Nutrition and Health: Setting Realistic Expectations and Changing Research Targets

In this issue of the Journal of Bone and Mineral Research (JBMR), along with Joshua Lewis and Trudy Voortman, we make some recommendations regarding how nutrition research for bone health can be best steered toward maximized reliability and impact.(1) Bone health is already one of the most sophisticated subfields of nutrition science. On average, it is more advanced in technological and methodological sophistication and better sensitized to the need of generating rigorous evidence compared to most other nutrition-related disciplines. Bone health nutrition is already informed by several thousands of publications from clinical trials; it uses data from large prospective cohort studies with the ambition to map the entire lifespan; and its substantial maturity in the genetic understanding of bone phenotypes offers solid ground for Mendelian randomization studies. Many researchers working in the field are often methodologically savvy, and they recognize the relative strengths and many limitations of the various research designs that they employ. New approaches such as evaluations of the impact of diet on the gut microbiome and its indirect implications to bone health, interventions from genetically modified food crops increasing specific nutritional content, and the broadly broadcasted future of "personalized" nutrition are offering tantalizing possibilities for future work. It is unknown which, if any, will prove to be useful and which will remain at the level of (refuted) speculation. At the same time, much like in other domains of nutrition research, when treatment (supplementation) for extreme nutritional deficiencies is set apart, there has been little or no progress in identifying silver bullets or even modestly effective nutritional interventions that clearly improve hard outcomes in bone health, specifically serious fractures. A main question is whether we can expect that major successes and moonshot-level advances are possible in the field of nutrition bone research in the mid-range future. Let us assume for a moment that all the promising tools, methods, and data which have accumulated and are likely to accrue in the near future are put to best possible use. Will we be able to achieve a clear paradigm change; eg, reduce the number of new fractures occurring per year globally by 50% by optimizing diet in pragmatic circumstances for individuals and for communities? Predicting the future is always precarious and prone to failure, but the most likely answer is "no". There are many reasons for such a pessimistic reply. Even if the best diet is found, uptake and adherence need to be secured, especially for long-term outcomes to be improved. Therefore, research on behavioral interventions will continue to be needed. Behavior is the rate-limiting step for achieving meaningful change. However, there are additional challenges. First, the expanding global population is becoming older. By 2050, the number of people over 65 years old will be about 1.5 billion, almost three times the current number. Obesity, the main devastating phenotype related to poor (excessive) nutrition, is likely to become twice as common, reaching levels of 50% to 60% or more in the global population by 2050.(2) This translates to 5 to 6 billion obese people, approximately four times as many as the current numbers. Concurrently, malnutrition is likely to continue to be a serious problem. On a global level, currently malnutrition is even more important than obesity and many key micronutrients have strong relationships to bone health. Third, the future of physical activity is more uncertain, but it is likely that physical inactivity will become globally more widespread, if the current negative trend continues, in particular in high-income countries.(3) Fourth, there is large uncertainty about evolutions in health systems, health care, social welfare, and wider environmental circumstances with evolving climate change, heavy pollution, and emerging new infectious and other threats that may change human behavior, nutrition, and risk profiles for diverse diseases. The major impact of diet choices may not be so much an issue of personal benefits, but global effects, eg, on the environment.(4) One can be an optimist and expect that somehow human ingenuity and last-minute efforts will revert some of these negative trends and major threats. Moreover, perhaps new technologies and progress in medicine and public health will come up with solutions that are currently hardly imagined in terms of their scope and potential. However, unqualified optimism is not solid ground for rational, objective thinking and effective action. With these prospects, one wonders whether efforts to identify minute differences in diet (either single nutrients or foods or even more comprehensive diet patterns) that improve bone or other nutrition-related outcomes are a fruitful avenue for research and public health—or, instead, should one focus on larger-scale questions and social determinants of a changing world? Do we need another million papers of nutrition research in the same genre as the previous ones? Of course not.(5) We know that most of the previous million papers published to date have been highly problematic and even flawed—with clear stellar exceptions that account for a minority. Do we then need more of the same type of research in terms of target questions, but done with more rigorous methods, with greater transparency, with more sharing of data and methods across investigators, with more rigorous outcomes, with better measurement tools? All of that would be nice, and there is genuine reason to suspect that we will have a better yield of reliable and implementable knowledge if we go down this path. However, even then, there may be a ceiling to possible success unless we also change, or at least expand, our research agenda radically. This means moving away from asking minute questions that lead to minute, barely incremental answers, even if the answers are perfect. In fact, we may have to acknowledge that some types of nutrition research answers are so complex and difficult to address with sufficient accuracy that the yield is likely to be limited or null. This does not mean that we should abandon the effort. Instead, we should tune our expectations toward realism. Adoption of the best research practices, with full transparency and reproducibility, may be helpful even in the worst scenario emerges; ie, if almost all the avenues pursued to date trying to find minute silver bullets are proven to be dead ends. A disillusionment may lead to a crisis, but in a healthy research community, this should quickly lead to a refocusing of our agenda on bigger questions. These may have less to do with precise subtle nutrition changes and more with large-scale determinants at a global level.(6) Do we know this for sure? No, we do not. This is why we need to try with our best tools and see what we can achieve that is reliable, trustworthy, and helps the lives of millions of people and patients. We should expect to find the truth, even if the truth thwarts some of our original expectations.