Article Text

Changes in tar yields and cigarette design in samples of Chinese cigarettes, 2009 and 2012
  1. Liane M Schneller1,
  2. Benjamin A Zwierzchowski1,
  3. Rosalie V Caruso1,
  4. Qiang Li2,3,
  5. Jiang Yuan3,
  6. Geoffrey T Fong2,4,5,
  7. Richard J O'Connor1
  1. 1Department of Health Behavior, Roswell Park Cancer Institute, Buffalo, New York, USA
  2. 2Department of Psychology, University of Waterloo, Ontario, Canada
  3. 3Chinese Centers for Disease Control and Prevention, Beijing, People's Republic of China
  4. 4Ontario Institute for Cancer Research, Toronto, Ontario, Canada
  5. 5School of Public Health and Health Systems, University of Waterloo, Waterloo, Ontario, Canada
  1. Correspondence to Dr Richard J O'Connor, Department of Health Behavior, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263, USA; Richard.Oconnor{at}


Background China is home to the greatest number of smokers as well as the greatest number of smoking-related deaths. An active and growing market of cigarettes marketed as ‘light’ or ‘low tar’ may keep health-concerned smokers from quitting, wrongly believing that such brands are less harmful.

Objective This study sought to observe changes in cigarette design characteristics and reported tar, nicotine and carbon monoxide (TNCO) levels in a sample of cigarette brands obtained in seven Chinese cities from 2009 to 2012.

Methods Cigarettes were purchased and shipped to Roswell Park Cancer Institute, where 91 pairs of packs were selected for physical cigarette design characteristic testing and recording of TNCO values. Data analysis was conducted using SPSS, and was initially characterised using descriptive statistics, correlations and generalised estimating equations to observe changes in brand varieties over time.

Findings Reported TNCO values on packs saw mean tar, nicotine and CO levels decrease from 2009 to 2012 by 7.9%, 4.5% and 6.0%, respectively. Ventilation was the only cigarette design feature that significantly changed over time (p<0.001), with an increase of 31.7%. Significant predictors of tar and CO yield overall were ventilation and per-cigarette tobacco weight, while for nicotine tobacco moisture was also an independent predictor of yield.

Conclusions The use of ventilation to decrease TNCO emissions is misleading smokers to believe that they are smoking a ‘light/low’ tar cigarette that is healthier, and is potentially forestalling the quitting behaviours that would begin to reduce the health burden of tobacco in China, and so should be prohibited.

  • Low/Middle income country
  • Surveillance and monitoring
  • Public policy

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Low-tar cigarettes are widely promoted throughout China via advertising and cigarette packs with prominently displayed tar values.1 The State Tobacco Monopoly Administration (STMA) has also reduced maximum limits on International Organisation for Standardization (ISO) tar ratings over time, from 17 mg in 2001 to 15 mg in 2004, to 12 mg in 2011.1 This has helped to promote ‘low-tar’ cigarettes as less dangerous.2 Indeed, China lags in broad public awareness that cigarettes marketed as ‘low tar/light’ are not necessarily less hazardous.3 ,4 The Global Adult Tobacco Survey (GATS) 2010 found that 86% of respondents thought that light/low-tar cigarettes were better for one's health compared to regular cigarettes.5 Even those with more education, such as healthcare professionals and teachers, had high levels of misperception.5

Reduced tar and ‘safer’ cigarettes remain a priority for STMA. In 2009, over 3 billion Yuan (∼US$490 million) was expended on ‘low-tar’ cigarette research and development.1 However, independent research shows that smoking cigarettes with lower machine yields of tar (according to ISO machine measures) is not associated with lower levels of nicotine metabolites or total polycyclic aromatic hydrocarbons (PAHs) in Chinese smokers’ urine.2 This is consistent with a large body of literature in Western countries pointing to little benefit from low-tar cigarettes.6 Reducing tar levels is typically achieved through cigarette engineering, changing features such as filter density, tipping paper length, filter ventilation levels and tobacco density.6 Our research group has shown that UK manufacturers complied with a 2 mg reduction in the tar ceiling enacted in 2004 (from 12 to 10 mg) by increasing filter ventilation, while other facets of engineering remained essentially unchanged.7 Typically, smokers of low-yield cigarettes can compensate for lowered yields of tar and nicotine by smoking more intensively (eg, more puffs, larger puffs, blocking filter vents, smoking more cigarettes).8 Thus a disconnect exists between machine-measured yields and exposures among smokers.

Following on our earlier report on Chinese cigarette design features,9 this paper presents data on the physical design characteristics of a variety of popular cigarette brands manufactured and sold in China in 2009 and 2012, coincident with Framework Convention on Tobacco Control (FCTC) implementation in China and a reduction in the tar limit from 15 to 12 mg.10 Thus, our study design was capable of assessing which changes in cigarette design were associated with this lowering of the limits on tar ratings, comparable to our earlier work on UK cigarettes.7 We sought to explore on which brands tar, nicotine and carbon monoxide yields (as reported on packages) declined and what changes in cigarette design might best explain those changes.


The cigarettes analysed for this study were purchased at two time points (2009 and 2012) from seven cities: Beijing, Changsha, Kunming, Shanghai, Shenyang, Guangzhou and Yinchuan.11 Field workers visited three large retail stores in each city and bought packs of cigarettes until every brand family and brand variety available in each year was purchased (total n=2,052; 2009, n=907; 2012, n=1145). The packs were shipped to the Tobacco Research Laboratory at Roswell Park Cancer Institute (RPCI), where they were catalogued and stored unopened at −20°C until analysis. From this larger set we randomly selected 91 pairs of cigarettes based on Universal Product Code matching to measure tar and ventilation at 2009 and 2012. Before testing, in accordance with ISO 3402:1999, the packs were conditioned for a minimum of 48 h at 22±2.0°C and 60±2.0% relative humidity in an environmental chamber. Cigarette physical and design characteristics were assessed following previously published procedures by the same laboratory.12–14 For a given brand, five cigarettes were selected randomly from each pack after conditioning for physical analysis and the data averaged. Digital callipers were used to measure the entire cigarette length, the tobacco rod length and diameter and the filter length and diameter. Filter weight measurements were made gravimetrically using a Mettler-Toledo analytical balance. The tipping paper was removed from each filter, and its length and the presence of any vent holes were performed using a light box and a transparent ruler. The tobacco moisture and tobacco weight were analysed using a halogen moisture analyser (HR83 or HB43-S, Mettler-Toledo, Columbus, Ohio, USA). The moisture content was determined as the per cent change in weight after heating the tobacco from five cigarettes with a halogen bulb at 125°C. Filter ventilation and pressure drop were measured using a KC-3 apparatus (Borgwaldt-KC, Richmond, Virginia, USA). The level of the paper porosity was measured using the vacuum method on a PPM1000M device (Cerulean, Milton Keynes, UK). Tar, nicotine and carbon monoxide yield values were obtained from the product packages. Analyses were conducted from January to March 2013.

Data analysis was conducted using Statistical Package for the Social Sciences V.21.0 (IBM, Armonk, New York, USA). The sample was initially characterised using descriptive statistics, and correlations of design characteristics were also analysed. Changes in brand varieties over time were assessed within-participants using generalised estimating equations (GEE). Models used a normal distribution with identity link function and exchangeable working correlation matrix unless otherwise noted.


Tar and ventilation levels

We identified eight brands as having major design changes in the 2-year study period (see online supplementary Table S1). These included the addition of a two-part filter or a charcoal filter (examples in see online supplementary Figure S1A), or implementing filter patterning (see online supplementary Figure S1B). Huanghelou implemented a channelled filter design reminiscent of Barclay cigarettes, introduced in 1980 in the USA which has been shown to be particularly susceptible to blockage by the smoker in the normal course of puffing, yielding much greater tar delivery than its ISO tar rating.15

Tar, nicotine and carbon monoxide (TNCO) and cigarette design characteristics

Table 1 presents the design characteristics of Chinese cigarettes from 2009 and 2012. Between 2009 and 2012, mean ISO tar levels decreased by 7.9% (from 11.9 to 10.9 mg), nicotine levels decreased by 4.5% (from 1.1 to 1.0 mg), and CO levels decreased by 5.7% (from 12.7 to 12.0 mg). Mean ventilation levels increased substantially, by 31.7% (from 8.3% to 10.9%). Of the unventilated cigarettes in 2009, 4.4% became ventilated in 2012, while 2.2% of ventilated cigarettes in 2009 lost ventilation in 2012. Furthermore, 1.1% of charcoal-containing filtered cigarettes did not have charcoal filters in 2012, and 2.2% of non-charcoal containing filtered cigarettes became charcoal-filtered in 2012.

Table 1

Average Chinese TNCO and cigarette design characteristics by year

Correlations of cigarette design characteristics and TNCO

Correlations among cigarette design characteristics and tar, nicotine and CO can be found in online supplementary Table S2. Tar, nicotine and CO were significantly intercorrelated (all higher than 0.68). Additionally, ventilation had by far the strongest correlations with tar (−0.66), nicotine (−0.54) and CO (−0.74).

Generalised estimating equations of cigarette design characteristics by year

We next examined whether ventilation, tobacco weight, rod and filter density, tobacco moisture and pressure drop changed as a function of time (Table 2). Significant changes with time were found for only ventilation (B=2.33, S.E.=0.77, p=0.002), which increased between 2009 and 2012. If the analysis is limited to only those cigarette brands with a tar yield above 12 mg in 2009 (ie, those needing to reduce tar to comply), mean ventilation shows an increase, although not statistically significant (B=0.355; p=0.086).

Table 2

Changes in various cigarette design features from 2009 to 2012

Relationship of cigarette design features to reported TNCO yields

Stepwise GEE models with unstandardised β estimates were conducted for each of tar, nicotine and carbon monoxide. Year was the primary predictor. Ventilation was the first design feature forced into the model because of its known association with TNCO, followed by tobacco weight, filter density, overwrap and tobacco moisture, based on their bivariate associations with TNCO (Table 3). For tar as well as nicotine, predictors were ventilation (p<0.001) and per cigarette tobacco weight (p=0.020 and p=0.011, respectively). Predictors of CO were also ventilation and per cigarette tobacco weight, as well as filter density. Other measured design features did not contribute significantly to the model.

Table 3

Primary predictors of TNCO yields


O'Connor and colleagues demonstrated that the UK tobacco industry met the 2001 European Union (EU) regulatory standard of lowering yields to tar (10 mg), nicotine (1 mg) and CO (10 mg) by increasing filter ventilation.8 The results of the present study, which examined changes in cigarette design characteristics and reported tar, nicotine and carbon monoxide levels in selected Chinese cigarette brands from 2009 and 2012, demonstrate that the Chinese tobacco industry met the 2011 requirement to reduce tar limits from 15 to 12 mg in a similar fashion. Cigarette ventilation, as expected, was the most important predictor of yields, and was the only design parameter that significantly changed (a 31.7% increase) from 2009 to 2012, replicating the results found in the UK in 2004.16 ,17

In addition, a subset of brands appear to be innovating in terms of filter design, employing filters with pictograms carved in the ends, channelled filters, dual filters and carbon filters. Future research will need to monitor such innovation as the Chinese market develops to determine if such products obtain significant market share. However, given the substantial evidence that increasing filter ventilation leads to no benefits in actual exposure and uptake, the public health consequences of the lowering of tar ratings in China are trivial if present at all. But in fact the publicity and marketing associated with China's increasing focus on lowering tar limits, coupled with the lack of real health benefits, portends a significant negative impact on public health because Chinese smokers are much more likely than smokers in other countries to believe that light/low-tar cigarettes are less harmful: 71% of Chinese smokers believe that light/low-tar cigarettes are less harmful,4 which suggests that the Chinese industry's campaign of promoting such cigarettes will continue to appeal to the 300 million Chinese smokers who are likely to be increasing in their awareness of the harms of smoking and will wrongly see light/low-tar cigarettes as a way to reduce their risk.1 As a result, Chinese smokers may well respond to increasing awareness of health harms of cigarettes by switching brands with lower tar levels rather than by quitting.

As in Western countries from the 1950s through the 1980s, when cigarette smoking was definitively linked to increased disease risks, the Chinese tobacco industry has responded to increasing health concerns around smoking by promoting cigarettes that delivered less tar in measurements made by machine smoking of cigarettes using a fixed pattern of smoking.6 ,18–20 In the US, Canada and EU, research and court cases have demonstrated the fallaciousness of these products in terms of health benefit, and efforts have been made to counter market the products.6 ,21–23 However, in the Chinese case, there are structural reasons for the continued marketing of low-tar cigarettes despite a large body of evidence that they will not reduce disease burden. Importantly, there are broader political considerations intertwined in the domestic tobacco trade, such as tax revenues, which have important impacts on tobacco control efforts.24 The decision-making structure of the implementation of the FCTC remains in the hands of the Ministry of Industry, Innovation and Technology (MIIT), of which STMA/CNTC is a part. STMA has invested heavily in ‘low-tar’ cigarette research, one of their important research and development goals.

The current study is limited in several ways. While significant differences were seen between TNCO yields and cigarette ventilation over time, the ISO yields reported on the packs were not tested directly by our laboratory. So it is possible for discrepancies to exist between labelled and measured values. ISO yields are problematic themselves, as they are not representative of smoking behaviour or exposure.6 ,19 Nonetheless, the consistency of findings with the existing literature is encouraging.

The promotion of light/low-tar cigarettes in China represents a significant barrier to reducing smoking in China, the most important preventable cause of death and disease in the world's largest country. China should not only ban misleading descriptors such as low tar, light as they already have, but also ban other misleading claims equating lower tar numbers with low harm or high technology. While removing misleading descriptors is helpful, it is not likely to lead to lasting reductions in misperceptions,25 which appear to be driven in large part by filter ventilation.13 As others have recommended, China should also remove TNCO numbers from the pack, as they are irrelevant indicators of human health risk. Otherwise, continued promotion of misperceptions about low-tar cigarettes, coupled with China's limited implementation of other FCTC recommended policies on taxation, smoke-free spaces and health warning labels, will likely prolong the tobacco epidemic.

What this paper adds

  • The marketing of ‘light’ and ‘low tar’ cigarettes in China has resulted in consumer misperceptions of their harmful effects. These lower International Organization for Standardisation (ISO) tar yields can be attributed to design features, such as filter ventilation.

  • We found that across 4 years, to comply with reduced ISO tar ceilings, Chinese cigarette manufacturers increased ventilation to reduce tar levels.

  • Misleading descriptors and ISO yields should be removed from the pack to help address consumer misperceptions.


The authors thank the China CDC staff and fieldworkers in each city for their assistance with data collection.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors RJO and GTF conceived the study. LMS, BAZ and RVC led data analysis. QL and JY contributed to data collection. All authors contributed to data interpretation and manuscript preparation.

  • Funding The ITC China Project was supported by grants from the US National Cancer Institute (R01 CA125116 and P01 CA138389), the Roswell Park Transdisciplinary Tobacco Use Research Center (P50 CA111236), the Robert Wood Johnson Foundation (045734); the Canadian Institutes of Health Research (57897, 79551 and 115016), and the Chinese Center for Disease Control and Prevention. Additional support was provided to Geoffrey T Fong from a Senior Investigator Award from the Ontario Institute for Cancer Research and a Prevention Scientist Award from the Canadian Cancer Society Research Institute.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.