Sun Xuejun : The relationship between diet and health, achieving a clear conclusion is still the current problem (reprinted)
2022-10-12 15:54 | System Classification: Overseas Observation
https://blog.sciencenet.cn/blog-41174-1359129.html
Carrying out such a discussion itself shows that we are not very clear about the relationship between diet and health. Because health is a complex problem and diet is also a complex problem, it is a very difficult task to causally connect two complex phenomena. As this article says, many relationships between health and diet cannot have clear conclusions, and of course some relationships are very clear. In the current situation, we can only choose to use diets with very clear relationships as guidance. Relationships that are not so clear about relationships should not be used blindly for the time being.
Studies linking diet with health must get a whole lot better (nature.com)
Will eating red meat reduce lifespan? Of course, some researchers think so. Global research on disease burden, injury and risk factors has prompted WHO and the USDA to recommend that people limit the consumption of raw red meat to protect themselves from diseases such as 2 diabetes and various cancers.
Other researchers are not sure. The red meat intake targets set by public health officials and the expert panel vary widely, with some suggesting that people should not exceed 14 grams per day, while others do not specify the recommended upper limit. This sends a confusing message that itself is not good for public health.
is more than just red meat: the evidence base for nutrition and broader health advice is equally controversial. Now, A new approach could help health decision makers better evaluate the quality of studies that assess potential health risks. A team from the Institute of Health Indicators and Assessment of the University of Washington (IHME) in Seattle has created a star-based metric that ratings the quality of evidence for a link between a specific behavior, such as eating red meat or smoking, and a specific health outcome. A five-star rating means that the connection is clear, and one-star means either there is no connection between these two factors or the evidence is too weak to draw a clear conclusion.
The "burden of proof" analysis that researchers call it does not itself solve troubling problems such as the risks of red meat or the benefits of vegetables. However, as a judgment on the quality of existing research, it can help indicate to research funders which areas require better evidence to draw more precise conclusions. How is the
star built? What are its parameters? Can this method itself be considered a rigorous study? The IHME team did several things to try to quantify the impact of various biases in the research being evaluated. For example, an epidemiology study may be biased in a different way to a study that examines the results of a health intervention. The researchers also eliminated the common sources of bias in the study, which assumed that health risks increased exponentially with the parameters studied, such as blood pressure or consumption of unprocessed red meat. They attempted to explain the possible bias of the sample size.
applies this framework to studies that evaluate a total of 180 questions, and the results produced are mostly expected. For example, studies that evaluated the relationship between smoking and multiple cancers received a five-star rating. Similarly, high systolic blood pressure (the power of the heart to pump blood) is associated with vascular stenosis (called ischemic heart disease).
studies that evaluate diet and its health results obtained significantly lower star ratings. For example, the IHME analysis found that there was only weak evidence that there was a correlation between consumption of unprocessed red meat and results such as colorectal cancer, type 2 diabetes and ischemic heart disease. It found no relationship in the study exploring whether eating unprocessed red meat can cause two strokes. There is stronger but not overwhelming evidence that eating vegetables can reduce the risk of stroke and ischemic heart disease.
In some cases, the lower star rating may be due to the effect size: for example, any health risk from red meat consumption can be small compared to the huge losses caused to the body by smoking.Most importantly, the results of lower ratings suggest that research in these areas needs to be done better if compelling results are to be achieved.
It is difficult to find the effects of a single dietary ingredient from a person’s various diets throughout his life. This also requires larger-scale research, requires diversified participants and strict control over their daily diet. Such research will require collaboration between research groups with different expertise and access to participants in different environments – an initiative that must be encouraged by funders. This is a priority. Low personal risk does not mean a small impact on public health: low risk behaviors, if common, may have a large impact on the population.
The literature in the field of responsible research and innovation emphasizes that measures in science must always be questioned by robustness and rigor. As indicated by initiatives such as the San Francisco Research Assessment Declaration and the Leiden Declaration, extensive consultations are required and the unintended consequences of the use of indicators must be anticipated as much as possible. This work should be done sooner rather than later.
We have evidence that inadequate clinical studies, lack of necessary controls to understand the data, does not help. If funders do not work hard to provide high-quality data, the public will remain confused, bored, distrustful, and deprived of the information needed to make informed health and lifestyle choices.
