Intended for healthcare professionals


How did UK cigarette makers reduce tar to 10 mg or less?

BMJ 2006; 332 doi: (Published 02 February 2006) Cite this as: BMJ 2006;332:302
  1. Richard J O'Connor, affiliate research scientist (Richard.Oconnor{at},
  2. K Michael Cummings, chair,
  3. Gary A Giovino, senior research scientist,
  4. Ann McNeill, honorary senior research fellow,
  5. Lynn T Kozlowski, chair
  1. Department of Health Behavior, Roswell Park Cancer Institute, Buffalo, NY 14263, USA
  2. Department of Health Behavior, Roswell Park Cancer Institute, Buffalo, NY 14263, USA
  3. Department of Epidemiology and Public Health, University College London, London WC1E 6BT
  4. Department of Biobehavioral Health, Pennsylvania State University, University Park, PA 16802, USA

    EDITOR—To try to reduce the harm caused by cigarette smoking, the European Commission established maximal values for tar (10 mg), nicotine (1 mg), and carbon monoxide (CO; 10 mg) per cigarette, as measured by the International Organization for Standardization (ISO) method, from 1 January 2004.1 The easiest way to reduce yields is by increasing filter ventilation,2 but this allows smokers easily to control the dose of smoke they can obtain, usually to facilitate increased intake from lower yield cigarettes.2 3 We compared yields and design features of 10 cigarette brands sold in the United Kingdom before and after the EC standard was implemented.

    Researchers at Roswell Park Cancer Institute performed analyses in September-October 2005. The 1999 brand versions had been stored unopened at room temperature since purchase; current versions were purchased in September 2005.

    Full details of methods and data are available on the web ( Filter ventilation was assessed using a KC-3 digital apparatus (Borgwaldt-KC, Richmond, VA, USA) following a published protocol.4 ISO tar, nicotine, and CO yields were obtained from packs (1999 CO values are from LGC5). Wilcoxon tests assessed average changes across brands.

    The table shows yields and filter ventilation for each brand. Originally rated at 11-13 mg tar (median 12 mg), all brands dropped to 10 mg tar (17% drop, P < 0.002), while reducing nicotine from a median of 1.0 mg to 0.9 mg (P < 0.008). The ratio of tar to nicotine did not change (P > 0.45). Carbon monoxide yields also dropped significantly, from a median of 13 mg to 10 mg (P < 0.01). Median ventilation increased by 479% (P < 0.006) from 1999 to 2005. None of the other design features measured showed consistent changes.

    Changes in ISO yields and ventilation for selected “full flavour” brands, United Kingdom, 1999-2005

    View this table:

    Our findings indicate that manufacturers complied with the EC's recent mandated yield reduction primarily by increasing filter ventilation rates on cigarettes—a design feature that promotes compensatory smoking.2 The current “10-1-10” standard is therefore unlikely to reduce smoke exposure for smokers.3 The EC, while recognising the compensation problem, has said that it will not revise the standard until solid evidence shows that better methods exist.1

    Our data suggest that removing erroneous yields from packs and adopting alternative approaches to reducing the harmfulness of cigarettes, such as banning filter vents, seems warranted. However well intentioned the EC's effort to make cigarettes less harmful, focusing solely on the use of maximum yields has served to promote increased levels of filter ventilation, which is both ineffective and misleading.


    • Contributors Susan Anderson provided the cigarettes for testing. Tammy Vance performed the cigarette measurements.

    • Funding This work was performed under a Transdisciplinary Tobacco Use Research Center grant to Roswell Park Cancer Institute from the US National Institutes of Health (1 P50 CA111236).

    • Competing interests KMC and LTK have provided expert testimony in court cases against the tobacco industry. KMC received travel expenses for speaking at a tobacco litigation seminar. RJO, AM, and GAG have no competing interests to declare.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.