Bittersweet harvest – herbicides and farmers’ health in Fiji

Sugar cane farmers in Fiji use high levels of herbicides, but remain unaware of the exposure hazards. Philip Szmedra explains the results of a recent study showing that farmers using pesticides have higher medical bills and poor health adversely affecting farm production. Farmers lack the information and training to reduce their dependence on costly and potentially harmful weed control strategies.

Fijian children in village close by sugar farms. Photo Philip Szmedra

Sugar farming in Fiji is strictly a smallholder, labour intensive enterprise. Much of the land area on the two principal islands in the Fiji group is mountainous, the eroded remnants of volcanic activity that created the islands. Sugarcane farming occurs along the narrow coastal plain regions of the generally dry northern and western portions of Viti Levu and the northern coast of Vanua Levu. The soils are extremely rich in organic matter and with sufficient rainfall and appropriate levels of management expertise are capable of producing in the range of 16 tonnes of cane per hectare(1).
    Approximately 40% of sugarcane farmers in Fiji rely on herbicides for weed control in their fields while using very few insecticides or fungicides(2). This fact probably lessens the health impacts of pesticide use, as the most acutely toxic pesticides are generally the latter. However, a number of the herbicides commonly used in Fiji sugarcane are classified by the World Health Organization (WHO) as toxicity level II (moderately hazardous) or III (slightly hazardous) and so would be expected to negatively affect the health of those exposed either directly or indirectly. Included among these materials are paraquat (Class II), 2,4-D (Class II) and diuron (Class III)(3).
    Farmers in Fiji and many other developing countries are generally unaware of the actual short term or long term exposure hazards associated with many pesticide products in common use in the production of sugarcane, aside from recognition of the very obvious dangers of taking pesticides internally, or of accidental splashes on the skin or in the eyes. The acute and chronic hazards therefore do not enter as criteria into the farmer’s decision of whether to use pesticides rather than alternative pest management methods. 
    Small farm size, relatively low income levels, as well as relatively low levels of investment in human capital in the rural areas of the country has led to a situation in which technologies such as chemical pesticides, which require some minimal level of knowledge that goes beyond traditional agricultural practices, may not be used most efficiently nor in a safe manner. This situation may lead to negative health impacts which in turn hinder the productivity of the farmer, farm worker and farm family members leading to overall diminished farm productivity. Rural dwellers affected by pesticide exposure could experience reduced work capacity in the field and a reduction in the ability to make rational farm management decisions. Farmers that do not recognize health impairment as one of the chronic effects of pesticide exposure will not balance the effectiveness of the pesticides being used on their farm against the risks involved. By not taking account of the costs of pesticide use, they will therefore tend to use these materials in excess – in economic terms this is known as usage beyond the private or socially optimal amount.

Surveying sugar farmers
There had been no work done in Fiji investigating the health impacts of pesticide use on the rural population in sugar-producing regions. As a member of the economics faculty at the University of the South Pacific (USP) in Suva between 1996 and 2000, I had the logistical support of the Fiji Sugar Corporation (FSC), the parastatal that controls the production, milling and export of sugar in the country, and the financial support of USP research funding to investigate the problem, or indeed whether a problem existed.
    Seven student enumerators and myself conducted a personal interview survey over a period of three days in December 1998 shortly after the completion of the sugarcane-harvesting season. The FSC identified 150 sugar farmers whose farms were located in the vicinity of the three FSC sugar mills situated on Viti Levu, Fiji’s principal island and largest producer of sugarcane. Farmers were interviewed during individual farm visits as well as at FSC field offices where many farmers had travelled to receive governmental drought assistance payments from FSC officers. The 1998 crop year in Fiji was one of severe El Nino-related drought in which the production of sugarcane decreased dramatically from average historical levels. As a consequence the government of Fiji provided farmers with cash subsidies to act as both income support and to assist in establishing the next season’s sugarcane crop. 
    The survey required farmers to recall from memory details of their use of pesticides. While formal records are a more reliable method of determining pesticide use patterns and practices, many sugarcane farmers in Fiji do not keep written records of their use of pesticides(4). Each student enumerator was fluent in both Hindi and English. Our survey contained 97% Indian farmers although the actual racial split in the sugarcane farm population is approximately 76% Indian and 24% indigenous Fijian(5). The racial split in the general population of Fiji is approximately 44% Indian and 52% indigenous Fijian, with other Pacific Islander and Chinese groups comprising the remaining 4%. 
    We also asked farmers to recall the medical costs incurred by them and their families as a result of pesticide exposure as well as average medical expenses incurred for all types of illnesses over the past five years. Of the 150 potential farmer contacts, 145 farmers were interviewed. The control group consisted of 31 farmers who did not use pesticides. These farmers were asked to recall production information, specifics of their medical histories, as well as family medical expenditures over the past five years. We classified the two groups of farmers as ‘users’ and ‘non-users’. The control group numbers were limited by the difficulty of finding farmers who were not using pesticides, and all farmers and controls were men, reflecting the actual situation.

Sugarcane cultivation in the rolling hills of Western Viti Levu. Photos Philip Szmedra

Results of the survey
The average user worked a farm whose land area comprised about nine hectares, about twice the typical Fiji sugar farm size. His family had been working this land for 36 years and he had been farming sugar for about 29 years. The non-user farm was significantly smaller comprising on average about 3.5 hectares. This may be a possible contributing factor in the farmer decision not to use pesticides: smaller land area and probably the associated constrained financial ability to purchase pesticide products. Average output was also correspondingly smaller on the non-user farm: about 24 tonnes per hectare for the users and 20 tonnes per hectare for the average non-user farm.

Weed infestation in sugarcane 
The tropical climate and rich soils of Fiji are conducive to the growth of generally all tropical plant species. Large weeds found in Fiji sugarcane fields such as guinea grass (Panicum maximum), Johnson grass (Sorghum halepense), or itchgrass (Rottboellia cochinchinensis) compete with cane for light and reduce the number of tillers and stalks that the cane plant produces. Unchecked weed growth in Fiji cane fields has been demonstrated to decrease cane production by between 20-50%(6).
    The Fiji sugar farmer uses an array of herbicides. Sampled farmers reported using 20 different commercially available products and spraying on average twice per season. The average cost of herbicides in the production year was F$515 (US$230). Four different compounds were most popularly used (table 1). The chlorophenoxy herbicide 2,4-D in two different formulations, 40% active ingredient (a.i.) (Weedkiller E40) and 80% a.i. (Weedkiller E80), was reported applied by 89% of the sampled farmers. These two formulations are selective and systemic herbicides applied mainly in post-emergence applications for the control of broadleaf weeds(7). 

    2,4-D is the most popular chemical pesticide by volume worldwide(8). It is also the oldest having been developed during WWII by American scientists. Its intended purpose was as a defoliant against Japanese food crops in preparation of a planned allied invasion of the Japanese home islands in an attempt to end the war in the Pacific. While not employed for this, 2,4-D was commercialised at the end of WWII and quickly adopted by farmers as a method to decrease labour input in farming. 2,4-D demonstrated the potential market for manufactured pesticide products. It paved the way for the intensification of research and development by both public agencies and private chemical firms into expanding the variety of chemical pesticide products available to agriculture.
    In recent years 2,4-D and the chlorophenoxy herbicides group have been implicated as causal factors in the development of certain cancers in agricultural workers in Northern Europe and the US(9,10,11). In addition, 2,4-D has been identified as having medium to high acute toxicity ratings as established by the WHO(12). Other studies have linked 2,4-D to a wide range of long term toxic effects including heart, liver, and kidney damage, and central nervous system disorders(13).
    Diuron 90 (diuron 90% a.i.), and Karmex (diuron 80% a.i.) were used by 71% of the survey respondents. Diuron is a residual herbicide used for both pre- and post-emergence applications. The herbicide has been classified as having low to medium acute toxicity while exhibiting carcinogenic, mutagenic, and growth inhibiting long-term effects(14). 
    Gramaxone (paraquat 20% a.i.) was used by 43% of respondents. Paraquat is a non-selective compound used to control grasses and broadleaf weeds and is one of the more toxic materials used for weed control. The acute toxicity ratings of formulations range from high to very high while chronic effects cover the entire spectrum of mammalian susceptibility to pesticidal compounds including mutagenicity, neurotoxicity and all major organ system damage, and may be a possible contributor to the onset of Parkinson’s disease(15). In addition, paraquat is often used in rural Fiji as a suicide agent, as it is in many other developing countries.

Above Siga toka street market, where vegetables are often treated with leftover sugarcane herbicides. Photo: Philip Szmedra

Assessing health effects
We asked users initially their attitude with regard to the use of pesticides in their production system and then to evaluate their health status from three different perspectives: (a) an overall subjective evaluation of their personal health; (b) health problems they had recognizably experienced from pesticides through either accidental or habitual exposure; and (c) long term health problems not necessarily associated with pesticide exposure. Most farmers (79%) expressed concern about the health impacts of using pesticides but in general thought the benefits outweighed the risks. Only 7% of farmers had no concern about the impact of pesticide use on their health. Despite this generally positive view of using pesticides significant numbers of surveyed farmers reported suffering acute episodes of what could be termed ‘pesticide poisoning’, or having experienced an adverse reaction to pesticide exposure. Approximately 10% of users took no precautions when applying or handling herbicides, and 85% of farmers polled used multiple protection devices such as gloves, overalls, and boots.
    Average medical expenses varied significantly between the two groups. The user group reported average yearly medical expenditures for drugs, doctor, and hospital visits of F$821 (US$370) over the previous five years. The non-user group’s average medical expenditures were F$328 (US$148). These costs are a heavy strain on the farm family’s income in a nation with a per capita GDP of US$2300 in 2001. The average annual medical expenditures from illness attributed to pesticide exposure over the past five years was F$36 (US$16).
    The following describes medical disorders associated with the use of the more popular herbicides in Fiji sugar cane production and reports the percentage of farmers in both groups experiencing chronic problems with specific organ systems.

Eye problems Herbicides such as 2,4-D and paraquat are known eye irritants(16). Chronic irritation can lead to a number of eye problems including cataracts and the development of a vascular membrane over the eye (pterygium) leading to decreased vision: 49% of user-farmers reported chronic eye problems compared with 13% of the non-users.

Dermal problems The main mode of entry into the body of pesticide materials is through the skin. Contamination occurs typically when mixing materials in preparation for application as well as during the application process. Approximately 16% of users reported skin problems, skin lesions or rash, which they traced directly to pesticide exposure; 30% described themselves as suffering from chronic skin problems compared to 10% of non-users.

Respiratory problems The lungs are an important site of impact of many pesticidal compounds. Typical complaints that could originate or be aggravated by pesticide exposure include asthma, chronic cough, excess sputum, and wheezing: 20% of users reported respiratory problems linked directly to pesticide exposure. Smoking cigarettes also contributes to respiratory problems. Approximately 37% of users smoked cigarettes regularly; 40% of users surveyed complained of long-term problems with their respiratory tracts; 39% of non-users smoked cigarettes though just 19% reported chronic respiratory problems indicating a possible synergy between cigarette smoke and pesticides leading to a higher incidence of chronic illness in the user group.

Neurological problems Many pesticide compounds disrupt neurological transmission. Most insecticides rely on this mode of action to be effective in controlling target insect pests. However, many herbicides also have neurotoxic characteristics as an adjunct to their principal action pathways for control of weeds. 2,4-D and paraquat exhibit neurotoxic effects(17,18). Sixteen per cent of users complained of tingling in fingers directly related to pesticide exposure; 5% complained of desensitisation in their extremities. Chronic neurological problems were reported by 26% of surveyed users but only 13% of the non-user group.

Gastro-intestinal (GI) problems The gastro-intestinal tract is another important site for the metabolism of many pesticides. Exposure to 2,4-D, paraquat and dichlorprop can cause the GI problems – nausea, vomiting, diarrhoea, pain and bleeding(19): 10% of the users reported nausea related to pesticide exposure while 32% reported chronic GI problems. Of the non-users, 36% reported similar problems. Diet may be a significant contributing factor in both groups in this instance. The predilection for ‘hot’ spices by the Indian population as a condiment and cooking ingredient makes the Indian more susceptible to GI distress. Many survey respondents in both groups blamed diet for their gastric problems. 

When compared with the control group the users had significantly greater incidence of organ disease. The user group’s exposure and proximity to pesticide materials may be an important factor in contributing to these higher rates of chronic illness and morbidity. Certainly there exist other contributory factors to ill health in the rural population besides pesticide exposure: diet; access to health care facilities for acute care needs and/or health maintenance; a hereditary predisposition to chronic illness; other social or environmental factors. However, the nature of the specific pesticide materials in common use in sugarcane production strongly suggests the culpability of pesticides in these higher rates of illness and disease in the user group.

Conclusion 
Our survey results provide sufficient evidence of the probability that the health of sugar farmers in Fiji is negatively affected by types of exposure to suggest that action should be taken. The levels of impairment have potential to negatively affect farm production. Reducing pesticide use may decrease overall sugarcane productivity but the improvement in short and long-term health levels in the rural population may offset these production losses from a social welfare perspective. 
    Insufficient resources are devoted to the development of efficient feasible alternatives to chemical pesticide products for sugarcane production in Fiji. Policies aimed at educating the farm population about both the acute and chronic effects of pesticide exposure are necessary to rationalize the use of pesticide products in sugarcane. Educational support could include better handling, disposal and protective measures that could be used to mitigate the effects of pesticide exposure. But a more effective way to decrease acute and chronic risks involved in pesticide use would be to devote resources toward research and implementation of alternative weed management measures. This approach could be paired with a demonstration to users of the negative health impacts and subsequent physical and financial costs involved in developing the dependency that characterises chemical intensive agricultural practices.

References
1. FSC, Sugar Cane Research Centre Annual Report 1997-98, Lautoka, Fiji, 1998.
2. Maharaj, G., CEO, Sugarcane Growers Council, Fiji, personal communication, December 1998.
3. WHO 1999 www.who.int/whosis/.
4. FSC, personal communication, December 1998.
5. Sugarcane Growers Council, Fiji Cane Grower, July/August 1998.
6. Sugar Technical Advisory Mission, Weed Control in Sugarcane Field, Tech Bulletin 5, Mission of the Republic of China, 1996.
7. Post-emergence application implies treating for weed infestations after germination; pre-emergence implies applying a herbicide after sowing but before the crop emerges from the ground.
8. US Department of Agriculture, National Agricultural Pesticide Impact Assessment Program, Biologic and Economic Assessment of Benefits from the Use of Phenoxy Herbicides in the US. Special NAPIAP report 1-PA-96, November 1996.
9. Hardell, L., Sandstrom, A. Case-control study: soft tissue sarcomas and exposure to phenoxyacetic acids and chlorophenols. British Journal of Cancer, Vol 39: 711-17, 1979.
10. Hoar, S.K., Blair, A., Holmes, F.F., et al, Agriculture herbicide use and risk of lymphoma and soft tissue sarcoma, Journal of the American Medical Association, Vol 256: 1141-7, 1986.
11. Hardell, L., Eriksson, M., The association between soft tissue sarcomas and exposure to phenoxyacetic acids, Cancer, Vol 62: 652-6, 1988.
12. World Health Organisation, Recommended classification of pesticide by hazard, WHO/PCS/01.4 (2000-01).
13. Sjöden, P.O., Söderberg, V., Phenoxyacetic acids: sublethal effects. In C. Ramel (ed) Chlorinated phenoxy acids and their dioxins, Royal Swedish Academy of Sciences, Ecological Bulletin 27, 1978.
14. Council on Scientific Affairs, Cancer Risk of Pesticides in Agricultural Workers, Journal of the American Medical Association, Vol. 260: 959-66, 1988.
15. Bocchett, A., Corsini, G.U., Parkinson’s Disease and Pesticides, Lancet II: 1163, 1986.
16. UK Pesticide Guide 2002, British Crop Protection Council, CABI, Wallingford, UK.
17. Extoxnet, 2,4-D, Pesticide Information Network, Cornell University, US, 1993.
18. Briggs, S. Basic Guide to Pesticides. Their Characteristics and Hazards. Taylor & Francis. Washington, D.C.1992 p133.
19. Hurst, P., Hay, A., and Dudley, N. The Pesticide Handbook. Journeyman, London, 1991. pp 100-01.

Philip Szmedra is an Assistant Professor, Department of Economics, Georgia Southwestern State University, Americus, GA 31709 USA, pszmedra@canes.gsw.edu. 

[This article first appeared in Pesticides News No. 55, March 2002, pages 12-14]