. For this, you may want to monitor the popularity of the law, and a public opinion poll would be appropriate. You also may want to show that the law has increasing public support over time. If you want to gauge the success of implementation, you will want to measure level of compliance with the law. If you want to monitor popularity of the law, you will need a public opinion poll. Similarly, measuring economic impact, and health impact on workers, will each require different, specific monitoring methods and tools. We’ll discuss these in the next slide. Depending on which of the indicators you are monitoring, you may need baseline information to make the information more meaningful. If you are monitoring level of compliance with a new law, there is obviously no baseline measurement from beforehand. But if you want to be able to evaluate the impact of the new law on the health of workers, you will need to take baseline data before implementation of the law. You may need to gather certain data even as you are planning your campaign. You will need good monitoring methodology - most of which you will be able to use or adapt from other successful jurisdictions. Finally, you need to get your data or information out in the public eye. Most monitoring information, assuming that the data is positive (it usually is) will have a greater impact if it is disseminated among and covered by the media. You may want to provide preliminary reports to politicians, both your allies, so they can continue to champion the legislation, and possibly to those who had doubts about the law, so they can be reassured that the impact is not the negative one painted by the opponents during the campaign. 4

无烟政策的研究设计 需控制其他控烟政策的效果 前后比较+干预组对照组比较 流行病学实验 “自然实验” 前后比较 – 自身对照 现况研究 5

无烟政策效果评估的概念模型 Let’s take the evaluation of smoke free policy as an example. In the recent IARC Handbook on measuring the effectiveness of population-based tobacco control interventions, there is a chapter devoted to smokefree policy evaluation. The conceptual framework for these evaluation is presented here. With policy adoption, the main proximal variable of interest is the extent to which there is compliance with the policy, which in turn impacts SHS exposure and subsequent adverse health effects of SHS. However, there are a number of moderating factors such as the presence of other tobacco control initiatives, for example. In addition, there are incidental outcomes of interest for policymakers, including whether smokefree policies have adverse economic effects to tourism or hospitality industries. 来源: IARC Handbook. Measuring Effectiveness of Population Based Tobacco Control Interventions

常用的监测指标 公众支持/意见 遵守的程度和难点 空气质量 不同工作环境对员工的健康影响 经济影响，特别是对酒店业的影响 吸烟行为（人均消费量、流行率、戒烟意愿、开 始时间） Usually, jurisdictions are interested in monitoring one or more of the following: Public opinion: to show that the public is in favour of the new legislation Level of compliance and any difficulties with compliance: the first to show that compliance is high, and the second to take corrective action as necessary Air quality: to show that the implementation of the law has made an objective difference in lowering levels of secondhand smoke (measured by nicotine or particulate matter) Health impact on workers in various settings: to show that tworkers has improved pre- and post-implementation of law. This can include measurement of symptoms, measurement of biomarkers such as urine or blood cotinine. Economic impact, particularly on hospitality sector: to show that the economic he health of sky did not fall, as the tobacco industry had falsely predicted. This research normally uses sales or tax data, particularly for sales in the hospitality sector as a whole, and possibly for sales of food and alcoholic beverages within those premises. Smoking behaviour: to see how much of an impact the policy has on smoking levels. If there is an impact (and there usually is), the policy can be promoted as a success not only for protection of non-smokers but also for reducing tobacco use You will find resources on your CD-Rom describing how all of these types of monitoring have been done elsewhere and can find links to more detailed information. 7

无烟环境的效果 评估无烟政策的有效性 IARC预防癌症手册2009年第13卷 这一卷是来自9个国家的17位科学家的文 献回顾。 www.iarc.fr

公众支持 利用现有的人群调查 对现有的公众意见调查增加的题目付费 常用的问题： 在下面的每一个公共场所，你认为应该有怎样的禁烟规 定? 不能在室内任何区域吸烟、只能在室内某些区域吸 烟、还是没有规定或限制？

我国七城市公众对全面无烟政策的支持率

我国六城市公众对全面无烟政策的支持率

酒吧吸烟禁止令的支持率颁布前和颁布后的对比：爱尔兰（０４）苏格兰（０６）， 英国（０７），法国（０８），荷兰（０８），德国（０７－０８） • • Ireland UK (Non-Scotland) Scotland France Netherlands Germany • 61 UK Apr,Jul 07 • FR Jan 08 • IE Mar 04 48 46 SC Mar 06 43 This slide shows support among SMOKERS for smoke-free laws before and after implementation. You can see that even among smokers, support shot up after the law was put into place. And again--in Germany and Netherlands, support for the law has not changed at all. Good implementation leads to strong effectiveness and strong support among even smokers. 39 • DE 07- 08 NL Jul 08 23 • • • 17 12 14 14 15 11 9 10 7 Jan 04 Oct 04 Jan 05 Oct 05 Jan 06 Oct 06 Jan 07 Mar 07 Oct 07 Mar 08 Oct 08 Dec 08 Jul 09

遵守法律 编辑和分析监察部门获得的遵守情况的数据 主要指标：在场所内观察到吸烟现象的比例 观察法 问卷法 其他指标：是否有禁烟标志，烟具，烟头，吸烟时 是否有人劝阻 组织志愿者对方便样本进行观察

场所观察

场所观察 Cross-sectional, longitudinal

问卷调查 常用的问题： 在过去30天内，您是否去过政府大楼或政府办公室（专指公务 员办公的地方）？ 是，否 在过去30天内，您是否去过政府大楼或政府办公室（专指公务 员办公的地方）？ 是，否 在那儿是否有人在室内吸烟？ 是，否

爱尔兰 (2004), 苏格兰 (2006), 英国 (2007), 法国 (2008)酒吧禁烟前在酒吧看到吸烟的比例 98 98 98 93 96 92 Jan 04 Oct 05 06 07 Mar 08

爱尔兰 (2004), 苏格兰 (2006), 英国 (2007), 法国 (2008)酒吧禁烟前后在酒吧看到吸烟的比例 Percentage of Bars in Which ANY smoking was observed 98 98 98 93 96 92 Ireland March 2004 Scotland March 2006 UK Apr,Jul 2007 France Jan 2008 5 4 2 5 4 Jan 04 Oct 05 06 07 Mar 08

工作场所:吸烟以及对全面无烟政策的支持 19 19

餐厅:吸烟以及对全面无烟政策的支持 20 20

中国6个ITC城市餐馆内吸烟率与其他国家/城市实施无烟法规前后的比较: 爱尔兰(04), 苏格兰(06), 法国(08), 德国(07-08), 荷兰(08), 墨西哥城(08), 毛里求斯(09) • • • • 86 29 DE 07- 08 • • • 84 IE Mar 04 2 83 NL Jul 08 5 • • • • • • • MXC Apr 08 9 78 China 87-96% 71 2 FR Jan 08 62 16 MU Mar 09 59 SC Mar 06 % of Restaurants in which there was ANY smoking 21 21 21

空气质量 租赁或借用一套空气质量监测器，请志愿者在相关地点对方 便样本收集数据 寻求经验丰富的机构的协助（可能是本地区以外的机构）分 析和总结数据。 Full protocols available at: www.shsmonitoring.org and www.tobaccofreeair.org

K/OSH Presentations/Template 测量烟草烟雾污染的暴露 卷烟、雪茄、烟斗等燃烧性烟草 制品是空气中直径小于2.5微米的 悬浮颗粒的主要来源，而这些颗 粒可被深吸入肺，造成疾病。 TSI SidePak AM510 Personal Aerosol Monitor (weight: ~1 lb) An approach that we’ve used ourselves and with our many partners is to use a TSI Sidepak for particulate monitoring. The Sidepak is portable and provides real time particulate matter readings – often for particles less than 2.5 microns in diameter. The Sidepak uses a laser photometer to obtain PM concentrations. An air sampling pump draws air through the plastic tube and into the device. A laser is passed through the air entering the Sidepak. The more particles in the air, the more the light will scatter and a photometer converts that into a PM level. Few particles mean less scatter. That’s it – the machine does the hard work.

烟草烟雾颗粒 大小 2.5µm颗粒 人类头发

PM2.5对健康的影响 主要影响呼吸系统和心血管系统，无论短期或长期暴露，都 会对健康产生负面影响： 长期暴露将导致肺功能下降，慢性支气管炎，甚至早亡。 短期暴露将诱发肺部疾病，如哮喘，急性气管炎，呼吸道感染。 对心脏病患者可导致心脏病发作、心律失常等。 多项研究发现，PM2.5水平可影响医院急诊患者数及住院病 人数。 世界卫生组织得出结论： PM2.5没有安全水平。 研究估计，欧盟居民因PM2.5暴露平均减寿8.6个月。

暴露后的主要症状 健康人：眼、鼻、咽喉部的刺激症状、咳嗽、痰多、胸部发紧、 气短气促。 患有肺部疾病的人：呼吸深度受限、咳嗽、胸部不适、哮喘、 气短、疲倦。 患有心脏病的人：胸痛、胸部发紧、痰多、气短、疲倦。易导 致急性心脏病发作。

为什么使用PM2.5作为二手烟污染的指标？ 较为敏感的指标 相对更具成本效益 可以作为二手烟中化合物的指标 有一些标准可以参照 例如. 2,000:1, PM2.5:PAH 有一些标准可以参照 We’ve chosen to focus on PM2.5 assessment in this talk for the reasons shown on the slide. It is a sensitive marker for SHS. The data are available in real-time, which we’ve found to be effective particular for communicating the results and for linking specific SHS exposures to specific PM2.5 peaks. Because no lab assays are required, the net cost for collecting samples is relatively inexpensive compared to airborne nicotine or biomarkers, for example. In addition, there are published health standards for PM2.5 exposure limits, which can be used to help communicate risk to the public.

环境空气PM2.5的标准 中国：没有PM2.5标准 美国EPA标准（PM2.5）： 世界卫生组织标准（ PM2.5 ）： 24小时暴露：35微克/立方米 年平均暴露：15微克/立方米 世界卫生组织标准（ PM2.5 ）： 24小时暴露：25微克/立方米 年平均暴露：10微克/立方米

世界卫生组织标准

US EPA Air Quality Index PM2.5 (g/m3) Health Advisory Good 0-50 ≤15 None. Moderate 51-100 16-40 Unusually sensitive people should consider reducing prolonged or heavy exertion. Unhealthy for Sensitive Groups 101-150 41-65 People with heart or lung disease, older adults, and children should reduce prolonged or heavy exertion. Unhealthy 151-200 66-150 People with heart or lung disease, older adults, and children should avoid prolonged or heavy exertion. Everyone else should reduce prolonged or heavy exertion. Very Unhealthy 201-300 151-250 People with heart or lung disease, older adults, and children should avoid all physical activity outdoors. Everyone else should avoid prolonged or heavy exertion. Hazardous ≥301 ≥251 People with heart or lung disease, older adults, and children should remain indoors and keep activity levels low. Everyone else should avoid all physical activity outdoors. I want you to focus on the 3rd column. This gives the range of PM2.5 for various health warnings that are given in the US for outdoor air PM2.5. Less than 15 is ‘good’. The worst air quality is more than 250 and is labeled ‘hazardous’. The ‘Hazardous’ health advisory is pretty much that everybody should stay inside and keep activity levels low. Now think back to the last slide. Venue D had an average level of 290 inside the pub. That can’t be good for the people inside! I want you to keep that figure of 250 for ‘hazardous’ air quality in mind as I show you some more results.

二手烟是室内PM2.5污染主要来源 国外研究发现，在有人吸烟的室内，来源于二手烟中的 PM2.5约占室内PM2.5总量的90%左右。

室内吸烟与PM2.5浓度的关系 吸第三支烟 吸第二支烟 吸第一支烟 注：研究时仪器与吸烟者距离：1.5米。空间容积：122.5立方米（5m*7m*3.5m)。 吸烟人数：1

Let’s look at some of the data Let’s look at some of the data. Here is what is called a ‘real time plot’. On the x-axis is the time in minutes of measuring. On the y-axis is the PM2.5 levels. This is essentially a record of that evening’s PM2.5 levels experienced by our German team. Starting at time 0, levels outdoors are low – near 0. Then venue A is entered and PM2.5 levels increase and stay elevated until venue A is vacated. PM2.5 levels go back near 0, then spike up to over 200 when venue B is entered. Levels stay elevated inside venue B, but then drop near 0 when the team leaves. The same story holds true for venues C and D. Note the average readings inside each venue. Let’s compare with to what the US EPA has for limits for outdoor air PM2.5…

北京市部分餐厅、酒吧、办公室内PM2.5浓度 （完全禁烟，部分禁烟 及不禁烟餐馆各6家） （完全禁烟及不禁 烟办公室各6家） （不禁烟酒吧6家）

Liu R, Yang Y, Travers MJ, Fong GT, O'Connor RJ, Hyland A, Li L, Nan Y, Feng Z, Li Q, Jiang Y. A cross sectional study on levels of secondhand smoke in restaurants and bars in five cities in China. Tob Control. 2009 Sep 11. [Epub ahead of print]

自愿禁烟规定是无效的 2008年，北京市开展无烟奥运 2007年2月，倡导北京市餐馆增设无烟区 2007年4月， “在北京市餐饮业开展控烟行动”启动会

北京餐饮场所两次PM2.5监测比较 参数 回归系数 标准误 t值 P值 常数 1.28 0.16 8.12 <0.001 基本情况 § 2006年2-8月调查的92家（去掉1家极值） 2007年7-10月调查的85家 随访监测23家 § 三部分在餐馆类型和档次构成上无统计学差异，可比 § 两次调查中，不同禁烟规定餐馆构成比无统计学差异 两次监测比较 （协方差分析） 参数 回归系数 标准误 t值 P值 常数 1.28 0.16 8.12 <0.001 室外PM2.5浓度对数值 0.51 0.07 7.01 调查年份 -0.05 0.05 -0.92 0.357

北京23家餐饮场所随访PM2.5监测结果 观察结果：23家餐馆仍是3家完全禁烟，1家部分禁烟 平均值(μg/m3) 配对样本浓度对数值之差 随访监测配对t检验结果： 平均值(μg/m3) 配对样本浓度对数值之差 t值 P值 均值 标准差 标准误 第一次测量 室外PM2.5浓度 88.75 -0.11 0.52 0.11 -1.04 0.310 第二次测量 115.00 室内PM2.5浓度 166.79 0.38 0.08 -1.44 0.165 216.07

Denver June 9, 2002 This is a satellite picture of the largest forest fire the US state of Colorado history. Right in the middle of the image you can see the smoke plume rising from the pinkish flames, visible from space. Other smaller fires can be seen burning as well. The city of Denver is right in the middle of this smoke plume. Just how bad is the PM2.5 levels during a forest fire?

Denver’s visibility – before Hayman fire This is an image of the Denver skyline the day before the Hayman fire. Clear blue sky – beautiful.

Denver’s visibility – 1 day after Hayman fire Hayman Fire – Worst in Colorado History, June 8 2002 – Worst PM2.5 day in Denver Was… Denver’s visibility – 1 day after Hayman fire 24 hours later the smoke plume rolls in and masks the Denver skyline. The maximum PM2.5 reading was 200 micrograms per cubic meter. Now recall venue D in Germany at 290 micrograms per cubic meter, or the 8 countries who’s average smoking bar had levels more than 200 – the PM air quality inside these venues like a forest fire raging inside. This is a powerful communication tool to convey the message that SHS is toxic. 43 µg/m3 (24-hr max) or 200 µg/m3 (hourly max)

PM2.5可用于 法规通过前：政策倡导，动员公众/决策者 法规实施后：倡导遵守法规。前后比较，评估效果，找出问题 所在并及时改进

空气中尼古丁浓度的检测 最为特异的烟草烟雾监测指标 测量累计浓度，而非实时浓度 实验室检测较为昂贵，费时

空气中尼古丁样品采集和含量检测 制备空气中尼古丁被动式采样器； 进行环境中二手烟尼古丁暴露现场样品的采 集 个体暴露现场样品的采集 运用已建立的GC-MS分析方法对采集到的空 气样品进行检测 测试结果的横向和纵向比较可以客观的评价 环境空气中尼古丁的暴露情况，同时也可以对 控烟效果进行评价

危地马拉室内禁烟后10家酒吧空气尼古丁变化

某疾控中心室内禁烟后空气尼古丁变化

员工健康 总结其它无烟场所的同类研究的结果，以说明本地区能够期望获 得的相似结果。 在相关地点（可能人们最容易违反法律的地点）工作的不吸烟员 工中开展调查，让他们自己收集并邮寄唾液样品供可的宁检测， 同时进行电话调查以评估这些员工自我报告的烟草烟雾暴露情况 和询问呼吸系统和感觉症状。 条件允许时，可研究无烟法规实施前后后当地居民心脏病发作数 的变化

生物标志物检测 指标：唾液、尿液、头发等样品中可替宁、NNAL等 ，建立用于检测这些化合物的GC-MS和LC-MS/MS分析方法，可以 满足灵敏度高、选择性好、特异性强等检测方法的要求，而且该 系列方法均采用同位素内标方法进行定量，可以有效的避免基质 效应和提取回收率对实验结果的影响。运用该系列方法的检测结 果可以更加客观的评价烟草暴露的危害。

唾液可叮咛的变化

尿液中特有亚硝胺代谢产物（NNAL）LC-MS/MS分析检测方法的线性范围和代表图谱 Y=0.000109X+0.000611， 线性范围：0.25pg/mL~50pg/mL 烟草中特有亚硝胺代谢产物（NNAL）LC-MS/MS分析代表图谱

健康指标 指标1：心脏病发病及死亡 推荐方法：通过疾病监测数据、死亡监测数据、医院入院或 急诊数据获得心脏病发病、死亡、急诊或住院相关数据 指标2：呼吸系统症状、哮喘发病 推荐方法：问卷调查、通过疾病监测数据、医院入院或急诊 数据获得相关疾病发病、急诊或住院相关数据

海丽娜, 蒙大拿州 城市人口: 28,306 52

急性心肌梗塞的入院记录

其他指标 非吸烟者呼吸道症状及儿童哮喘入院数等 Mackay等发现苏格兰实施全面无烟立法后，15岁以下儿童 哮喘入院率较立法前下降了18.2%。 Ayres等报道苏格兰无烟立法后酒吧工作人员呼吸系统症 状发生率由69%下降至57%（p=0.02）。

经济影响 总结其它无烟场所的同类研究的结果，以说明本地区 能够期望获得的相似结果。 指标：餐厅、酒吧、旅馆营业额及税收收入数据 推荐方法：1）问卷调查法询问餐厅、酒吧、旅馆营业收 入数据；2）收集和分析相关的经济数据，例如：税收 、销售和就业状况。

经济影响 Cowling及Bond对美国加利福尼亚州餐厅、酒吧禁烟 前后这两个行业的税收数据进行了分析，结果发现，加 州1995年餐厅禁烟后餐厅业的收入较禁烟前上升，而 1998年酒吧禁烟后酒吧业的收入也有所增加。 Scollo等发现，未得烟草业资助的研究多发现全面无烟 政策不会影响餐饮业收入，而声称无烟政策会导致餐 饮业收入下降的研究均得到了烟草业的资助。

吸烟行为 总结其它无烟场所的同类研究的结果，以说明本地区能够期望获 得的相似结果。 法律实施后，应用烟草销售和制造的数据估测人均消费量的变化。 法律实施前后进行烟草使用（流行率和/或消费量）的详细研究， 同时考虑与其它可能因素（例如增税）相比，法律对消费量产生 的可能影响。 指标：吸烟者成功戒烟率、戒烟热线电话数、人群吸烟率、人均消 费量等 推荐方法：1）从相关渠道获得戒烟热线电话数、人均消费量；2）使 用问卷调查法获得人群吸烟率以及吸烟者成功戒烟情况

吸烟行为

吸烟行为 Bauer等对美国6603名吸烟者从1993年到2001 年进行了随访，发现与工作场所没有无烟政策的 吸烟者相比，工作场所实施了无烟政策的吸烟者 戒烟的可能性更高（OR=1.9，95%可信区间1.1- 3.3）。 Nagelhout等发现荷兰实施全国无烟立法后，人群 吸烟率较立法前明显下降（OR=0.91，p<0.001）。

结果的发布 发布评估结果应当作为宣传策略的一部分 好的结果应当被广泛宣传

传播策略 年度报告，充分对外公布 新闻发布 同合作组织开展的活动 选择并宣传成功的案例（强调正面的结果）

一年执行报告 Ireland United Kingdom 英格兰 爱尔兰 英国苏格兰Scotland England This is just a sample of the one-year reports that have been released following the implementation of smoke-free laws in New Zealand, Ireland, Scotland and England. The reports compile all of the evaluation and monitoring data available, even though results from the separate studies may have been released independently. Office of Tobacco Control, Ireland, Smoke-Free Workplaces in Ireland – A One Year Review Department of Health, U.K. Smokefree England – one year on Smoke-free Successes, ASH Scotland presents the Scottish experience 62

赞扬 This is a testimonial from a New Orleans, USA, restaurant owner talking about the success or her business following implementation of the smoke-free law in that city. Source: http://www.tobaccoscam.ucsf.edu/ads/ 63

无烟政策效果评估的概念模型 Let’s take the evaluation of smoke free policy as an example. In the recent IARC Handbook on measuring the effectiveness of population-based tobacco control interventions, there is a chapter devoted to smokefree policy evaluation. The conceptual framework for these evaluation is presented here. With policy adoption, the main proximal variable of interest is the extent to which there is compliance with the policy, which in turn impacts SHS exposure and subsequent adverse health effects of SHS. However, there are a number of moderating factors such as the presence of other tobacco control initiatives, for example. In addition, there are incidental outcomes of interest for policymakers, including whether smokefree policies have adverse economic effects to tourism or hospitality industries. 来源: IARC Handbook. Measuring Effectiveness of Population Based Tobacco Control Interventions

结论 监测和评估是测量政策影响的重要手段 有很多节约成本的监测方式 监测和评估结果可以帮助改进政策并提供更多的持续支持 Remember: [review bullet points] Are there any questions?