In 1999, the American geneticist Francis Collins predicted that within 10 years doctors would routinely genetically screen healthy individuals to identify those with genes that increased their susceptibility to develop common diseases, such as cancers and heart disease.

Those at higher-than-average genetic risk would be advised to change their behaviour (e.g. exercise, eat a healthier diet) or be offered drugs or other medical treatment to reduce their risk of developing these diseases. Collins (recently appointed director of the National Institutes of Health) predicted that this would transform medical practice and dramatically reduce chronic disease.

This prediction seemed plausible at the time because over the preceding decades 1500 disorders had been identified in which a mutation in a single gene predicted a high risk of developing a serious disease. But genetic testing for these mutations has had a limited impact in practice because there are few effective interventions to offer for persons at risk of developing these diseases and collectively they account for less than 5% of the disease burden in developed societies.

The major causes of disease burden in developed countries like Australia are heart disease, cancers, major depression, arthritis, asthma, and diabetes. These are multifactorial disorders in which environmental risk factors and genetic susceptibility contribute to the risk of developing them.

Family and twin studies indicate that there is a substantial genetic contribution to the risk of developing these diseases. Large genome wide association studies, in which whole genomes can be searched for candidate gene variants associated with diseases, and meta-analyses combining data from these studies have found associations between certain alleles or chromosomal regions and the risk of developing common diseases.

But, with a few exceptions, most susceptibility alleles have only weakly predicted disease risk: persons with most of these susceptibility genes have a 10% to 30% higher risk than those without.

Combining information on multiple individual genetic variants could potentially improve predictions of disease risk. But modelling has shown that in the case of type 2 diabetes, coronary heart disease and cancer information from multiple susceptibility genes does not predict risk better than risk factors such as age, sex, BMI, smoking, and blood glucose.

Even if genetic prediction is able to do better than this in the future, it is unclear whether such information will provide a stronger motivation for behaviour change (e.g. stop smoking tobacco) than advice based on risk behaviour or family history. It is also unclear whether genetic risk information will produce sustained changes in behaviour.

Some critics fear that genetic risk information may promote fatalism in individuals less confident about their ability to change their behaviour. Others are concerned that genetic screening might lead to anxiety about disease risk and unnecessary interventions in those who are at low risk of developing disorders.

Public health professionals are concerned that “high risk” strategies based on predictive genomics will displace effective population-based strategies that aim to shift the population distribution of risk by reducing, for example, the prevalence of cigarette smoking, per capita alcohol consumption or average blood pressure.

It is more sensible policy to reduce cigarette smoking by high taxation on tobacco products and restrictions on cigarette advertising than to spend resources on identifying those at increased genetic risk of nicotine dependence or tobacco-related diseases, if they smoke tobacco.

Genomic medicine illustrates the common trajectory of many new biotechnologies: enthusiastic basic researchers overstate their likely impact and foreshorten the time frame within which their benefits will be delivered by failing to take account of the substantial research and development required to translate them into routine use.

This is an edited version of a speech due to be delivered tonight at the University of Sydney by Wayne Hall, NHMRC Australia Fellow and Professor of Public Health Policy, School of Population Health, University of Queensland.