The history of the cellulose acetate filter

Most available historical information is USA oriented, including key documents from the University of California San Francisco Truth Tobacco Industry Documents Library (https://www.industrydocuments.ucsf.edu/tobacco/). Filters were first used in the 1860s to keep loose tobacco out of smokers’ mouths.11 In the 1930s and 1940s, they were marketed to protect smokers from ‘poisons’, such as nicotine.12 Although they were a popular novelty at first, filtered cigarettes did not affect unfiltered cigarette sales in the early 20th century.12 By the mid-1950s, carcinogens such as arsenic, 3,4-benzpyrene and radioactive polonium were identified in cigarettes,13 implicating them as a contributor to the reported increase in lung cancer incidence.14 Internal tobacco industry documents reveal that their research units also identified carcinogens in tobacco and tobacco smoke,15–17 but the industry did not publicly report or acknowledge these findings. As tobacco companies focused on ways to eliminate carcinogens from their product, the research on filters increased in intensity.12 Implicit and explicit product marketing focused on the presumed ‘safety’ of filtered cigarettes.12 However, according to a 1958 internal company memo, Philip Morris scientists recognised that selective filtration of harmful compounds was ‘a thermodynamic impossibility’.18 As for the cellulose acetate filter, industry researchers evaluated this in 1932 and determined that there was very little difference in nicotine content delivered between regular and filtered cigarettes.19 There were also other problems with the filter, such as fibres disbursing and being inhaled by smokers.20 21

By the 1950s, cellulose acetate was the most commonly used filter material. In cooperation with tobacco companies, major chemical manufacturers (Hoechst Celanese and Tennessee Eastman) provided these filters.22 Cellulose acetate production for filters increased from 3 million tons in 1953 to 22 million tons in 1955. In 1957, a congressional committee investigation addressed ‘false and misleading’ advertisements by tobacco companies on the implied benefits of cigarette filters and concluded that the tobacco companies deceived the public regarding the safety of their products.12

The industry’s overall shift to filtered cigarettes continued into the 1960s because of two important historical events. First, in 1962, the UK’s Royal College of Physicians published Smoking and Health, highlighting the link between smoking and lung cancer and other diseases.23 Second, the US Surgeon General’s Advisory Committee on Smoking and Health published the landmark Report on the Health Consequences of Smoking in 1964, concluding that cigarette smoking is a cause of lung cancer and laryngeal cancer and the most important cause of chronic bronchitis.24 These reports resulted in enormous press attention and likely encouraged tobacco industry marketing to emphasise the value of filtered cigarettes.

Almost all commercial cigarette filters are now made of cellulose acetate fibres along with paper and plasticisers. Some filters include activated charcoal, which may remove some gas-phase chemicals but not particulates or carbon monoxide.25 26 According to the 2020 Federal Trade Commission (FTC) Cigarette Report, the US market share for filtered cigarettes across all major manufacturers was 99.8%.6

In Robert Proctor’s historical volume, Golden Holocaust, a chapter entitled ‘Filter Flimflam’ summarised the three reasons why filters are now part of almost all commercial cigarettes. These are: (1) to lower the cost of manufacturing (cellulose acetate is cheaper than tobacco leaf); (2) to keep tobacco bits from entering the mouths of smokers; and (3) to convince people into thinking that filtered brands were somehow ‘safer’ than unfiltered brands.12

Environmental concerns regarding cigarette filters

Cigarette butts, mainly the cellulose acetate filter, have been the most commonly picked up item on International Coastal Cleanup, held worldwide each September, for almost all of the last 30 years.8 In 2020, nearly a million were collected, but this was far less than the more than 5 million picked up globally in 2019 (likely a result of the COVID-19 pandemic, with widespread reduction in group activities, non-essential travel and socialisation).8 Clean-up activities call attention to the problem of tobacco product waste, but because of the ubiquity of this waste source, they do not represent a valid surveillance system for quantifying tobacco waste or an intervention to substantially reduce it; they collect only a small proportion of the discarded tobacco waste products.

The negative impacts of cigarette filters on ecosystems and the organisms inhabiting them are now a growing field of research (table 1).27 Recent reports mainly involve laboratory studies with microorganisms as well as with larger aquatic and terrestrial organisms. There is increasing concern about environmental microplastic contamination in general, including that derived from discarded cigarette filters.28 A cigarette filter has 12 000–15 000 cellulose acetate strands, and if discarded into aquatic or terrestrial environments, these fibres can disperse into ecosystems. Belzagui et al29 modelled this process in a laboratory setting and estimated that a typical filter releases approximately 100 microfibres per day, most of which are less than 0.2 mm in size. They estimated that roughly 0.3 million tons of cellulose acetate filters are disposed of annually worldwide. The released microfibres may harm small aquatic organisms. In fact, microplastics have been found in fish and shellfish that may become part of the human food chain. Although the human health risks of aquatic microfibre pollution are uncertain, it is clear that humans are exposed to them and to the toxins that adhere to them through aquatic contamination.30 Toxic chemicals, pharmaceuticals and microorganisms may sorb onto microplastics disposed into the environment, providing potential risks to human health.31

Approximately 800 chemical constituents were detected in one laboratory study of fresh and saltwater cigarette butt leachates;32 nicotine was the most abundant, which is a hazardous chemical previously used as a pesticide. Also found were diacetin and triacetin, which are plasticisers used in filter fabrication and attachment. Thirty-eight compounds found in the saltwater leachates were also identified in laboratory-exposed mussels.32 These leachates appeared to produce positive in vitro responses in these organisms for genotoxicity (increased activation of the aryl hydrocarbon receptor (a transcription factor that regulates gene expression)) and cytotoxicity (on the oestrogen receptor-p53 loop). In another leachate study, some leached nicotinic alkaloids were found to bioaccumulate in exposed rainbow trout.33 These and other laboratory studies suggest a potential risk to human health through consumption of cigarette butt-exposed biota.34

Novotny et al35 reviewed human and animal poison centre data for reports of accidental ingestion of tobacco products, including filters. The authors found that cigarette butt consumption by small children was a health concern likely due to indiscriminate eating behaviours and modelling adult behaviours. Veterinary reports of nicotine poisoning suggest that domestic animals have consumed butts and suffered serious gastrointestinal, central nervous system and cardiovascular effects.

Health concerns regarding use of filtered cigarettes

In 2001, the US National Cancer Institute (NCI) Monograph 1336 asserted that changes in machine-measured tar and nicotine yields in cigarette smoke due to filtration or other design changes (with the so-called ‘FTC Method’37) did not reduce smokers’ actual exposure to tobacco toxicants. Chapter 6 (on ‘Cancer’) in the 2014 US Surgeon General’s Report14 extensively reviewed the way changes in cigarette design have changed smoking-attributable lung cancer patterns. The population risks for lung cancer associated with smoking have increased over time, and for a particularly aggressive cell type of lung cancer (adenocarcinoma), incidence has increased. Incidence of other cell types (small cell, in particular) declined due to widespread smoking cessation. The evidence was sufficient to conclude that the increased risk of lung adenocarcinoma among smokers results from changes in the design and composition of cigarettes since the 1950s; however, the Report did not specify which changes these were.

A refinement to the filter, intended by the tobacco industry to lower the machine-measured tar and nicotine yields, is filter ventilation. This involves providing small holes in the filter that allow the dilution of the smoke when the cigarette is puffed. Because smokers need to extract sufficient nicotine to maintain their addiction, they obstruct the vents (so-called compensatory smoking), and puff more deeply, thereby obviating any benefits from reduced toxin or nicotine delivery. The addition of ventilated filters has clearly changed the pattern of smoking, including more intense puffing, and this has changed the pattern of lung cancer incidence (ie, adenocarcinoma in particular). The elasticity of compensatory smoking negates any benefit from smoking low-tar, low-nicotine yield (filtered) cigarettes.38

Despite the accumulating evidence regarding the inability of ‘filters’ to eliminate toxic tobacco chemicals and the increase in lung adenocarcinoma incidence that is likely attributable to cigarette design changes,39 there still seems to be uncertainty among the public40 and some scientists about the health value of ‘filters’. Notably, the 2014 US Surgeon General’s Report on the Health Consequences of Smoking14 and the US NCI Monograph 13 reviewed previous longitudinal epidemiological studies showing potentially reduced risks from low-tar, low-nicotine yield cigarettes and concluded that filters and changes in design did not make cigarettes safer.36 A more recent Research Letter concluded that smoking unfiltered cigarettes was ‘more harmful than smoking filtered cigarettes’.41 The researchers found that, ‘After adjustment, unfiltered cigarette smokers were nearly 40% (hazard ratio, 1.37; 95% CI, 1.10–1.17) more likely to develop lung cancer and nearly twice (hazard ratio, 1.96; 95% CI, 1.46–2.64) as likely to die of lung cancer compared with those who smoked filtered cigarettes’. However, it is important to recognise that this was a secondary analysis of data from a cohort study set up to assess the efficacy of tomographic (radiologic) screening in detecting lung cancer among high-risk smokers and not to assess the population-based risks of smoking unfiltered cigarettes.42 The limitations of that Research Letter are: (1) the population included in the study was extremely high risk (men and women aged 55–74 with >30 pack-year history or quit within the last 15 years); (2) there was limited adjustment for socioeconomic status; and (3) the reported prevalence of unfiltered cigarette use in the study population was 11.4% (less than 1% of cigarette sales in the USA are of unfiltered cigarettes). Hence, that recent study mainly suggests there are multiple confounding factors that determine lung cancer mortality among those at the highest risk for lung cancer.

Conducting a study to measure the health effects of filtered versus unfiltered cigarettes would involve a clinical trial comparing exposures and disease incidence among those randomly selected to smoke filtered and unfiltered cigarettes. The logistical and ethical challenges for such a study would be extraordinary. To date, only a small pilot, proof-of-concept study has attempted such a controlled trial. It assessed perceptions, changes in smoking topography (inhalation and puffing patterns) and changes in exposure to nicotine and some carcinogens comparing filtered and unfiltered cigarette smoking.43 Preliminary data from this trial suggest that committed smokers, when switched to unfiltered cigarettes, smoke fewer cigarettes per day and experience less satisfaction from their smoking.44 They did not differ with respect to urinary cotinine (the main metabolite of nicotine) or selected carcinogen exposure (Eyal Oren, personal communication, principal investigator).

Challenges in addressing the cellulose acetate filter as a health and environmental hazard

Many smokers and non-smokers still believe that filtered cigarettes are safer than unfiltered cigarettes, and most do not know that almost all commercial cigarettes have plastic filters. Using a population-based sample of 2979 adult non-smokers, former smokers and current smokers, Patel et al. (2021) studied knowledge and beliefs around cigarette filters. The authors evaluated how these factors might inform support for policies aimed at reducing the environmental impact of discarded plastic filters.40 Only about a quarter of the participants (28.9%) thought that cigarette filters contained plastic. Stratified by smoking status, 33.2% of smokers compared with 21.3% of non-smokers believed that filters reduce the harmful effects of smoking (p<0.001). Epperson et al sought to assess knowledge, attitudes and beliefs about the environmental impact of filters among a sample of young adults.45 Most respondents (89%) agreed that filters are harmful to the environment and not biodegradable, but only 43% knew that filters are made of plastic. Using data from a 2019 representative household survey of the German population aged 14 years and over, Kotz and Kastaun reported that the majority of both smokers and non-smokers did not know that cigarette filters were made of synthetic materials.46

Based on the history and anatomy of the filter, it now seems appropriate to consider defining this additive differently. According to Google’s Oxford Languages online dictionary (https://languages.oup.com/google-dictionary-en/), the definition of ‘filter’ is ‘a porous device for removing impurities or solid particles from a liquid or gas passed through it’. Given this specification for how filters should function, it may be better to consider cigarette filters as product additives. It is clear that although filters may change the machine-smoked measures of nicotine and other toxic chemicals as well as reduce some of the particulates produced by combusted tobacco, they have not prevented the severe human harms due to smoking. If these additives had effectively functioned as ‘filters’ (ie, removing impurities or all solid particles from cigarette smoke), there would be evidence that the risks for smoking-attributable diseases have declined since filtered cigarette smoking became normalised over the last 50 years. This is clearly not the case.47 The filter ultimately has become nothing more than a fraudulent marketing tool, designed specifically to mislead smokers and young initiators that they are doing something to reduce their risks.21