Put the words "genetics" and "public health" together and the word "eugenics" will almost certainly not be far behind. This jaundiced view of what "public health genetics" means is a sad legacy of the inhumane and scientifically misguided eugenics programs of countries including Nazi Germany, Sweden, the United States, and China. Public health has always been about the prevention of disease, and the assumption has been that once this is allied with genetics, it will inevitably mean intervention by the state to kill, sterilize, or prevent the birth of individuals who are regarded as genetically unfit.

Certainly the history of the 20th century teaches us that we must constantly be on guard against the misuse of genetics by the state. Recent laws in China show that the specter of eugenics has not disappeared. Some would even argue that allowing individual couples the choice of aborting a baby with a genetic disorder amounts to eugenics. However, many people, at least in Western countries, draw a distinction between the decisions of individuals and families threatened by genetic disease, and state interference in the area of human reproduction. Most would see antenatal screening for Down's syndrome, for example, as morally justifiable provided individuals can choose whether they wish to take part and can decide without coercion what action (if any) to take on the basis of the test result. Care and treatment must also be freely available to those born with genetic diseases.

Reproductive decisions made on the basis of genetic information are a matter for individuals and their clinicians and are not the province of public health, except insofar as public health authorities may be involved in ensuring that any prenatal genetic tests that are made available are properly used and evaluated. The diseases that individuals and couples may wish to avoid by this means, such as cystic fibrosis or beta thalassemia, are entirely, or almost entirely, genetic in origin and cannot be avoided by any other means. They are also relatively rare. In contrast, all the major public health scourges that threaten the health and well-being of populations in the developed world - cancer, coronary heart disease, and diabetes, for example - have both genetic and environmental causes. The proper interface between public health and genetics lies in working toward an understanding of how genes and environment act together to cause these diseases and how the environment, rather than genes, might be manipulated to help prevent or delay the onset of disease.

That goal is, in reality, still some distance in the future. Although we know about many of the individual genetic and environmental factors that are instrumental in causing hypertension, for example, we still understand very little about how they interact in individuals to cause disease. "Environment," in the context of gene-environment interactions, can mean not only such factors as diet, lifestyle, or air and water quality, but also medical interventions such as drug treatments. It seems likely that the first successes in applying an understanding of gene-environment interactions may come in the area of individual drug therapy. It's often said that of a group of people given a specific drug to treat a medical condition, 30 percent will improve, 30 percent will see no change, 30 percent will discontinue treatment, and 10 percent will suffer some sort of adverse reaction, possibly life-threatening. The causes of at least some of these different responses to the drug are likely to be genetic.

Drug companies are trying to identify the genetic variants that are responsible for these differences, hoping eventually to develop genetic tests that can be used to match each individual to appropriate drugs. The question of whether these tests will be adopted by, for example, the National Health Service in the UK will be very much an issue for public health scrutiny. Public health physicians and policy makers will have to consider the costs of testing in relation to factors such as the burden of ill health caused by the adverse drug reaction, the costs of mis-prescribed drugs, and the sensitivity and specificity of the test(s) that are available.

In the meantime, many workers in the field of public health genetics see an important part of their role as protecting the public against ill-advised genetic testing and screening. It has been suggested, for example, that population-level screening should be introduced for the disease hereditary hemochromatosis, a disease of iron overload that can lead to severe ill health, usually beginning in late middle age. Hemochromatosis seems, at first sight, to have all the hallmarks of an ideal candidate for population-level screening: the genetic basis of the disease is known, the susceptible genotype is quite common (about 1 in 250 people of white European extraction are homozygous for the major disease-causing mutation), suitable tests are available (either DNA based or involving measurements of iron overload in the blood), and there is a cheap and effective treatment (regular phlebotomy).

But there are still some unanswered questions. For example, the disease appears to be incompletely penetrant - that is, an appreciable percentage of people who are homozygous for the major disease-causing mutation either have no symptoms or have symptoms that are apparently so mild that they are not aware of ill health. Population-level screening therefore runs the risk of labelling as "ill" - and incurring the costs of monitoring and treating - a group of people who might otherwise live out their lives in reasonably good health. The burdens of possible adverse psychological reactions, insurance problems, and implications for other family members have not been assessed. Public health geneticists have so far argued successfully that, although eventually the balance sheet may tip in favor of population screening, we do not as yet have all the necessary information to embark upon such a course.

We do not yet know how large the impact of new genetic knowledge on medical practice will be, or when it will begin to be felt in earnest. The role of public health genetics must be to counteract the "hype" while at the same time making sure that suitable policies and services are in place so that genetic advances can be incorporated into mainstream medical practice with proper evaluation and when the time is right. As these preparations are made, it is also vital that all those associated with the provision of health care - policy makers and managers as well as clinicians - understand enough about genetics to enable them to assess critically the contribution of genetic knowledge to the prevention and treatment of disease.