Journal of Public Health

, Volume 15, Issue 4, pp 255–264

Health risks in tobacco farm workers—a review of the literature


    • Research Association Public Health Saxony, Medizinische Fakultät Carl Gustav CarusTechnische Universität Dresden
  • Jochen Schmitt
    • Department of Dermatology, Unit of Dermato-Epidemiology and Health Services Research, University Hospital Carl Gustav CarusTechnische Universität Dresden
  • Dimitris J. Kouimintzis
    • Department of Hygiene and Epidemiology, School of MedicineNational and Kapodistrian University of Athens
  • Wilhelm Kirch
    • Research Association Public Health Saxony, Medizinische Fakultät Carl Gustav CarusTechnische Universität Dresden
Review Article

DOI: 10.1007/s10389-007-0122-4

Cite this article as:
Schmitt, N.M., Schmitt, J., Kouimintzis, D.J. et al. J Public Health (2007) 15: 255. doi:10.1007/s10389-007-0122-4



To systematically describe and summarize the specific occupational health impact of tobacco farm working.

Subjects and methods

Qualitative systematic review applying a standardized electronic literature search strategy (MEDLINE, TOXNET, EMBASE, AGRICOLA, Science Citation Index; until January 2007) and hand search within eligible articles. Double assessment of eligibility and study results. Inclusion of all studies on specific work-related health risks of tobacco farm workers, e.g. risk factors for and frequency of green tobacco sickness (GTS) and other specific health risks, as well as intervention studies.


We identified 31 relevant publications. Concerning GTS substantial heterogeneity was observed in terms of the case definition and frequency of disease occurrence. Seasonal prevalence ranged between 8 and 89%. Incidence was reported to be 1.9 cases per 100 person days. In the two studies included on primary prevention of GTS hand washing and the use of protective clothes were effective. Studies conducted on cancer as well as on urticaria and contact eczema in tobacco harvesters are scarce.


Health problems due to transdermal nicotine absorption are frequent among tobacco harvesters. GTS as well as chronic health effects in this population need to be further investigated. The toxicity to the cardiovascular system and carcinogenicity of chronic dermal nicotine exposure are likely to exist as non-smoking tobacco harvesters show similar cotinine and nicotine levels compared to active smokers in the general population. Effective intervention strategies addressing the specific needs of the tobacco harvesters in different regions around the world must be developed, evaluated and further improved. It is important to educate tobacco farm workers, farmers and health care providers on GTS and other health risks.


Farm workerGreen tobacco sicknessHealth riskOccupational diseaseTobacco


Tobacco production continues around the world and is increasing especially in developing countries. Consequently, the characteristics of tobacco farm workers are changing and their occupational health risks are of great relevance to public health.

Health risks associated with tobacco farming were identified as early as 1713 by Bernardino Ramazzini who is considered by many as the father of occupational medicine. He recorded various symptoms in Italian tobacco workers, such as headaches and stomach ailments, attributing them to exposure to tobacco dust (Eckholm 1978). In 1970, green tobacco sickness (GTS)—a disease specific to tobacco farm workers—was first described in Florida (Weizenecker and Deal 1970).

Tobacco farming is becoming more and more commercialized. Today, maintaining a viable income from tobacco requires a large production base with hired personnel working in tobacco almost exclusively during the harvesting season (8–12 weeks). These workers are more extensively exposed to the risk of tobacco farming than the former, typical farming families with allotments of a few acres only. In the USA, nowadays most of the labour force is comprised of foreign-born, low-paid Hispanics. In Europe, family labour still strongly prevails with about 80% of the total labour force employed in the sector (Commission 2003). In the future, there will be a significant shift in the world tobacco supply towards developing countries in which the health effects of tobacco farm workers have only been studied rarely so far (McBride et al. 1998; Quandt et al. 2000).

Tobacco farming around the world

Tobacco (Nicotiana tabacum) is grown in more than 100 countries around the world (FAO 2006). The top six tobacco producers, which account for two-thirds of worldwide tobacco production, are China (35% of worldwide production), Brazil (13%), India (10%), USA (6%), Malawi and Indonesia (both 2%). There are about 33 million tobacco farm workers around the world (USDA 2004).

European countries produce nearly 5.4% of the total world production. In the EU 25, Greece and Italy account for more than half of the total area harvested. About 2.4% of workers in the EU agricultural sector are employed on tobacco farms, 25% of whom are women (Commission 2003; FAO 2006).

Different types of tobacco grown

Flue-cured tobacco is grown largely in North and South Carolina, in Virginia and Georgia in the USA and in Europe. Burley tobacco, which contains more than 10% more nicotine, is mainly grown in Kentucky and Tennessee in the USA (McBride et al. 1998). In comparison to burley tobacco, which is harvested as a whole stalk, the harvest of flue-cured tobacco presents a greater opportunity for exposure to wet tobacco leaves as ripe tobacco leaves are picked in stages. The whole body is engulfed in the plant—and potentially exposed to nicotine—while the harvester breaks off tobacco leaves. Leaves are mostly gathered under the worker’s arms until no more can be carried. Hand harvesting remains the preferred method as tobacco leaves mature at different rates, machinery is expensive and leaves picked mechanically are of lower quality (Quandt et al. 2000). Shade tobacco leaves, which are grown e.g. in Connecticut, USA, are used as wrappers for premium cigars and are therefore harvested more carefully and mostly under dry conditions—limiting workers’ exposure to nicotine (Trapé-Cardoso et al. 2005). Air-, fire- and flue-cured tobacco originally grown in Virginia, USA, is referred to as Virginia tobacco (bright leaf). Virginia and non-Virginia tobacco have morphological differences. Matured leaves of non-Virginia tobacco contain about threefold to fourfold more nicotine compared to Virginia tobacco leaves (Gosh et al. 1986).

Other types of tobacco include light air-cured tobacco (high quality, for cigarettes), dark air-cured and sun-cured tobacco (low quality, dark cigarettes), Basmas, Katerini and Kulak (oriental types, for aroma enrichment of cigarettes) as well as fire-cured tobacco (cigars and Toscani).

Description of green tobacco sickness (GTS)

Green tobacco sickness is frequently defined as a disease characterized by vomiting or nausea and dizziness or headache during or after exposure to the agent Nicotiana tabacum in tobacco leaves. However, GTS may also result in severe conditions such as dehydration and consequently in the need for emergency medical care (Arcury et al. 2001a). The different aspects of GTS are shown in Table 1. In the USA, the economic impact of GTS seems to be limited (outpatient treatment US $250, hospital admission US $566, intensive care US $2041). However, decreased productivity and lost income due to GTS may be significantly important to tobacco farmers and their farm workers (McBride et al. 1998).
Table 1

Summary of characteristics of GTS: exposure, signs and symptoms, natural history, possible treatment and prevention strategies

Aspect of disease



Skin contact (hands, forearms, axilla, feet, thighs, back) with dissolved nicotine from (wet) tobacco leaves

Signs and symptoms

Common: nausea, vomiting, dizziness, headache

Others: weakness abdominal cramps, shortness of breath, pallor, diarrhoea, chills, fluctuations in blood pressure and heart rate, increased perspiration and salivation

Natural history

Latency: 3–17 h after exposure (median 10 h)

Duration: 2.4 days mean (<12 h–72 h), self-limiting


Rest, rehydration, symptomatic treatment (antiemetics, dimenhydrinate, diphenhydramine, meclizine)


Water-resistant clothing (gloves, boots, rain suit), avoidance of working in wet conditions (early morning, rain), change out of wet clothes and wash clothes soaked with tobacco sap, prophylactic use of dimenhydrinate and diphenhydramine (Ives 1983)

Figure 1 presents the pathway from dermal exposure to tobacco leaves via dermal absorption of nicotine—both mediated by several factors—to plasma nicotine levels and green tobacco sickness. The rate of transdermal absorption is not only determined by the amount of skin exposure to tobacco leaves but also by working in wet tobacco, with pre-existing cuts and rashes and after alcohol consumption. The axilla, where harvesters typically hold the leaves, is particularly susceptible to chemical absorption. High plasma nicotine levels in smokers have been reported to decrease dermal absorption of nicotine (Arcury et al. 2001a).
Fig. 1

Pathway of exposure to tobacco leaves to green tobacco sickness (GTS) (adapted from Arcury et al. 2001a; Quandt et al. 2000)

The main differential diagnoses of GTS are pesticide poisoning and heat exhaustion. Pesticide poisoning occurs earlier in the season than GTS as the last application of pesticides usually occurs several weeks before harvest during which workers are exposed to tobacco leaves (McBride et al. 1998). In contrast to GTS, patients suffering from heat exhaustion show fever, but no increased secretions. GTS also occurs on rainy days on which heat exhaustion is less likely (Gehlbach et al. 1974; Swinker and Meredith 2000).

In 1998, McBride et al. (1998) summarized the evidence on GTS. However, this review exclusively focused on GTS and excluded other health risks of tobacco farm workers. Additionally, several studies on GTS have been published since then.

This systematic literature review aims to give an update on health risks associated with tobacco farming, particularly, but not exclusively, focusing on the exposure to tobacco leaves during harvest.


We systematically reviewed all studies on specific work-related health risks of tobacco farm workers. Studies on the relative importance of risk factors, studies on the frequency of disease occurrence (e.g. incidence, prevalence) as well as intervention studies were considered eligible. We excluded studies on work-related health risks that are not specific to tobacco farm workers, such as accidents, injuries and musculoskeletal disorders.

Literature search and eligibility criteria

A standardized electronic literature search was performed using MEDLINE and TOXNET (both from inception until January 2007). We used combinations of keywords related to tobacco farming (tobacco AND agriculture* OR grow* OR harvest* OR farm), work-related conditions (environment*, work*, occupation*) and exposure/risk factors (risk, exposure, hazard). We limited the literature search to journal articles presenting original data with an English abstract.

We additionally searched the database of the DIRERAF project (Development of Public Health Indicators for Reporting Environmental/Occupational Risks Related to Agriculture and Fishery), which includes relevant articles on the topic from EMBASE, AGRICOLA and Science Citation Index.

Hand searching of the reference lists of eligible articles as well as of review articles on the topic identified one additional study.

Using this systematic search strategy we identified a total of 423 articles (Fig. 2).
Fig. 2

Identification of relevant studies for inclusion in the systematic review

The abstracts of all articles identified were reviewed for eligibility by two independent reviewers (NS, JS). If it was not clear from the abstract whether an article was eligible the full text paper was reviewed. Disagreement was resolved by discussion. Exclusion criteria comprised no original data reported, case reports/case series (n < 5), no work-related condition, outcome not specific to tobacco farm workers and exclusively laboratory data reported (e.g. urinary cotinine levels); 31 articles met the eligibility criteria and were included in the systematic review (Fig. 2).

Data extraction and qualitative review

All articles included were abstracted using a standardized data abstraction form designed for this systematic review. The form included information on the study design, sample size, geographical region, study duration, participant characteristics, exposure (e.g. type of tobacco), definition of outcome assessed (e.g. GTS) and study results. Data abstraction was solely based on the methods and the results sections. Heterogeneity was assessed for all relevant study and population characteristics. Odds ratios, incidences and prevalences were calculated from the data given in the publications if not reported.


Green tobacco sickness (GTS)

Study and participant characteristics of included investigations on GTS are summarized in Table 2. Studies published on GTS are mostly descriptive. We identified five cohort studies, six case-control studies, two intervention studies (one on hand washing and another on protective gloves) and two retrospective uncontrolled studies. The majority of studies were conducted in the USA (n = 10), four in India and one in Malaysia. The most common tobacco types were under investigation in the different studies. Sample size ranged between 12 and 1,340. In studies conducted in the USA, the study populations consisted predominantly of Latino migrants.
Table 2

Summary of study and participant characteristics, case definition and main results of studies on GTS


Study characteristics

Participant characteristics

Case definition


Study type

Sample size (n)




Quandt et al. 2000

Cohort study

n = 144

General illness after exposure

41% GTS prevalence during summer


Tobacco harvesters, mainly Mexican

Flue-cured tobacco


Arcury et al. 2001a

Cohort study

n = 182

Exposure + vomiting or nausea + dizziness or headache

0.082 prevalence


Tobacco harvesters, mainly Latino farm workers

1.88/100 person days incidence

Flue-cured tobacco

Arcury et al. 2001b (see Quandt et al. 2000)

Cohort study

n = 182 (see Arcury et al. 2001a)

Exposure + vomiting or nausea + dizziness or headache

65 cases in 44 workers, 0.242 (44/182) prevalence


Tobacco harvesters, mainly Latino farm workers

1.88/100 person days incidence

Flue-cured tobacco


Arcury et al. 2002

Case-control study

36 cases, 40 controls (farm workers in clinic, no GTS)

See Arcury 2001 + not pregnant, no fever, no exposure to pesticides

OR: 14.6 wet conditions w/ rain suit vs dry conditions


Tobacco harvesters, Mexican men

Not specified

Arcury et al. 2003a

Cohort study

n = 182 (see Arcury et al. 2001a)

Exposure + vomiting or nausea + dizziness or headache

1.9/100 person days incidence


Tobacco harvesters, mainly Latino farm workers

Flue-cured tobacco

Gosh et al. 1979

Case-control study

197 tobacco workers vs controls (institute staff, number not specified)

Neurological and/or respiratory complaints

0.8883 prevalence (175/197) presumably seasonal


Tobacco harvesters, population from district, smokers

Non-flue-cured Virginia tobacco

Gosh et al. 1986

Case-control study

289 tobacco workers, 150 controls (non-tobacco workers in tobacco farm)

Weakness or giddiness or abdominal pain or nausea and vomiting or increased sweating or shivering or increased salivation

60.6% cases/0% controls


Tobacco harvesters, both sexes

Virginia tobacco

Gosh et al. 1987

Intervention study

n = 85

Nausea, vomiting, dizziness, prostration, weakness, headache, etc.

82% reduction in GTS (75% stop, rubber gloves), 82% reduction (61% stop, cotton gloves)


Tobacco harvesters, both sexes

Non-flue-cured Virginia tobacco

Ballard et al. 1995

Retrospective chart review/case-control study

n = 47 review (tobacco workers from hospital and OHNACa)/40 cases (80% follow-up), 83 controls (non-ill tobacco workers)

Diagnosis consistent with GTS, characteristic symptoms, exposure at time of illness

OR infinite for risk factor working in wet clothes


Tobacco harvesters, both sexes

Burley tobacco

Curwin et al. 2005

Intervention study

n = 12

Nicotine level on hand

96% reduction of nicotine on hands after hand washing


Tobacco harvesters, Hispanic men

Flue-cured tobacco

Gehlbach et al. 1974

Case-control study

53 cases, 49 controls (tobacco harvesters without GTS)

Characteristic symptoms + recurrence if reexposure + self-limiting illness + pesticide exposure not responsible

OR GTS non-smoker vs smoker: 17.2; most frequent symptoms (> 80%): nausea, vomiting, pallor, weakness, dizziness, headache


Tobacco harvesters, mainly White, African American, Indian

Flue-cured tobacco

Onuki et al. 2003

Retrospective uncontrolled study

80 cases, 40 controls (officers)

Exposure + runny eyes, blurred vision, increased salivation, nausea, stomach cramps, bloating, cough, shortness of breath, nervousness, unusual tiredness, confusion, dizziness, headache, skin rash, numbness, muscle weakness

Runny eyes and blurred vision significantly more often if urinary cotinine = 50 ng/ml per m2


Tobacco harvesters, men, registered farmers

Significant differences in urinary cotinine cases vs controls only in non-smokers

Not specified

Parikh et al. 2005

Retrospective case-control study

685 cases, 655 controls (farm workers not in tobacco)

Exposure + one symptom: nausea, vomiting, giddiness, headache, weakness, loss of appetite

47% prevalence (presumably per season) of GTS


Tobacco harvesters, both sexes


No chronic effects (blood pressure, ECG, eye disease, reproduction)

Non-flue-cured Virginia tobacco


Trapé-Cardoso et al. 2003

Retrospective chart review

450 visits, 331 tobacco workers

(1) Exposure + vomiting or nausea + dizziness or headaches

4-15% of visits due to GTS


Tobacco harvesters, men, mostly migrants

(2) ICD-9 diagnosis consistent with GTS

Shade tobacco

Trapé-Cardoso et al. 2005

Case cohort study

53 cases, 59 controls (nurses) (follow-up: 55% cases, 74% controls)

Exposure + vomiting or nausea + dizziness or headaches + increase in salivary cotinine

No GTS, no significant increase in salivary cotinine in harvest season


Tobacco workers (only 46% picking tobacco)

Shade tobacco

Cases: migrant men, controls: 24% women

aOHNAC Occupational Health Nurses in Agricultural Communities programme (surveillance system)

Substantial heterogeneity was observed in terms of the case definition of GTS (see Table 2). While Quandt et al. (2000) defined GTS as any general illness after exposure to tobacco leaves and Gosh et al. (1979) as any neurological and/or respiratory complaints in tobacco farm workers, other used more specific definitions (e.g. Arcury et al. 2002; Gehlbach et al. 1974; Trapé-Cardoso et al. 2005). The most frequently applied case definition of GTS was vomiting or nausea and dizziness or headaches during or after exposure. The sensitivity and specificity of the case definitions have to be acknowledged when interpreting the results of the individual studies. There is a wide range of prevalence of GTS in the literature: 8–89% per season. The highest reported prevalence may be due to the unspecific case definition (Gosh et al. 1979). Incidence was reported to be about 2 cases per 100 person days. Two studies found very high relative odds of GTS when working in wet conditions (Arcury et al. 2002 OR: 14.6; Ballard et al. 1995 OR: infinite). A case-control study conducted in North Carolina revealed an odds ratio of GTS of 17.2 in tobacco farm workers who were non-smokers compared to those who were smokers (Gehlbach et al. 1974).

Both intervention studies identified a significant, clinically relevant reduction in GTS and nicotine levels on farm workers’ hands. Rubber gloves seem to be more effective in the prevention of GTS compared to cotton gloves (elimination of symptoms in 75 and 62% of workers, respectively) (Curwin et al. 2005; Gosh et al. 1987).

The studies conducted with shade tobacco harvesters revealed lower risks of GTS, but there were limitations in the study designs. In contrast to other types of tobacco, shade tobacco is harvested more carefully, skin exposure to the tobacco plant is much more limited and harvest under wet conditions is avoided because of the decreased quality of the picked leaves (Trapé-Cardoso et al. 2003; Trapé-Cardoso et al. 2005).

Health beliefs concerning GTS

There are inconsistencies in health beliefs related to GTS in tobacco farmers, farm workers and research-based knowledge of the disease, at least in the USA. Some years ago, both farmers and farm workers lacked a specific name for the illness. Whereas 50% of the farmers correctly suspect nicotine to cause GTS, the other half of them attributed GTS to the posture during harvest, the heat, the smell of the tobacco plants or a combination of these factors. Farm workers mostly believed that the chemicals applied to the plants (pesticides, growth regulators, etc.) cause GTS. Farmers seem to be convinced that tobacco harvesters develop a tolerance and GTS is not prevalent any more after being used to working in tobacco. Farmers also underestimate the duration of illness. Farm workers, on the other hand, are aware of the fact that tolerance concerning GTS is not developed necessarily: so far, there is no evidence that workers susceptible to GTS build up a tolerance. Farmers’ discounting of the seriousness of GTS allows them to also discount their responsibilities in terms of provision of prevention strategies for their employees. Both farmers and farm workers believe in the protective effects of tobacco use and work experience. Farmers even think that smoking is most effective in the prevention of GTS and suggest advising the mostly non-smoking Latino farm workers to start smoking tobacco (Arcury et al. 2003b; Rao et al. 2002).

Other threats to tobacco harvesters’ health


Literature focusing on the association between exposure to nicotine during tobacco farming activities and malignancies is scarce. A meta-analysis of case-control studies on female tobacco farm workers in Europe revealed a significantly increased risk (OR: 3.1, 95% confidence interval: 1.1-9.3) of bladder cancer in this population. The researchers adjusted for age and smoking status but not for pesticide exposure, which also has been linked to increased cancer risk (Mannetje et al. 1999).

Urticaria and contact eczema

Ueda et al. (1987) reported a prevalence of 45% of allergic or irritant skin disorders in tobacco farmers. Contact eczema was observed more frequently than urticaria. From the data presented it is not clear whether the reported symptoms were caused by pesticides, tobacco leaves, other work-related or non-occupational exposures. In another descriptive analysis, three of seven tobacco farm workers treated for allergic skin rashes and allergic-toxic dermatitis had positive skin tests. Delayed skin reaction (eczema) may result both from handling tobacco leaves and pesticide use. Urticaria may be triggered by direct skin contact with tobacco leaves, e.g. planting and drying tobacco leaves (Nakamura 1984; Szarmachz and Poniecka 1973).

Although the absolute magnitude of the problem is not clear from the evidence, published data suggest that tobacco farm workers are at risk for occupational allergic and irritative dermatological symptoms.

Pesticide exposure

Only one study (Curwin et al. 2003) investigated the health effects of the exposure to pesticides during tobacco harvest. Most publications on the association between pesticide exposure and farm workers’ health have been conducted in pesticide sprayers who are differently exposed to pesticides compared to tobacco harvesters. Nevertheless, it seems like a substantial amount of pesticides (e.g. organophosphate) is transferred to the hands of tobacco farm workers during harvest. It has also been shown that hand washing significantly removes pesticides from the skin (reduction by 23–96% depending on washing method, solvents, time interval between exposure and decontamination). In order to prevent health effects of exposure to pesticides it is critically important to make protective equipment and hand washing facilities available to all tobacco farm workers (Curwin et al. 2003). Pesticide sprayers have an increased risk of suffering from transient neurological and psychological conditions explained by the toxicity to the peripheral and central nervous system (Kimura et al. 2005; Salvi et al. 2003). Pesticides used in tobacco plantations are not specific to tobacco and are therefore not regarded as an issue of this literature review.


Despite a number of observational studies focusing on GTS, literature on the health effects of tobacco harvesters is limited to descriptive rather than analytical publications.

GTS constitutes a significant public health problem. It is frequent even when a specific case definition is applied. An incidence of 2 cases in 100 person days exposed and an OR of 17 for non-smokers compared to smokers leads to the conclusion that particularly non-smoking tobacco harvesters face a daily health risk of intoxication due to transdermal absorption of nicotine. However, smoking is a major public health threat and therefore by no means a solution in the prevention of GTS.

Protective equipment has been shown to decrease the magnitude of GTS significantly. However, in order to be accepted, protective suits and gloves should be lightweight and comfortable while working in hot climate. Providing on-site hand washing facilities and allowing workers time to change if their clothes are sodden are also key elements in primary prevention of GTS/nicotine absorption in tobacco harvesters (Gehlbach et al. 1975, 1979; Rao et al. 2002).

The majority of studies were conducted in the USA whereas most, and the percentage is increasing, tobacco harvesters work in developing countries. The magnitude of the major health risk in tobacco harvesting—GTS—as well as of long-term effects of chronic dermal tobacco exposure (e.g. cardiovascular diseases, cancer) in harvesters needs to be verified and updated around the world in order to encourage and persuade policy makers to intervene.

Strengths and weaknesses of the existing evidence and implications for future research

One limitation of the current evidence on GTS is a possibly imprecise estimate of its true prevalence and incidence. The reason for that lies in the fact that the numbers reported in the individual studies depend on the case definitions applied and the health beliefs, e.g. the awareness of stakeholders that the condition GTS exists at all. Only efforts to raise awareness of GTS on the part of tobacco harvesters and health care providers may help to fully recognize the magnitude of GTS. In Kentucky, for example, no cases of GTS had been reported to public health agencies before 1992, the year in which a surveillance system for agricultural workers and farmers was established. The number of reports increased between 1992 and 1993, indicating the success of surveillance and awareness raising activities (Ballard et al. 1995; CDC 1993).

Additionally, the self-limiting nature of the illness and low health insurance coverage among migrant and seasonal farm workers may explain the difference between results based on cases reported to surveillance programmes or hospital chart reviews in contrast to observational cohort studies on tobacco harvesters.

There are no established criteria for the diagnosis of GTS. Exposure to nicotine from tobacco leaves—a prerequisite for the diagnosis of GTS—should be verified in non-smokers in blood, saliva (cotinine only) or urine tests for nicotine and cotinine (McBride et al. 1998). In addition, a validated questionnaire is needed, for which the case definition of Arcury and colleagues (2002) with additional consideration of pregnancy, fever and pesticide exposure appears to be a good basis for an expert consensus.

Despite the fact that GTS poses an important public health problem, large and high quality cohort studies with adjustment for potential confounders (e.g. smoking, weather conditions, benefits of adequate clothing and protective equipment) have not been published. The effects of handling wet tobacco leaves on long-term morbidity and mortality also still have to be investigated. No effects of chronic toxicity have been identified so far. Pregnant women who are chronically exposed to tobacco during harvest may be at increased risk of preterm delivery and stillbirth but do not show an increase in abortions. In animal studies there is some evidence of teratogenic effects of exposure to tobacco leaves, but prospective studies of pregnant women harvesting tobacco or retrospective studies of children with congenital defects are necessary before drawing any conclusion (Parikh et al. 2005; Priester and Mulvihill 1975).

The cumulative seasonal exposure to nicotine has been projected to be equivalent to smoking at least 180 cigarettes. Although neither tar, carbon monoxide nor other combustion products identified in cigarette smoke are absorbed in the harvesting process, the toxicity to the cardiovascular system and carcinogenicity of the chronic nicotine exposure are likely to exist and should therefore be further investigated (D’Alessandro et al. 2001). Researchers focusing on long-term effects of active and passive smoking should pay additional attention to and put increased efforts into the investigation of long-term health effects of transdermally absorbed nicotine.

The association between chronic exposure to nicotine in tobacco farming and increased risks of bladder cancer, lung cancer and other cancers linked to passive and active smoking needs to be further analysed. Table 3 summarizes cotinine and nicotine levels to which smokers and non-smokers among farm workers and the general population are typically exposed. Professional tobacco harvesters seem to be at lower risk of nicotine absorption compared to inexperienced, seasonal minority workers (D’Alessandro et al. 2001).
Table 3

Cotinine and nicotine levels in tobacco farm workers compared to the general population in smokers and non-smokers (salivary cotinine, urinary cotinine and nicotine)


Saliva cotinine (ng/ml) (mean; range)

Urine cotinine (μg/ml in 24-h urine) (mean)

Urine nicotine (μg/ml in 24-h urine) (mean)

Tobacco farm workers

Active smoker

145; 104–179a

GTS, no gloves: 3.74

GTS, no gloves: 5.35

GTS, gloves: 1.96c

GTS, gloves: 2.54c



46; 5–90a

Without GTS: 1.43

without GTS: 1.61

With GTS: 3.71d

with GTS: 4.15d

GTS, no gloves: 2.71

GTS, no gloves: 3.14

GTS, gloves: 0.90c

GTS, gloves: 1.62c

General population

Active smoker

100; 60–140b



Cut-off smoker/non-smoker: 7–13b


1.8; NAa




2.4; NAb


aData from Quandt et al. 2000

bData from Etter et al. 2000

cData from Gosh et al. 1987

dData from Gosh et al. 1986

eData from Gosh et al. 1979

fData from Behera et al. 2003

Other aspects of tobacco farming

Our review does not focus on musculoskeletal disorders, injuries and respiratory illness in tobacco workers. The first are not specific to tobacco farming and the latter is a condition occurring in workers involved in curing and stacking tobacco rather than harvesting tobacco. Injuries include accidents with farm machinery, cuts and piercings, as well as repetitive movement joint injuries. Musculoskeletal disorders may result from the workers’ posture during harvest (Browning et al. 1998). Exogenous allergic alveolitis (synonym: hypersensitivity pneumonitis) is a type of parenchymal lung disease which is labelled “tobacco worker’s lung” as it may be caused by inhalation of tobacco moulds (Nefedov et al. 1991; Olade and Lessnau 2006). Tobacco farm workers seem to be at increased risk of suffering from disorders of the upper airways (e.g. nasal dysfunction). In contrast, the association between tobacco processing activities and chronic diseases of the lower respiratory systems could not be proven yet, although non-significant reductions in lung function have been detected (Chloros et al. 2004; Osim et al. 1998).

Harvesting tobacco does not only have direct adverse effects on health leading to decreased productivity and lost income. By harming the environment (e.g. deforestation), growing tobacco also affects human health indirectly (WHO 2004). The fact that personal and family involvement in tobacco farm work increases the likelihood of tobacco use is another argument in favour of restricting tobacco cultivation around the world (Noland 1990).

Implications for public health policy

Occupational safety programmes should cover tobacco farm workers around the world and should not neglect the working conditions of low-income workers in tobacco. GTS seems to be a highly preventable disease which mostly affects vulnerable populations such as children and migrant and/or seasonal Hispanic men in the USA (McKnight and Spiller 2005).

Cost-effectiveness studies of primary prevention strategies need to be conducted. Positive results may push farmers to advocate for better protection of their employees.

International organizations as well as national and regional agencies must improve occupational health information and communication concerning the health risks related to working on tobacco farms. Health care providers as well as farmers and farm workers themselves need to be adequately educated about GTS and other occupational health risks before and during the harvesting season. Modules on GTS may e.g. be added to courses on pesticide safety training which are already attended by most tobacco workers in some countries. It is critically important that education is culturally appropriate and accessible to all tobacco farm workers. In the USA, where the majority of employed tobacco harvesters today are Latino farm workers, educational materials should e.g. include low literacy booklets, photonovels, comics and videotapes (Arcury et al. 2003b). On the other hand, awareness raising activities concerning GTS should not lead to increased misdiagnoses in cases of pesticide poisoning.


The DIRERAF study was supported by partial financing from the EU Health and Consumer Protection Directorate and the Stavros Niarchos Foundation.

Conflict of interest statement

We declare that we have no conflict of interest.

Copyright information

© Springer-Verlag 2007