WPC: 2B R ZV,#|KHP LaserJet 4HPLAS4.PRSx  @\ubX@ 3'3'Standard6&6&StandardHPLAS4.PRSx     USUK#x6X@8;2X@#  X` hp x (#%'0*,.8135@8:DgFdddgF9CgggEEggzzddd~rrEgdF"m^gCdddgC9CgggEEggzzdddsrrEgdC2Vvpkka8DocumentgDocument Style StyleXX` `  ` a4DocumentgDocument Style Style . a6DocumentgDocument Style Style GX  a5DocumentgDocument Style Style }X(# 2v-ta2DocumentgDocument Style Style<o   ?  A.  a7DocumentgDocument Style StyleyXX` ` (#` BibliogrphyBibliography:X (# a1Right ParRight-Aligned Paragraph Numbers:`S@ I.  X(# 2   - a2Right ParRight-Aligned Paragraph Numbers C @` A. ` ` (#` a3DocumentgDocument Style Style B b  ?  1.  a3Right ParRight-Aligned Paragraph Numbers L! ` ` @P 1. ` `  (# a4Right ParRight-Aligned Paragraph Numbers Uj` `  @ a. ` (# 2 Ga5Right ParRight-Aligned Paragraph Numbers _o` `  @h(1)  hh#(#h a6Right ParRight-Aligned Paragraph Numbersh` `  hh#@$(a) hh#((# a7Right ParRight-Aligned Paragraph NumberspfJ` `  hh#(@*i) (h-(# a8Right ParRight-Aligned Paragraph NumbersyW"3!` `  hh#(-@p/a) -pp2(#p 2"ua1DocumentgDocument Style StyleXqq   l ^) I. ׃  Tech InitInitialise Technical Style. k I. A. 1. a.(1)(a) i) a) 1 .1 .1 .1 .1 .1 .1 .1 Technicala5TechnicalTechnical Document Style)WD (1) . a6TechnicalTechnical Document Style)D (a) . 2 .) a2TechnicalTechnical Document Style<6  ?  A.   a3TechnicalTechnical Document Style9Wg  2  1.   a4TechnicalTechnical Document Style8bv{ 2  a.   a1TechnicalTechnical Document StyleF!<  ?  I.   2['!!)"/,&a7TechnicalTechnical Document Style(@D i) . a8TechnicalTechnical Document Style(D a) . PleadingHeader for numbered pleading paperP@n   $] X X` hp x (#%'0*,.8135@8:a) ;ppA"p Techn actifActiver style document technique(e I. A. 1. a.(1)(a) i) a) 1 .1 .1 .1 .1 .1 .1 .1 Technique22)50*0+1,02a2TechniqueStyle document technique)<  yO  A.   a3TechniqueStyle document technique*9Xh  yO  1.   a4TechniqueStyle document technique+8X| yO  a.   a1TechniqueStyle document technique,Ff!<  e  I.   26-/$3.S4/405Doc actifActiver style documentle-H  p-p-p-  I. A. 1. a.(1)(a) i) a) I. 1. A. a.(1)(a) i) a)Documentga5TechniqueStyle document technique.9: yO  (1) . a6TechniqueStyle document technique/9z yO  (a) . a7TechniqueStyle document technique08 k yO  i) . 21I626:a8TechniqueStyle document technique18n yO  a) . Papier marg)R/gle de num)rotation de lignes2    X X` hp x (#%'0*,.8135@8:" y9v\'dddy x y\'dddy Proportion of the population 15 years of age and over that smokes ee365    X&  It must be borne in mind that the nature of the estimate of the proportion of smokers in the population will affect the accuracy of the results since the estimate of annual per capita consumption includes cigarettes smoked by all smokers, whether daily)-++ or occasional. If the proportion of smokers used to prepare the consumptionbased estimate is daily smokers, the procedure will yield an overestimate of cigarettes smoked per day by daily smokers. If the proportion used refers to all smokers, this will underestimate the number of cigarettes smoked per day by daily smokers. This consumptionbased estimate of the mean number of cigarettes smoked per day by smokers is likely to be substantially higher than the surveybased estimate, (calculated from smokers' self reports), consistent with the finding that smokers tend to underreport the number of cigarettes they smoke each day. Figure VII.1 illustrates the extent of this underreporting based on data for the United States. For the majority of countries, the consumptionbased estimate of the mean number of cigarettes smoked each day by smokers should fall between about 10 and 30; if numbers outside this range are obtained, data should be reexamined. SOURCES OF DATA TO CALCULATE CONSUMPTION In principle, data on the sale of tobacco products in a given year will provide the most direct indication of the amount of tobacco consumed by the population in that year. Sales data are generally closely monitored by governments for tax collection purposes and can be obtained from government publications or agencies. Where reliable sales data are not available, consumption can be indirectly estimated from data on the production and trade of tobacco products. These statistics are most commonly available for manufactured cigarettes. Government statistical offices collect and publish these data. Statistics on the production and trade of tobacco products are routinely reported to the United Nations Statistical Office and published in the Statistical Yearbook of the United Nations. The United States Department of Agriculture (Tobacco, Cotton and Seeds Division) also collect these data through a network of country offices.  For each country, population data are generally available from the Census Department or Central Statistics Office. These data are typically available for geographic or administrative regions of a country and can be used for calculating variations in per capita consumption within a country, where the appropriate cigarette C*-++ consumption data are also available. Periodic estimates of population size and agesex structure are published biennially by the United Nations Population Division. These estimates are a convenient and comparable source for making comparisons between countries and may be used where reliable national population estimates are not readily available.  X.  PROBLEMS INHERENT IN THE USE OF PER CAPITA DATA Although per capita consumption is a valuable indicator of overall smoking levels, there are some limitations that must be considered. For example, although cigarette production figures are typically based on the assumption of one gram of tobacco per cigarette, this is not always the case. In the past, cigarettes frequently contained more than one gram of tobacco. Conversely, in some countries in more recent years, one cigarette now contains less than one gram of tobacco (See Figure VII.2).  For some countries, cigarette production data may not be available and any published statistics may have been estimated from the weight of tobacco produced. This will increase the possibility of built in biases depending on the sources. If the estimate of cigarette production remains constant for many years, or if it appears to be unrealistically high or low compared with similar or neighbouring countries, or compared with other data sources, the estimate may be unreliable. In other countries, although there may be an abundance of data from different sources, they may not be in agreement. Therefore,  X one should consider all data for their apparent validity and also for their consistency  X[ among different data sources.#Xi\  P6ƒXP# Smuggling can seriously distort estimates of apparent consumption. For example, smuggling has been estimated to account for 10% to 30% of consumption in Canada and Belgium, and consumption estimates for these countries have been adjusted to take into account smuggling. Such an adjustment is only possible in a few countries. There are, however, many other methods which may be used to develop more accurate estimates of the level of smuggling. Law enforcement and tax collection officials may have such estimates, as might the tobacco industry (though biases need to be considered in usingC*-++ such data). Efforts to reduce total smuggling, such as prominent tax markings and increased fines and law enforcement, can not only reduce the level of smuggling but make its detection (and therefore overall measurement of tobacco consumption) much easier. In 1994, smuggling represented about 5% of world manufactured cigarette consumption. It is possible to make this approximate estimate of global smuggling by examining the excess of world cigarette exports over imports. Smuggling has been increasing during the 1980s and 1990s. By 1994 roughly 30% of all cigarettes that were "exported" by tobacco companies appear to have ended up going through the hands of smugglers rather than legal distribution channels. For those countries with a small indigenous population but a very large transient (e.g. tourist) population, care must be taken to distinguish between the consumption of the two groups. Sometimes this is possible because different taxes are levied on sales to the indigenous and transient populations, and a ratio of consumption for the two groups can be established. For example, in Cyprus, the indigenous adult population of 522 000 buy cigarettes taxed at the national rate but a transient population of about 1.5 million tourists per year and a nonindigenous population of military personnel and their families all buy untaxed cigarettes. Countries in such a situation can develop a more accurate estimate of domestic consumption by requiring different accounting for, taxation of, or markings on, such products. Excessive yeartoyear fluctuations in per capita data are unusual except in times of warfare or national catastrophe and when fluctuations do appear in trends, they should be viewed with caution. Minor fluctuations in per capita consumption trends can be removed by using statistical smoothing techniques such as 3year moving averages. It must be kept in mind that cigarette sales data only approximate true cigarette consumption in any given year. Not all cigarettes produced or imported into a country will be consumed in that year. If beginning and endofyear cigarette inventories do not vary greatly from each other, the estimated trend in consumption, based on production and trade of cigarettes, may be reasonably reliable. Annual sales data may also be affected by the timing of shipments from producers and stock piling among wholesalersC*-++ and retailers. Such actions can be an effort to undermine health measures by making them look less effective or less necessary. They can also be related to tobacco control measures as companies ship out additional cigarettes in order to avoid approaching tax increases or improved package health messages. Consumption figures, based on cigarette sales data may also suffer inaccuracies due to the use of homegrown tobacco and other forms of cigarettes which are generally at least as hazardous as manufactured cigarettes. These include "rollyourown" (RYO)  X or handrolled cigarettes, which are popular in a number of countries, as well as bidis (traditional locallyproduced cigarettes) which are extremely common in countries of South Asia. In India, for example, bidis account for between 80 and 90% of all smoked tobacco products, including cigarettes. To ignore them would lead to a massive underestimation of current, and particularly future health effects of tobacco use in India. Wherever reliable estimates of the consumption of these types of cigarettes are available, they should be included in the overall population estimate of cigarette consumption.  X  ESTIMATING PER CAPITA CONSUMPTION: THE EXAMPLE OF AUSTRALIA As an illustration of how these various issues affect estimated consumption trends, different estimates of cigarette consumption in Australia are summarized in Figure VII.3. Three time series are shown. The series labelled A is based on national sales data. Series B are estimates of cigarette consumption based on production and trade data ((production + imports exports/population 15+), and series C is the same source (i.e. production and trade data) with the inclusion of estimates for "roll your own" cigarettes. The overall trend pattern is broadly similar between series A and series B (sales and production/trade data), indicating that indirect methods to estimate consumption may well provide a good approximation to actual consumption in many countries. The effect of "roll your own" cigarettes (estimated as 85% of loose tobacco sales in this case), is to push up the trend line estimated from production and trade data, with the effect progressively diminishing over time as the popularity of these cigarettes in Australia has declined.C*-++Ԍ X  Ù CONCLUSION: KEY ISSUES IN MEASURING CONSUMPTION  X  * XConsumption data are useful as an indicator of the overall severity of the tobacco epidemic (as measured by the extent of tobacco use in a population). #"  X  * XFor countries without recent prevalence data, consumption data may be relatively easy to calculate from readily available sources. #"  X  * XMonitoring the tobacco epidemic through the use of consumption estimates is a relatively inexpensive procedure.#"  X  * XIn countries where it is possible to get historical data from sources such as the branch of government that monitors tobacco taxation, or tobacco sales, trends in tobacco consumption can be directly estimated. #"  Xs  * XWhen interpreting cigarette consumption data, limitations such as smuggling and the exclusion of "roll your own" cigarettes need to be considered.#"-++  X  READING LIST FOR CHAPTER VII: 1.XJoosens L, Raw M. Smuggling and crossborder shopping of tobacco in Europe.  X British Medical Journal, 1995, 310: 13931397. #"  X 2.XNicolaidesBourman, Wald, N, Forey, B & Lee,P (Eds). 1993. International  Xz Smoking Statistics. Oxford, Oxford University Press.#"  XN 3.XUnited Nations, Statistical Yearbook, ( various years), New York.