Nutritional epidemiology is notoriously complex and the results difficult to interpret. The scientific literature is replete with contradictory and unconfirmed findings. The lay literature is even more confusing. Coronary heart disease (CHD) epidemiology began seriously over fifty years ago, and from the beginning the pattern of diet was identified as a key risk factor operating through blood lipid levels. Although the original diet heart hypothesis is now considered simplistic, the field has developed enormously.

Despite the confusion generated by different results from different study designs, a consensus has emerged around the main nutritional determinants of the modern CHD epidemics.1 Evidence from a variety of scientific disciplines confirms that diets with a high unsaturated to saturated fat ratio, and high in whole grains, fruits and vegetables – a typical Mediterranean diet – protect against CHD. Differences of scientific opinion still exist on the roles of total fat intake, and saturated and trans saturated fats. However, there is no doubt that an improved diet, regular physical activity and nonsmoking will prevent the majority of new CHD events in New Zealand, other similar countries, and probably in developing countries as well.2 Of equal importance is the evidence that these risk factors, along with an excessive total caloric intake, are responsible for a significant burden from other chronic diseases, including Type 2 diabetes.3

This compelling evidence base is critical to health policy in New Zealand, given our high burden of CHD and other cardiovascular disease. This high burden continues despite the impressive declines in death rates over the last three decades. These diseases also remain responsible for a large part of the ethnic and socioeconomic inequalities in health in New Zealand.4,5 Two questions arise. How much effort should now be directed into further research into the aetiology of CHD, and how can we more fully implement existing knowledge?

Laugesen and Elliott are clearly of the opinion that more aetiological research is required. Their paper in this issue6 presents evidence on the possible role of cow milk A1 β-casein in the aetiology of CHD and Type 1 diabetes. The authors are to be congratulated on the way they have followed up earlier work in this area because of the importance of the dairy industry and the relative high content of A1 β-casein in New Zealand milk. On the basis of the strong correlations found between per capita consumption of A1 β -casein and milk protein and national CHD rates, and of A1 with Type 1 diabetes, the authors suggest the need for further animal research and clinical trials of A1-free versus ‘ordinary’ milk.

How should we respond to this paper and its suggested research agenda? It is important to point out that the authors note the general limitations of the research design. Correlations at the population level tell us nothing about the diets of individuals who get CHD and those who do not; the epidemiological literature has many examples of this ecological fallacy. Questions can also be raised about the inclusion criteria for the countries (only 20 “healthcare-affluent” countries out of 36 likely to be in this category), the validity of the basic data upon which the conclusions are based, and the justification for the five-year time lags – an earlier paper used a ten-year period.7 The authors appreciate that the ecological study design is only a starting point for the generation of hypotheses. It would be prudent, however, to suggest other observational study designs before embarking on the difficult, complex and expensive clinical trials, even if they could be designed and implemented satisfactorily. Further animal studies alone will never be sufficient for public policy decisions.

An important question concerns the nature of further chronic disease research, given existing knowledge and the shortage of research resources.8 The research likely to have the most immediate impact on improving population health and reducing inequalities in New Zealand, will be directed towards identifying policies and programmes that implement existing knowledge. In the short term, the greatest gains are likely to come from ensuring that all New Zealanders at very high risk of cardiovascular disease receive the appropriate long-term management (lipid and blood pressure reduction, smoking cessation therapy), with the goal of substantially reducing the individual risks of disease. If these relatively cost-effective therapies were available equitably, they could significantly and quickly reduce ethnic inequalities in chronic disease outcomes. Of course, this strategy will do little to reduce the long-term burden of chronic disease; only the population approach to primary prevention has this potential.9 Despite progress over recent decades, much more could be done to fully implement this strategy in New Zealand. This will require further concentrated effort, guided by the New Zealand Health Strategy,10 and involve effective partnerships among government, nongovernmental organisations and civil society more generally.

The attraction of the A1/A2 hypothesis is the simplicity of the potential public health intervention, if the authors are proved correct. It would be reasonably straightforward to change New Zealand dairy herds to produce only A2 milk. The intervention would require no change in behaviour by New Zealanders and could be implemented with little personal difficulty for substantial health gain. For these reasons, we encourage Laugesen and Elliott to pursue this research. However, this research should not be at the expense of policy and programmatic research and must not distract us from the pressing need to act now upon what we already know.

Author information: Robert Beaglehole, Professor, Evidence and Information for Policy, World Health Organization, Geneva; Rod Jackson, Professor, Head of Division of Community Health and Director of the Effective Practice Institute, Faculty of Medical and Health Sciences, University of Auckland.

Correspondence: Professor Robert Beaglehole, Human Resources for Health Team, Department of Health Service Provision, World Health Organization, 20 Avenue Appia, CH-1211 Geneva 27, Switzerland. Fax: +41 22 791 4747; email: beagleholer@who.int