5. Ecological and human impacts

Ecological Impacts

Although the ecological impacts of chemicals are complex, some effects are well-documented. The effects on various animals, birds (Campbell and Cooke, 1997) and fish (Cameron & Berg, 1994; Stebbing et al., 1992), include birth defects, cancers, and damage to nervous, reproductive and immune systems (see Box 4). For example, dichlorodiphenyl trichloroethane (DDT) was implicated in the early 1970s as the cause of reproductive failure in eagles and other birds, due to the thinning of egg shells. Since then, a number of other cases involving wildlife have been studied, including large fish kills and declines in sea mammal populations. Contamination of fish by mercury, PCBs and other toxic chemicals appears to be increasing in the USA (NRDC, 1998) and evidence about the effects of low- level but possibly wide-spread contamination of fish is increasing in Europe (Matthiessen, 1998; Tyler, 1998). Recent results from the UK, for example, suggest that the incidence of feminisation and other sexual disruption in fish "is higher than previously thought and is associated with discharges from sewage treatment works" (EA, 1998).

The complexity of ecosystems, such as the North Sea, makes risk assessment very difficult without extensive multi-disciplinary research and integrated assessments (MacGarvin, 1994; Neal, et al., 1998).

Human Impacts

Clear scientific evidence for many impacts of manufactured chemicals on human health (except for some occupational exposures) is also complex and difficult to identify. This is partly because people are exposed to many different substances and their breakdown products via indoor and outdoor pollution from several pathways, including air, water, food and passage through the skin. Since the 1970s, there has been increasing concern in particular about consumer goods, including food, which can be one of the main routes of exposure to chemicals for many people. Major stationary and mobile sources of exposure, such as factory chimneys, may now account for less than 25% of total exposure, according to US estimates (Wallace, 1993).

Another problem in identifying risks from chemicals arises from the need to account for the effects of other causal agents such as smoking, radiation, and natural toxins, which can also cause ill-health or ecological damage, either separately, or sometimes in combination with manufactured chemicals (EEA, 1998b). Furthermore, there are usually large gaps in time between an exposure to a chemical, the observation of possible ill effects, and a medical or scientific assessment about association and causation (Box 3).

Box 3 Association and causality

It is sometimes fairly easy to show that a measure of ill-health (e.g. the number of admissions to hospital per day) is associated with a possible cause, such as the day-to-day variation in levels of air pollutants. However, to show that a causal relationship exists is more difficult. A number of guidelines or tests have been developed to help assess this. These include identifying whether there is a "dose-response relationship" between the proposed causal factor and the effect, whether the sequence of events makes sense (i.e. the cause always precedes the effect), checking the consistency of results between dif-ferent studies, and the way in which the results of different studies fit together (coherence).

Proof of causality is often very difficult but, by the application of these and other criteria, an expert judgement as to whether an association is likely to be causal can often be made. Where effects are likely to be serious and/or irreversible, then a low level of proof, as in the "precautionary principle", may be sufficient to justify actions to remove or reduce the probable causes

causes (WHO & EEA, 1997).

Health problems, such as cancer or allergies, are difficult to under-stand when they involve several, often inter-dependent causes, of which "chemical cocktails" may only be a part. The level, and burden of proof of harm, are also critical issues in risk assessment (Gee, 1995; Bro-Rasmussen, 1997). People at risk and other stakeholders including consumer and trade union groups need to be involved in risk assessments (NRC, 1996; Consumers’ Association, 1997).

Box 4: Some examples of ecological impacts and possible causes

The association/causation is assessed on the scale: 1 = no observed association, 2 = suspected association, 3 = weak association, 4 = clear association, 5 = significant association.

Observation/impact Sensitive species Substance Association/
Large scale
Eggshell thinning guillemot, eagle, osprey, peregrine falcon DDT 5
Reproduction seal, otter PCB 4
Skeleton malformation grey seal DDT, PCB 2
Pathological changes seal PCB, DDT, metabolites 3
Reproduction mink PCB 5
Reproduction osprey DDT, PCB 4-5
Reproduction eagle DDT, PCB 2-3
Reproduction salmon chlorinated substances 2
Large scale - pulp and paper industry
Induction of metabolising enzymes perch chlorinated/unchlorinated organic mixture/ Dioxin compound 3
Local/regional - pulp and paper industry
Induction of metabolising enzymes perch chlorinated/unchlorinated organic mixture/ Dioxin compounds 3-4
Spine malformations four-horned sculpin chlorinated/unchlorinated organic mixture 3-4
Local, forest industry
Induction of metabolising enzymes   perch chlorinated/unchlorinated organic mixture Dioxin compounds 4-5
Spine malformations four-horned fish chlorinated/unchlorinated organic mixture 4-5
Larvae damages sea mussel chlorinated/unchlorinated organic mixture 3

Source Swedish EPA, 1993


Despite the difficulties of identifying and assessing potential risks, there is evidence about the health effects of manufactured chemicals in humans, including cancer, cardiovascular and respiratory diseases, allergies and hypersensitivity, reproductive disorders, and diseases of the central and peripheral nervous systems. These potential health impacts and some of their possible causes are summarised in Box 5.

Box 5 Some health effects of chemicals

This is a summary of the main health effects of chemicals. The link with chemicals varies from well-known causal relationships such as benzene and leukaemia, to suggestive associations, such as chemical sensitivity and pesticides. Most harmful effects are the result of many causes acting together, such as genetics, lifestyle, radiation, diet, pharmaceuticals, chemicals (manufactured and natural), smoking and air pollution, including indoor and outdoor exposures. It is also important to consider sensitive groups, such as the elderly, children, the embryo, the sick, and pregnant women, who may be affected at much lower doses than others.

Health effect Sensitive group Some associated chemicals*
Cancer All asbestos
Polycyclic aromatic hydrocarbons (PAHs)
some metals
some pesticides
several hundred animal carcinogens
some solvents

natural toxins

Cardiovascular diseases especially elderly carbon monoxide
Respiratory diseases children, especially asthmatics inhalable particles
sulphur dioxide
nitrogen dioxide
some solvents
Allergies and   hypersensitivities all, especially children particles
Reproduction adults of reproductive age polychlorinated biphenyls (PCBs)
Developmental foetuses, children lead
other endocrine disruptors
Nervous system disorders foetuses, children PCBs
methyl mercury
organic solvents

* Examples only
EEA, based on Swedish EPA
(1996); WHO (1995);
EHP (1997b);
Ashford, (1998);
Williams (1997) and
Kilburn (1998)

Some chemicals clearly cause cancers in some exposed groups, but the role of chemicals in overall cancer causation is unclear and disputed (Doll, 1992; Epstein, 1998). Any excess cancer mortality from a chemical pollutant is likely to be restricted to a section of the population, so mortality rates for entire populations can often be weak and insensitive indicators of environmental health effects from pollution. Low levels of exposure to chemicals, including pesticides, may suppress the immune response defences of the body, leaving people more susceptible to diseases from viruses, parasites, bacteria and tumours (WRI, 1996).

The potentially hazardous effects of pharmaceutical chemicals, such as greater resistance to animal antibiotics and contamination of water supplies (Envirolink, 1998) are not covered further in this summary.

Chemical pollutants that may affect reproductive health and new-born children include certain metals (e.g. lead and methyl mercury), pesticides (e.g. DDT), industrial chemicals (e.g. PCBs), solvents and other substances (Foster & Rousseaux, 1995; CJPH, 1998, in press). Exposures can occur through the placenta and breast milk (Jensen, 1996; Rogan, 1996), and some may cause small abnormalities of the immune response system. However, the WHO and others conclude that the benefits of breast-feeding outweigh the risks of pollutants in breast milk (Weisglas-Kuperus et al., 1996; WHO, 1996).

Children may be particularly at risk from chemicals because of their greater biologic-al sensitivity and greater exposure to environmental pollution relative to body weight (NRC, 1993; McConnell, 1992; Bearer, 1995). Their physiological and intellectual development may be impaired by exposure to chemicals (Rodier, 1995; Rylander et al., 1995; Jacobson, 1996; -Grand Jean et al., 1997). Low-level pesticide contamination of food (infants consume eight times more food per kilogram of body weight than adults, making this a more significant exposure pathway; CICH, 1997) and of residential surfaces and toys in the UK and USA is being reported (Pesticides Trust, 1998, Gurunathan et al., 1998). Some regulatory authorities are giving special attention to the higher levels of risk to children from pollution (US EPA, 1996). For example, the Food Quality Protection Act in the USA requires the government to add an extra margin of safety to the risk assessment of chemicals that children may be exposed to.

Cancer in children in the USA is increasing (Pogoda, 1997; EHP, 1998; Rachel’s EHW, 1998), and a large-scale study of childhood leukemias and other cancers in the UK has found them to be associated with living close to industrial plants, particularly where fossil fuels were being used or processed (Knox & Gillman, 1997).

The causes of an increased incidence of testes cancer and breast cancer in humans, and of the effects on wildlife reproduction of endocrine-disrupting chemical substances that have been observed in many countries, are largely unknown. Changes in the environment, as well as in lifestyle, may be responsible (Colburn, 1993, CEC et al., 1997; EA, 1998 – see Box 6).

Box 6: The "Weybridge Report" on endocrine disruptors

The EEA has summarised the results of the Report from the European Workshop on the impact of Endocrine Disrupters on Human Health and Wildlife (CEC et al., 1997) as follows:

There is increasing evidence and concern about rising trends of reproductive ill health in wildlife and humans, and some substances have been implicated, but there are great uncertainties about the causes of reproductive ill health.

Key conclusions are:

Sufficient evidence exists that testicular cancer rates in humans are increasing.

The apparent decline in human sperm counts in some countries was likely to be genuine.

There is insufficient evidence to definitely establish a causal link between the health effects seen in humans with exposure to chemicals.

The major route of exposure to Endocrine Disrupting Substances (EDS) is usually by ingestion of food, and to a lesser extent water. It is valid for terrestrial animals, birds and mammals, including humans.

Compared with the situation in the US, there are few cases of reproductive ill-health in wildlife in the EU where the effects could be definitely associated with endocrine-disrupting substances.

However, some cases exist within the EU area where adverse endocrine effects, or reproductive toxicity, in birds and mammals coincide with high levels of anthropogenic substances, shown to have endocrine-disrupting properties in some test systems.

The considerable uncertainties and data gaps could be reduced by the recommendations on research and monitoring into exposure and effects in wildlife and humans.

Current eco-toxicological tests, studies and risk assessments are not designed to detect endocrine-disrupting activities.

Meanwhile, consideration should be given to reducing the exposure of humans and wildlife to endocrine disrupters in line with the "precautionary principle".


Source: European Environment Agency based on CEC et al., 1997

It is the widespread presence of small amounts of many chemicals which is causing increasing concern, because alone, or in combination with other agents, they may contribute to cancer, allergies (UCB, 1997), impacts on reproduction and the immune response system, and neurotoxic effects (NRC, 1992; Kilburn, 1998). The timing of exposure to some chemicals is important, particularly for endocrine-disrupting substances, where exposures during the first 30 days after conception seem to be critical. Although exposure to chemicals may be very low for most people (i.e. in parts per million or trillion of air, water or food), some chemicals at such low doses can be potent. For example, estra-diol, the body’s key oestrogen hormone, operates at concentrations of parts per trillion-equivalent to one drop of water in 660 "rail tankers" (Brekine, 1997). And some chemicals may be more harmful at lower doses than at higher doses, since, for ex-ample, damaged cells may cause more harm than dead cells, and higher doses can trigger detoxification activity which is not triggered by lower doses (Lodovic et al., 1994).

A recent and comprehensive review of the risk assessment of new and existing chemicals concluded: "At the present level of understanding we cannot adequately predict adverse effects on ecosystems, nor can we predict what part of the human population will be affected. We are only able to assess risks in a very general and simplified manner" (van Leeuwen et al., 1996).

Despite the difficulties of risk assessment, there are many government and industry policies in place which have been designed to protect people and the  environment by minimising the risks of manufactured chemicals.


Document Actions